Boat acceleration temporal structure of thestroke cycle and effectiveness in rowingV Kleshnev

BioRow Ltd 4 Egerton Road Slough Berkshire SL2 2LD UK email klevalbtinternetcom

The manuscript was received on 25 January 2009 and was accepted after revision for publication on 23 September 2009

DOI 10124317543371JSET40

Abstract The purposes of the study were to explain the double peak in the boat accelerationduring the drive phase to analyse the temporal structure of the stroke cycle and to findcorrelations between the temporal structure and the boat type rower gender stroke rate forceprofile and effectiveness of rowing Measurements of the boat acceleration boat velocityhandle force oar angle and velocity of body segments were made in competitive rowing boatsusing a telemetry system The accelerations of the whole system and of the rowerrsquos centre ofmass (CM) were derived and used to define the temporal structure of the stroke cycle Sixmicrophases were defined during the drive phase and three microphases were defined duringthe recovery phase The relative magnitudes of the accelerations of the boat and of the rowerrsquosCM switched twice during the drive phase During the lsquoinitial rowerrsquos accelerationrsquo and thelsquorowerrsquos accelerationrsquo microphases the acceleration of the rowerrsquos CM was higher and duringthe lsquoinitial boat accelerationrsquo and lsquoboat accelerationrsquo microphases the acceleration of the boatwas higher The presence of the initial boat acceleration microphase is an important indicatorof the effectiveness of a rowerrsquos technique

Keywords boat acceleration temporal structure of the stroke cycle effectiveness in rowing

1 INTRODUCTION

In a temporal or phase analysis amovement is dividedinto a sequence of phases and microphases Eachphase has a clearly defined biomechanical functionand clearly identified phase boundaries (often calledlsquokey momentsrsquo or lsquokey eventsrsquo) [1] A temporal analysisis a very versatile method of biomechanical analysisand can be applied to many sports Temporal analysiscan play an integrating role for other biomechani-cal techniques eg by linking kinetic and kinematicmeasurements obtained using various instrumentssuch as force platforms with video analysis Tem-poral analysis can also reduce the complexity of ananalysis of a sporting technique and can aid a coachrsquosor athletersquos understanding of a technique

Temporal analysis is a well-developed technique ina number of cyclic locomotion sports such as runningcycling swimming skating and skiing The mostcommon definition has two main phases in the cyclefirst the lsquopropulsiversquo phase (the lsquodriversquo or lsquostrokersquophase) where the athlete actively interacts with thesupport substance (ground water snow or ice) andexecutes an effort to propel the centre of mass (CM)

of the athletendashequipment system forwards secondthe lsquonon-propulsiversquo phase (the lsquoglidersquo or lsquorecoveryrsquophase) where resistance forces act alone to decreasethe speed of the system CM [2] These two phases maybe further divided into microphases For example inrunning the propulsive phase has been divided intolsquofoot strikersquo lsquomidsupportrsquo and lsquotake-offrsquo microphasesand the recovery phase has been divided into lsquofollow-throughrsquo lsquoforward swingrsquo and lsquofoot descentrsquo micro-phases [3]

Temporal analysis of rowing is not as well devel-oped as it is in other cyclic sports Muscular activityand body segment sequencing has been used todefine microphases (lsquolegs pushrsquo lsquotrunk driversquo andlsquoarm pullrsquo) during the drive phase [4 5] Howeverthis approach does not link the main propulsion taskwith the analysis of the timing of the phases

In the present study the acceleration of the boatwas used to link the main propulsive task with thetemporal analysis Typical patterns of boat accelera-tion have one period of negative acceleration (iereducing boat velocity) at the catch and two periodsof positive acceleration during the drive and reco-very [6] The negative acceleration at the catch and

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

SPECIAL ISSUE PAPER 63

the positive acceleration during the recovery areexplained by the transfer of momentum between therowerrsquos mass and the boat mass (which can movesignificantly relative to each other) Positive accel-eration during the drive phase is explained by thepropulsion forces produced by the oars Boat accel-eration during the drive phase is usually not constantit typically has a first peak then a short gap andfinally a longer positive peak Coaches have expres-sed their concerns about this variation in boatacceleration during the drive but the reason for thisphenomenon is not clear

The purposes of this study were first to explain thedouble peak in the boat acceleration during the drivephase in rowing second to analyse the temporalstructure of the stroke cycle in terms of the accel-erations of the boat and of the rowerrsquos CM and thirdto find correlations between the temporal structureof the stroke cycle and the boat type rower genderstroke rate force profile and effectiveness of rowing

2 METHOD

21 Test subjects

The measurements were conducted during 1999ndash2004 as a part of regular biomechanical assessmentsof athletes at the Australian Institute of Sport andfrom the Australian National Rowing Team Thestudy was approved by the ethics commission andthe athletes were informed about the nature andoutcomes of the measurements A total of 294 crewsboth male and female were measured Table 1 gives abreakdown of the rowers into the normal rowinggroups and key average parameters of those groupsThese data were used for the statistical analysis ofthe temporal structure

In addition a series of more detailed assessmentswere used in this study with measurements of thestretcher and gate forces in addition to the standard setof measurements Seven male single scullers with amean height of 189ndash 004m and body mass of864ndash 31 kg took part in addition to their regular bio-

mechanical assessment These data were used for thedefinition and explanation of the temporal structure

22 Data acquisition

A radio telemetry systemwas used for data acquisitionThe system consisted of transducers an electronic unitwith signal conditioners and analogue-to-digital con-verters (12 bit 25Hz sampling frequency) and a pairof radio modems Acquired data were transmitted to amotorboat and stored in real time on the hard disk ofa laptop computer The measurements were madeduring on-water rowing in standard racing boats [7]The total mass of each boat with riggers oars andtelemetry system but without the moveable seat wasestimated for each session All crews performed anincremental test consisting of four to eight efforts over250ndash500m at a pace of 16ndash46 strokesmin

Several relevant mechanical parameters were meas-ured The boat velocity Vboat was measured using anelectromagnetic impeller (NielsenndashKellerman Booth-wyn Pennsylvania USA) which was linked to the tele-metry system The impeller was calibrated duringeach session by timing the boat over a known dis-tance The linear calibration regression was calcu-lated using at least two different average boat speeds(accuracy ndash10 per cent)

The boat acceleration Aboat was measured alongthe horizontal axis using an accelerometer (AnalogDevices Norwood Massachusetts USA accuracyndash1 per cent) The accelerometer was calibrated usingthree-point linear regression where the horizontalorientation of the accelerometer was taken as zeroand the vertical orientations of the accelerometer inopposite directions were taken as ndash gfrac14 9796ms2 [8]The accelerometer was built into the electronics unitmounted on the boat deck and aligned along theboatrsquos longitudinal axis Positive acceleration wastaken to be towards the bow of the boat (ie when theboat increases its velocity)

The horizontal oar angle Ah and vertical oarangle Av were measured using conductive-plasticpotentiometers (linearity ndash 01 per cent) connectedto the oar shaft with a light arm and a bracket The

Table 1 Subjects of the study (the values ndash the standard deviations are given for the heightsand the body masses)

Gender Weight Boat type n Height (m) Body mass (kg)

Male Heavyweight Sweep 46 1943ndash 0031 915 ndash24Male Heavyweight Scull 45 1915ndash 0045 907 ndash41Male Lightweight Sweep 36 1830ndash 0021 730 ndash10Male Lightweight Scull 29 1833ndash 0028 734 ndash13Female Heavyweight Sweep 52 1804ndash 0028 741 ndash26Female Heavyweight Scull 44 1794ndash 0040 742 ndash35Female Lightweight Scull 42 1702ndash 0032 591 ndash10

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

64 V Kleshnev

oar angles were calibrated at four or more pointsusing a protractor

The normal component of the force Fh applied tothe handle of the oar was measured using a custom-made strain-gauge-based transducer attached to theoar shaft (accuracy ndash 05 per cent) Each instru-mented oar was dynamically calibrated before eachsession using a precision load cell to apply a load tothe handle The oar was supported 002m from thecollar (at the point corresponding to the centre ofcontact with the gate) the blade was fixed and theforce applied perpendicular to the oar shaft at themiddle of the handle (015m in from the end for asweep oar 006m for a sculling oar)

The seat position Ls was measured using a custom-made device which consisted of a spring-loadedten-turn potentiometer connected to the seat with alow-stretch line (accuracy ndash 01 per cent) The posi-tion Lt of the top of the trunk was measured in smallboats using a device similar to the seat positionsensor The device was attached to the boat deck andthe line was passed up through a pulley mounted ona mast and attached to the trunk of the athlete levelwith the sternum and clavicle (C7) joint

23 Data processing

The data collected during one test effort were con-verted into a form representing one typical strokecycle for the sample [9] In this lsquonormalizationrsquo pro-cess the average cycle time was calculated over allthe strokes in the sample and then each cycle of the

sample was extended or compressed to have thesame average cycle time The cycle start time wasdefined as the moment when the oar (right oar insculling) of the stroke rower crosses the zero angle(perpendicular to the boat) during the recoveryphase This moment was used as the start of thestroke cycle for the whole crew The data were thenprocessed (averaged) into 50-point data arrays foreach measured variable (oar angle force etc)

Discrete data values (extremes or microphases)were then derived using second-order polynomialinterpolation based on the four nearest points ofthe arrays The normalization algorithm was checkedfor validity by means of a comparison between thederived values (such as catch and release anglesmaximum force work and power) calculated usingnormalized data and an average of those values fromeach cycle in the sample using raw data The differ-ences ranged from 002 per cent to 085 per centwhich was considered satisfactory

24 Data analysis

A new method was developed to define the accel-eration of the rowerrsquos CM The total blade force Fbl ofthe crew was derived from the sum of the measuredhandle forces Fh multiplied by the ratio of the actualinboard length Lin-a to the outboard length Lout-a(Fig 1)

Neglecting oar inertia (which is very small duringthe drive) the total blade force is given by

Fgate

Fh

Fbl

X

OVboat Aboat

Ah

Lin-a

Lout-a

Fdrag

Lin

Y

Lout

Fig 1 Coordinate system and forces acting on the boat and oar

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 65

Fbl frac14 FhLin-aLout-a

eth1THORN

The effective inboard length was calculated from

Lin-a frac14 Lin Lh

2thornWg

2eth2THORN

where Lh is the length of handle grip (012m for scullsand 030m for sweep oars) Lin is the measuredinboard length and Wgfrac14 004m is the gate widthThe effective outboard length was calculated from

Lout-a frac14 ethLoar LinTHORN Lsp

2Wg

2eth3THORN

where Loar is the overall oar length and Lsp is thelength of the blade spoon It was assumed that thispoint is the centre of water pressure on the bladewhich is a limitation of the method

The drag force Fdrag acting on the boat shell wasderived as [4]

Fdrag frac14 K dragV2boat eth4THORN

where the drag factor Kdrag was derived as the ratio ofthe integral of the blade propulsive force to the inte-gral of the square of boat speed over the stroke cycleinterval T according to

K drag frac14ETH tfrac14Ttfrac140 Fbl cos u dtETH tfrac14T

tfrac140 V 2boat dt

eth5THORN

It was assumed that the blade force acts perpendi-cularly to the oar shaft which is a limitation of themethod The propulsive force Fsys acting on therowerndashboat system was defined as

F sys frac14 Fbl cosAh thorn Fdrag eth6THORNThe acceleration Asys of the CM of the system wascalculated as

Asys frac14 F sys

msysfrac14 F sys

mboat thornmroweth7THORN

where msys mboat and mrow are the masses of thesystem boat and rower respectively The accelera-tion Arow of the rowerrsquos CM was calculated as

Arow frac14 F row

mroweth8THORN

The boat acceleration Aboat and propulsive force Fboatare related as

Aboat frac14 Fboat

mboateth9THORN

Using the boat propulsive force Fboat derived inthe detailed measurements and the measured boatacceleration Aboat it was found that the acting boatmass mboat was higher than the measured boat mass

The reason for this could be the hydrodynamic addedmass or other factors (aerodynamics dynamics ofoars force on seat etc) which were difficult to esti-mate Therefore a correction factor was used andmboat was assumed to be 30 per cent heavier than wasmeasured

The force Frow applied to the rower was derived as

F row frac14 F sys Fboat

frac14 F sys Aboatmboat eth10THORN

The deviations Vd of the instantaneous velocities ofthe system boat and rowerrsquos CM from the averagevelocity of the system (and each component) duringthe stroke cycle were calculated as

V d frac14ethA dt eth11THORN

where A is the corresponding acceleration (Asys Aboator Arow) In other words a frame of reference wasused which moved with a constant velocity equal tothe average velocity of the boat during the strokecycle The X axis was parallel to the boatrsquos long-itudinal axis and directed towards the bow of theboat

The velocities of the body segments were derivedfor use in qualitative analyses of microphases Thevelocity Vlegs of the legs was derived as the rate ofvariation in the measured seat position Ls The velo-city Vtrunkb of the trunk relative to the boat wasderived in the same way using the trunk position LtThe velocity Vtrunk of the trunk relative to the seat wascalculated by subtracting Vlegs from Vtrunkb Thevelocity Varms of the arms was derived as the differ-ence between the handle velocity Vh and Vtrunkbwhere Vh was derived from the measured horizontaloar angle and actual inboard according to

V h frac14 dAh

dtLin-a

p

180eth12THORN

25 Statistical analysis

An independent-samples t test was used to deter-mine whether any differences between sampleswere significant Significance was tested at the 005level Pearson correlation was used to define therelationships of the measured variables in the wholesample

3 RESULTS AND DISCUSSION

31 Boat acceleration patterns

The magnitude of the boat acceleration Aboat washighly dependent on the stroke rate Typical patterns

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

66 V Kleshnev

(Fig 2(a)) usually had one negative peak at the catchand two positive peaks during the drive The accel-eration Arow of the rowersrsquo CM (Fig 2(b)) had twopositive peaks one just after the catch and one in themiddle of the drive The system acceleration Asys

reflected the force application pattern and followedthe shape of the force curve

The negative peak of the boat acceleration Aboat

significantly increased in magnitude with increasingstroke rate (Fig 3(a)) (correlation with stroke raterfrac14 ndash085plt 001) This can be explained by thesubstantial increase in inertial forces at higher strokerates In contrast the first positive peak of Aboat

increased less at higher stroke rates (Fig 3(b)) Thesystem acceleration Asys usually increased its positivepeak at higher stroke rates and decreased its negativevalues during recovery This reflected the increaseddrag resistance at higher boat speeds

32 Definition of the microphases of thestroke cycle

The accelerations of the CM of the boat rower andthe whole system (Aboat Arow and Asys) were used to

define the microphases of the stroke cycle Thesevariables were chosen because they are related to theaccumulation of kinetic energy in the two majorcomponents of the rowerndashboat system and hence tothe effectiveness of rowing technique The wordlsquoeffectivenessrsquo here means the capability of produ-cing an effect ie the manner of application of therowerrsquos efforts is such that it provides the maximalaverage speed of the boatndashrower system The effec-tiveness differs from the lsquoefficiencyrsquo which is a ratioof the output effect (boat speed) to the input energy(rowerrsquos power) and is not discussed in this paperWhile the efficiency is easy to measure the effec-tiveness is difficult to quantify because the boatspeed depends on many factors which are beyondour control namely weather conditions rowerrsquosefforts and capabilities Therefore in this article theterm effectiveness is used in a qualitative way and itsevaluation is based on the overall performance of thecrew

The handle velocity Vh was used to define theoverall drive phase boundaries Figure 4 shows typi-cal biomechanical variables of a single sculler

Fig 3 Dependences of (a) the negative peak of the boat acceleration at catch and (b) the first positivepeak of the boat acceleration during the drive (b) on the stroke rate

Fig 2 Patterns of acceleration of (a) the boat and (b) the rowersrsquo CM at different stroke rates for anOlympic Champion menrsquos pair

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 67

Six microphases of the drive phase (D1 to D6) andthree microphases of the recovery phase (R1 to R3)were defined Table 2 presents the microphase iden-tifications descriptions and key events which definetheir start and end points

33 Dependence of the microphase structureon the boat type rower gender strokerate and force profile

Microphases D1 D2 and D3 were slightly longer insweep boats than in sculling boats and microphases

Fig 4 Typical biomechanical parameters and microphases of the stroke cycle (M1x menrsquos single scullrate 32 strokesmin) Key events are indicated as open circles

Table 2 Characteristics of the microphases of the stroke cycle

Identification Microphase Start event Description

D1 Blade immersion Catch beginning of the drive Vh changes sign topositive

Asys and Aboat are negative but Arow is positiveFast increases in Vh and Vlegs

D2 Initial rowersrsquoacceleration

Asys becomes positive The centre of the bladecrosses the water level downwards

Fh and Aboat increase but Aboat is still negativeand lower than Arow

D3 Initial boatacceleration

Aboat becomes higher than Arow First positive peak of Aboat which becomeshigher than Arow Maximum of Vlegs

D4 Rowersrsquoacceleration

Aboat decreases and becomes lower than therowerrsquos acceleration

Fh Arow and Asys increase slowly Vlegs decreases

D5 Boat acceleration Aboat again becomes higher than Arow Fh Ffoot Arow and Asys decrease but Ffootdecreases faster than Fh which produces thehighest Aboat

D6 Blade removal Asys becomes negative The centre of the bladecrosses the water level upwards

Arow is negative and Aboat is close to zero Vh isstill positive Varms is maximal

R1 Arms and trunkreturn

Finish end of the drive Vh changes sign tonegative

A quick positive peak of Aboat and negative Arowcaused by transfer of the moment of inertiafrom the rower to the boat

R2 Legs return Seat starts to move towards the stern Increase inAboat and decrease in Arow

Aboat is positive but Arow and Asys are negativeVlegs increases towards the stern

R3 Catch preparation Aboat becomes negative and Arow becomespositive

Ffoot increases which causes a decrease in Vlegsdeceleration of the boat and acceleration of therower

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

68 V Kleshnev

D4 D5 and D6 were slightly shorter (plt 005)(Fig 5) The longer duration of D1 in sweep rowingmight be due to the greater width of the sweep oarblade which requires more time to insert into thewater The shorter duration of D2 and D3 in scullingmight be due to the longer catch angles which cre-ates heavier effective gearing [9 10] A heavier gear-ing produces a lower ratio of handle to gate forceswhich creates higher acceleration of the boat com-pared with the rowerrsquos CM

There were no significant differences in the dura-tions of the D1 D2 D3 and D6 microphases amongboats of different sizes (Fig 6) D4 was significantlylonger in large boats (eights and quads) and D5 waslonger in small boats (singles and pairs plt 001) butthe reasons for these differences are not known

Male crews had a longer D3 microphase thanfemale crews did (plt 001) which is probably due totheir more rapid generation of forces Differencesbetween other microphases for the two genders werenot statistically significant

The absolute duration of all microphases exceptfor D3 tended to decrease at higher stroke rates Thecorrelation coefficients ranged from ndash010 for D4 tondash070 for D1 This decrease in duration was stronglycorrelated with a decrease in total drive time athigher stroke rates (rfrac14 ndash087)

The relative share of the D1 microphase within thedrive phase tended to decrease at higher stroke rates

(rfrac14 ndash049) whereas the relative share of the D6microphase tended to increase (Fig 7) Sculling crewstended to exhibit an absolute maximum in the rela-tive share of the D3 microphase at a stroke rate of30 strokesmin and sweep crews at 36 strokesminThe reason for this phenomenon is not known

To analyse the relationship between microphaseproportions and the force curve profile there was thetime from the catch until the moment when the forceVh achieved 10 per cent 20 per cent 100 per centof its maximal value The following statistically sig-nificant relationships were found (Fig 8)

1 The D1 period had the strongest positive correla-tion with the time for achieving 30 per cent of themaximal force (rfrac14 091 plt 0001)

2 D3 had the strongest negative correlation with thetime for achieving 70 per cent (rfrac14 ndash045 plt 005)and 80 per cent (rfrac14 ndash046 plt 005) of the max-imal force

3 D6 had a negative correlation with the timetaken to achieve 30 per cent of the maximal force(rfrac14 ndash046 plt 005)

34 Case study the importance of theD3 microphase

To discuss the influence of different rowing techni-que factors on the temporal structure of the drive

Fig 6 Time structures of the drive phase in small (1middot single scull 2pair) medium (2middotdouble scull4 four) and large (4middotquadruple scull 8thorn eight) boat types

Fig 5 Time structures of the drive phase in sweep and sculling crews

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 69

phase two crews of international level were selectedThe first crew consisted of Olympic Champions andwas considered by experts to be highly technicallyeffective The second crew was in the finals of theWorld Championship had very good physiologicalwork capacity but could not achieve excellent per-formance on the water Both samples were taken at astroke rate close to 32 strokesmin The first crewproduced a much quicker rise in the handle force(Fig 9(a)) achieving 70 per cent of their maximalforce at 231 per cent of the drive time The secondcrew had a relatively flatter force curve (70 per cent ofthe maximal force at 291 per cent of the drive time)but higher maximal and average forces

The faster rise in the force of the first crew coin-cides with a faster increase in the velocity Vlegs of thelegs (Fig 9(d)) (the maximal velocity of the legsoccurred at 224 per cent of the drive time) andwas accompanied by a steady increase in the trunkvelocity Vtrunk (Fig 9(f)) The second crew had aslower velocity of the legs in the first half of the drive

(a)

(b)

(c)

10

15

20

25

30

15 25 35

Rate (minndash1)

0

5

10

15

20

25

16 20 24 28 32 36 40

Sweep

Scull

Rate (minndash1)

)( e

mit evird ni 3D fo erah

S

0

5

10

15

20

15 25 35Rate (minndash1)

45

45

)( e

mit evird ni 6D f o erah

S )

( emit evird ni 1

D fo erahS

Fig 7 Dependences of the microphase time shares in thedrive time on the stroke rate (a) (b) linear trends(c) second-order polynomial trend

Fig 8 Dependences of the proportion of microphasewithin the drive phase on the stroke rate with lineartrend lines

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

70 V Kleshnev

(the maximal velocity of the legs at 322 per cent ofthe drive time) and a double-peaked trunk speedprofile Consequently the handle velocity Vh of thefirst crew increased more rapidly during the first halfof the drive (Fig 9(b))

The boat acceleration Aboat of the first crew had adeeper and later negative peak at the catch (84ms2

compared with 73ms2 in the second crew) but amuch quicker increase afterwards (Fig 9(c)) The firstpeak of Aboat was 27ms2 in the first crew but only03ms2 in the second crew which had Aboat dropbelow zero after the first peak

The first crew had a higher acceleration Arow ofthe rowerrsquos CM at the catch but the second crew

Fig 9 Biomechanical parameters and temporal structures of the drive phase for two examples of rowingtechnique

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 71

accelerated their body mass more intensively duringthe middle of the drive (Fig 9(e)) As a result thesecond crew did not have microphase D3 (initial boatacceleration) at all but it had a much longer D4(main rowersrsquo acceleration)

4 DISCUSSION AND IMPLICATIONS TEMPORALANALYSIS AND THE EFFECTIVENESS OF THEROWING TECHNIQUE

The main measure of rowing performance is theaverage speed of the rowerndashboat system Rowers can-not increase the system velocity during the recoveryphase because its deceleration is affected by suchenvironmental factors as water and air resistanceThe drive phase is the only time when rowers canincrease the system velocity (more precisely onlythat part of the drive when the propulsive force ishigher than the drag force) The higher accelera-tion of the system during the drive phase meansa higher average speed and hence better perfor-mance The system accumulates kinetic energy dur-ing the drive phase and loses it during the recoveryphase

The gain dE in kinetic energy of any body can bedefined using the equation

dE frac14 mv222

mv212

frac14 m

2ethv22 v21THORN eth13THORN

wherem is the mass of the body and v1 and v2 are thestart and end velocities respectively of the bodyrsquos CM

The mass of the rower is significantly greater thanthat of the boat (the ratio ranges from 400 per cent inlightweight women to 600 per cent in heavyweightmen) To maximize the average speed the rowersmust maximize the gain in kinetic energy which canbe achieved most effectively by means of accelerationof the heaviest part of the system (ie the rowerrsquosCM) Therefore the effectiveness of the rowing tech-

nique is related to the magnitude and timing of theacceleration of the rowerrsquos CM

The acceleration of any body depends on the forceacting on it The force Frow applied to the rowerrsquos CM(Fig 10) is equal to the sum of the foot-stretcherreaction force Frfoot which is positive ie pushesthe athletersquos CM forwards and the handle reactionforce Frhandle which is negative ie pulls the rowerbackwards according to

F row frac14 F rfoot thorn F rhandle eth14THORNBoth these reaction forces have the same magnitudeand the opposite direction to the action forces Ffootand Fh

The force Fboat applied to the boat hull is equal tothe sum of the gate force Fgate the stretcher forceFfoot and the drag force Fdrag (ignoring the small seatfrictional force) according to

Fboat frac14 Fgate thorn F foot thorn Fdrag eth15THORNTherefore emphasis on the boat accelerationrequires rowers to produce a higher gate force Fgate(and a correspondingly higher handle force Fh) andto push on the stretcher less Vice versa emphasison accelerating the rowerrsquos CM requires a higherstretcher force Ffoot and lower handle force FhGraphs of the gate and stretcher forces in Fig 4 con-firm that this switching of the emphasis happenstwice during the drive

The stretcher force Ffoot is the primary focus duringthe drive because it increases the acceleration Arow ofthe rowerrsquos CM which provides the most significantaccumulation of kinetic energy and hence increasesthe systemrsquos average speed

However the effectiveness of the rowerrsquos push onthe stretcher is affected by the boat velocity at thattime If the stretcher (boat) has a higher velocity thenthe rowers can accelerate their CM more efficientlythan if the boat is moving more slowly It can bespeculated that using the elastic energy stored in the

Frower

Fhandle

F foot

Fgate

Fbl

M

mFb

F drag

Fig 10 The main forces in the rowerndashboatndashoar system

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

72 V Kleshnev

oar handle and in the rowerrsquos tendons and musclescan play a significant role at the beginning of thedrive However this topic is beyond the scope of thispaper and can be the subject of further research Theuse of elastic energy is similar to jumping on atrampoline the point of support moves in the direc-tion of acceleration of the athletersquos CM and adds thevelocity of the support point to that of the athletersquosCM which makes the acceleration more effectiveThe presence of D3 could be called a lsquotrampoliningeffectrsquo and can be regarded as an important indicatorof the effectiveness of the rowing technique [11]In addition earlier force application could help toutilize the hydrolift effect better [12]

The practical implications of the microphases ofthe drive phase are as follows

During microphases D1 and D2 the rowers empha-size push with legs because they have to changethe direction of their movement from the sternto the bow at the catch This accelerates their bodymass and decelerates the boat At the same timethe blade obtains a grip on the water

During D3 the rowers emphasize pulling thehandle to accelerate the boat This creates a faster-moving support on the stretcher This microphase isimportant for performing an effective drive phase Itwas found that in 43 per cent of the crews thismicrophase was absent but a longer duration of D3does not mean better technique because D3 was thelongest in some low-level crews (according to anexpertrsquos opinion)

During D4 the rowers emphasize the stretcherpush again to accelerate themselves and hence toaccumulate kinetic energy The effectiveness of thismicrophase depends on the amount of boat velocitygained during the previous microphase D3 and on afast powerful leg drive

The microphases D5 and D6 occur after the legs aresubstantially extended and use trunk and arm workinstead The force levels and the total system accel-eration decrease but the stretcher force decreasesmore quickly and becomes negative (pull on thestretcher) This puts emphasis on the boat accelera-tion again The rowerrsquos CM acceleration becomesnegative which means transferring kinetic energyfrom the rowerrsquos mass to the boat

5 CONCLUSIONS

The phenomenon of a double peak in the boatacceleration during the drive phase can be explainedby double switching of the emphasis from pushingthe stretcher (more rowersrsquo CM acceleration) topulling the handle (more boat hull acceleration)Acceleration of the boat and of the rowersrsquo CM andthe velocity of the oar handle were used to define the

temporal structure of the stroke cycle Six micro-phases were defined during the drive phase and threemicrophases were defined during the recovery phaseThe presence of the microphase D3 (initial boatacceleration) was an important indicator of the eff-ectiveness of a rowerrsquos technique The initial boatacceleration creates a faster-moving platform for thestretcher to allow better acceleration of the rowerrsquosCM

ACKNOWLEDGEMENT

This research project was kindly supported by theAustralian Institute of Sport and Rowing Australia

Author 2010

REFERENCES

1 Bartlett R Sport biomechanics preventing injury andimproving performance 1999 pp 11ndash21 (RoutledgeNew York)

2 Toussaint H M and Beek P J Biomechanics ofcompetitive front crawl swimming Sports Med 199213 8ndash24

3 James S L and Brubaker C E Biomechanical andneuromuscular aspects of running Exercise Sport SciRev 1973 1 189ndash216

4 Dal-Monte A and Komor A Rowing and scullingmechanics In Biomechanics of sport (Ed C L Vaughan)1989 pp 53ndash119 (CRC Press Boca Raton Florida)

5 Rodriguez R J Rogriguez R P Cook S D andSandborn P M Electromyographic analysis of rowingstroke biomechanics J Sports Med Phys Fitness 199030 103ndash108

6 Shimoda M Kawakami Y and Fukunaga T Anapplication of acceleration analysis to evaluation of arowing technique In Proceedings of the 13th Interna-tional Symposium on Biomechanics in sports (EdT Bauer) Thunder Bay Ontario Canada 18ndash22 July1995 pp 95ndash100 (Lakehead University Thunder BayOntario)

7 Federation Internationale des Societes drsquoAviron (Inter-national Rowing Federation) FISA rules of racingand related bye-laws Part IV Boats and construction2005ndash2006 pp 60ndash66 available from httpwwwworldrowingcommediasdocsmedia_352459pdf

8 Geoscience Australia Applying geoscience to Australiarsquosmost important challenges 2009 available from httpwwwgeogovau

9 Kleshnev V Technology for technique improvementIn Rowing faster (Ed V Nolte) 2004 pp 209ndash228(Human Kinetics Champaign Illinois)

10 Kleshnev V Facts Did you know that Rowing Bio-mech Newslett 2007 7(72) available from httpwwwbiorowcomRBN_en_2007_files2007RowBiomNews03pdf

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 73

the positive acceleration during the recovery areexplained by the transfer of momentum between therowerrsquos mass and the boat mass (which can movesignificantly relative to each other) Positive accel-eration during the drive phase is explained by thepropulsion forces produced by the oars Boat accel-eration during the drive phase is usually not constantit typically has a first peak then a short gap andfinally a longer positive peak Coaches have expres-sed their concerns about this variation in boatacceleration during the drive but the reason for thisphenomenon is not clear

The purposes of this study were first to explain thedouble peak in the boat acceleration during the drivephase in rowing second to analyse the temporalstructure of the stroke cycle in terms of the accel-erations of the boat and of the rowerrsquos CM and thirdto find correlations between the temporal structureof the stroke cycle and the boat type rower genderstroke rate force profile and effectiveness of rowing

2 METHOD

21 Test subjects

The measurements were conducted during 1999ndash2004 as a part of regular biomechanical assessmentsof athletes at the Australian Institute of Sport andfrom the Australian National Rowing Team Thestudy was approved by the ethics commission andthe athletes were informed about the nature andoutcomes of the measurements A total of 294 crewsboth male and female were measured Table 1 gives abreakdown of the rowers into the normal rowinggroups and key average parameters of those groupsThese data were used for the statistical analysis ofthe temporal structure

In addition a series of more detailed assessmentswere used in this study with measurements of thestretcher and gate forces in addition to the standard setof measurements Seven male single scullers with amean height of 189ndash 004m and body mass of864ndash 31 kg took part in addition to their regular bio-

mechanical assessment These data were used for thedefinition and explanation of the temporal structure

22 Data acquisition

A radio telemetry systemwas used for data acquisitionThe system consisted of transducers an electronic unitwith signal conditioners and analogue-to-digital con-verters (12 bit 25Hz sampling frequency) and a pairof radio modems Acquired data were transmitted to amotorboat and stored in real time on the hard disk ofa laptop computer The measurements were madeduring on-water rowing in standard racing boats [7]The total mass of each boat with riggers oars andtelemetry system but without the moveable seat wasestimated for each session All crews performed anincremental test consisting of four to eight efforts over250ndash500m at a pace of 16ndash46 strokesmin

Several relevant mechanical parameters were meas-ured The boat velocity Vboat was measured using anelectromagnetic impeller (NielsenndashKellerman Booth-wyn Pennsylvania USA) which was linked to the tele-metry system The impeller was calibrated duringeach session by timing the boat over a known dis-tance The linear calibration regression was calcu-lated using at least two different average boat speeds(accuracy ndash10 per cent)

The boat acceleration Aboat was measured alongthe horizontal axis using an accelerometer (AnalogDevices Norwood Massachusetts USA accuracyndash1 per cent) The accelerometer was calibrated usingthree-point linear regression where the horizontalorientation of the accelerometer was taken as zeroand the vertical orientations of the accelerometer inopposite directions were taken as ndash gfrac14 9796ms2 [8]The accelerometer was built into the electronics unitmounted on the boat deck and aligned along theboatrsquos longitudinal axis Positive acceleration wastaken to be towards the bow of the boat (ie when theboat increases its velocity)

The horizontal oar angle Ah and vertical oarangle Av were measured using conductive-plasticpotentiometers (linearity ndash 01 per cent) connectedto the oar shaft with a light arm and a bracket The

Table 1 Subjects of the study (the values ndash the standard deviations are given for the heightsand the body masses)

Gender Weight Boat type n Height (m) Body mass (kg)

Male Heavyweight Sweep 46 1943ndash 0031 915 ndash24Male Heavyweight Scull 45 1915ndash 0045 907 ndash41Male Lightweight Sweep 36 1830ndash 0021 730 ndash10Male Lightweight Scull 29 1833ndash 0028 734 ndash13Female Heavyweight Sweep 52 1804ndash 0028 741 ndash26Female Heavyweight Scull 44 1794ndash 0040 742 ndash35Female Lightweight Scull 42 1702ndash 0032 591 ndash10

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

64 V Kleshnev

oar angles were calibrated at four or more pointsusing a protractor

The normal component of the force Fh applied tothe handle of the oar was measured using a custom-made strain-gauge-based transducer attached to theoar shaft (accuracy ndash 05 per cent) Each instru-mented oar was dynamically calibrated before eachsession using a precision load cell to apply a load tothe handle The oar was supported 002m from thecollar (at the point corresponding to the centre ofcontact with the gate) the blade was fixed and theforce applied perpendicular to the oar shaft at themiddle of the handle (015m in from the end for asweep oar 006m for a sculling oar)

The seat position Ls was measured using a custom-made device which consisted of a spring-loadedten-turn potentiometer connected to the seat with alow-stretch line (accuracy ndash 01 per cent) The posi-tion Lt of the top of the trunk was measured in smallboats using a device similar to the seat positionsensor The device was attached to the boat deck andthe line was passed up through a pulley mounted ona mast and attached to the trunk of the athlete levelwith the sternum and clavicle (C7) joint

23 Data processing

The data collected during one test effort were con-verted into a form representing one typical strokecycle for the sample [9] In this lsquonormalizationrsquo pro-cess the average cycle time was calculated over allthe strokes in the sample and then each cycle of the

sample was extended or compressed to have thesame average cycle time The cycle start time wasdefined as the moment when the oar (right oar insculling) of the stroke rower crosses the zero angle(perpendicular to the boat) during the recoveryphase This moment was used as the start of thestroke cycle for the whole crew The data were thenprocessed (averaged) into 50-point data arrays foreach measured variable (oar angle force etc)

Discrete data values (extremes or microphases)were then derived using second-order polynomialinterpolation based on the four nearest points ofthe arrays The normalization algorithm was checkedfor validity by means of a comparison between thederived values (such as catch and release anglesmaximum force work and power) calculated usingnormalized data and an average of those values fromeach cycle in the sample using raw data The differ-ences ranged from 002 per cent to 085 per centwhich was considered satisfactory

24 Data analysis

A new method was developed to define the accel-eration of the rowerrsquos CM The total blade force Fbl ofthe crew was derived from the sum of the measuredhandle forces Fh multiplied by the ratio of the actualinboard length Lin-a to the outboard length Lout-a(Fig 1)

Neglecting oar inertia (which is very small duringthe drive) the total blade force is given by

Fgate

Fh

Fbl

X

OVboat Aboat

Ah

Lin-a

Lout-a

Fdrag

Lin

Y

Lout

Fig 1 Coordinate system and forces acting on the boat and oar

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 65

Fbl frac14 FhLin-aLout-a

eth1THORN

The effective inboard length was calculated from

Lin-a frac14 Lin Lh

2thornWg

2eth2THORN

where Lh is the length of handle grip (012m for scullsand 030m for sweep oars) Lin is the measuredinboard length and Wgfrac14 004m is the gate widthThe effective outboard length was calculated from

Lout-a frac14 ethLoar LinTHORN Lsp

2Wg

2eth3THORN

where Loar is the overall oar length and Lsp is thelength of the blade spoon It was assumed that thispoint is the centre of water pressure on the bladewhich is a limitation of the method

The drag force Fdrag acting on the boat shell wasderived as [4]

Fdrag frac14 K dragV2boat eth4THORN

where the drag factor Kdrag was derived as the ratio ofthe integral of the blade propulsive force to the inte-gral of the square of boat speed over the stroke cycleinterval T according to

K drag frac14ETH tfrac14Ttfrac140 Fbl cos u dtETH tfrac14T

tfrac140 V 2boat dt

eth5THORN

It was assumed that the blade force acts perpendi-cularly to the oar shaft which is a limitation of themethod The propulsive force Fsys acting on therowerndashboat system was defined as

F sys frac14 Fbl cosAh thorn Fdrag eth6THORNThe acceleration Asys of the CM of the system wascalculated as

Asys frac14 F sys

msysfrac14 F sys

mboat thornmroweth7THORN

where msys mboat and mrow are the masses of thesystem boat and rower respectively The accelera-tion Arow of the rowerrsquos CM was calculated as

Arow frac14 F row

mroweth8THORN

The boat acceleration Aboat and propulsive force Fboatare related as

Aboat frac14 Fboat

mboateth9THORN

Using the boat propulsive force Fboat derived inthe detailed measurements and the measured boatacceleration Aboat it was found that the acting boatmass mboat was higher than the measured boat mass

The reason for this could be the hydrodynamic addedmass or other factors (aerodynamics dynamics ofoars force on seat etc) which were difficult to esti-mate Therefore a correction factor was used andmboat was assumed to be 30 per cent heavier than wasmeasured

The force Frow applied to the rower was derived as

F row frac14 F sys Fboat

frac14 F sys Aboatmboat eth10THORN

The deviations Vd of the instantaneous velocities ofthe system boat and rowerrsquos CM from the averagevelocity of the system (and each component) duringthe stroke cycle were calculated as

V d frac14ethA dt eth11THORN

where A is the corresponding acceleration (Asys Aboator Arow) In other words a frame of reference wasused which moved with a constant velocity equal tothe average velocity of the boat during the strokecycle The X axis was parallel to the boatrsquos long-itudinal axis and directed towards the bow of theboat

The velocities of the body segments were derivedfor use in qualitative analyses of microphases Thevelocity Vlegs of the legs was derived as the rate ofvariation in the measured seat position Ls The velo-city Vtrunkb of the trunk relative to the boat wasderived in the same way using the trunk position LtThe velocity Vtrunk of the trunk relative to the seat wascalculated by subtracting Vlegs from Vtrunkb Thevelocity Varms of the arms was derived as the differ-ence between the handle velocity Vh and Vtrunkbwhere Vh was derived from the measured horizontaloar angle and actual inboard according to

V h frac14 dAh

dtLin-a

p

180eth12THORN

25 Statistical analysis

An independent-samples t test was used to deter-mine whether any differences between sampleswere significant Significance was tested at the 005level Pearson correlation was used to define therelationships of the measured variables in the wholesample

3 RESULTS AND DISCUSSION

31 Boat acceleration patterns

The magnitude of the boat acceleration Aboat washighly dependent on the stroke rate Typical patterns

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

66 V Kleshnev

(Fig 2(a)) usually had one negative peak at the catchand two positive peaks during the drive The accel-eration Arow of the rowersrsquo CM (Fig 2(b)) had twopositive peaks one just after the catch and one in themiddle of the drive The system acceleration Asys

reflected the force application pattern and followedthe shape of the force curve

The negative peak of the boat acceleration Aboat

significantly increased in magnitude with increasingstroke rate (Fig 3(a)) (correlation with stroke raterfrac14 ndash085plt 001) This can be explained by thesubstantial increase in inertial forces at higher strokerates In contrast the first positive peak of Aboat

increased less at higher stroke rates (Fig 3(b)) Thesystem acceleration Asys usually increased its positivepeak at higher stroke rates and decreased its negativevalues during recovery This reflected the increaseddrag resistance at higher boat speeds

32 Definition of the microphases of thestroke cycle

The accelerations of the CM of the boat rower andthe whole system (Aboat Arow and Asys) were used to

define the microphases of the stroke cycle Thesevariables were chosen because they are related to theaccumulation of kinetic energy in the two majorcomponents of the rowerndashboat system and hence tothe effectiveness of rowing technique The wordlsquoeffectivenessrsquo here means the capability of produ-cing an effect ie the manner of application of therowerrsquos efforts is such that it provides the maximalaverage speed of the boatndashrower system The effec-tiveness differs from the lsquoefficiencyrsquo which is a ratioof the output effect (boat speed) to the input energy(rowerrsquos power) and is not discussed in this paperWhile the efficiency is easy to measure the effec-tiveness is difficult to quantify because the boatspeed depends on many factors which are beyondour control namely weather conditions rowerrsquosefforts and capabilities Therefore in this article theterm effectiveness is used in a qualitative way and itsevaluation is based on the overall performance of thecrew

The handle velocity Vh was used to define theoverall drive phase boundaries Figure 4 shows typi-cal biomechanical variables of a single sculler

Fig 3 Dependences of (a) the negative peak of the boat acceleration at catch and (b) the first positivepeak of the boat acceleration during the drive (b) on the stroke rate

Fig 2 Patterns of acceleration of (a) the boat and (b) the rowersrsquo CM at different stroke rates for anOlympic Champion menrsquos pair

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 67

Six microphases of the drive phase (D1 to D6) andthree microphases of the recovery phase (R1 to R3)were defined Table 2 presents the microphase iden-tifications descriptions and key events which definetheir start and end points

33 Dependence of the microphase structureon the boat type rower gender strokerate and force profile

Microphases D1 D2 and D3 were slightly longer insweep boats than in sculling boats and microphases

Fig 4 Typical biomechanical parameters and microphases of the stroke cycle (M1x menrsquos single scullrate 32 strokesmin) Key events are indicated as open circles

Table 2 Characteristics of the microphases of the stroke cycle

Identification Microphase Start event Description

D1 Blade immersion Catch beginning of the drive Vh changes sign topositive

Asys and Aboat are negative but Arow is positiveFast increases in Vh and Vlegs

D2 Initial rowersrsquoacceleration

Asys becomes positive The centre of the bladecrosses the water level downwards

Fh and Aboat increase but Aboat is still negativeand lower than Arow

D3 Initial boatacceleration

Aboat becomes higher than Arow First positive peak of Aboat which becomeshigher than Arow Maximum of Vlegs

D4 Rowersrsquoacceleration

Aboat decreases and becomes lower than therowerrsquos acceleration

Fh Arow and Asys increase slowly Vlegs decreases

D5 Boat acceleration Aboat again becomes higher than Arow Fh Ffoot Arow and Asys decrease but Ffootdecreases faster than Fh which produces thehighest Aboat

D6 Blade removal Asys becomes negative The centre of the bladecrosses the water level upwards

Arow is negative and Aboat is close to zero Vh isstill positive Varms is maximal

R1 Arms and trunkreturn

Finish end of the drive Vh changes sign tonegative

A quick positive peak of Aboat and negative Arowcaused by transfer of the moment of inertiafrom the rower to the boat

R2 Legs return Seat starts to move towards the stern Increase inAboat and decrease in Arow

Aboat is positive but Arow and Asys are negativeVlegs increases towards the stern

R3 Catch preparation Aboat becomes negative and Arow becomespositive

Ffoot increases which causes a decrease in Vlegsdeceleration of the boat and acceleration of therower

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

68 V Kleshnev

D4 D5 and D6 were slightly shorter (plt 005)(Fig 5) The longer duration of D1 in sweep rowingmight be due to the greater width of the sweep oarblade which requires more time to insert into thewater The shorter duration of D2 and D3 in scullingmight be due to the longer catch angles which cre-ates heavier effective gearing [9 10] A heavier gear-ing produces a lower ratio of handle to gate forceswhich creates higher acceleration of the boat com-pared with the rowerrsquos CM

There were no significant differences in the dura-tions of the D1 D2 D3 and D6 microphases amongboats of different sizes (Fig 6) D4 was significantlylonger in large boats (eights and quads) and D5 waslonger in small boats (singles and pairs plt 001) butthe reasons for these differences are not known

Male crews had a longer D3 microphase thanfemale crews did (plt 001) which is probably due totheir more rapid generation of forces Differencesbetween other microphases for the two genders werenot statistically significant

The absolute duration of all microphases exceptfor D3 tended to decrease at higher stroke rates Thecorrelation coefficients ranged from ndash010 for D4 tondash070 for D1 This decrease in duration was stronglycorrelated with a decrease in total drive time athigher stroke rates (rfrac14 ndash087)

The relative share of the D1 microphase within thedrive phase tended to decrease at higher stroke rates

(rfrac14 ndash049) whereas the relative share of the D6microphase tended to increase (Fig 7) Sculling crewstended to exhibit an absolute maximum in the rela-tive share of the D3 microphase at a stroke rate of30 strokesmin and sweep crews at 36 strokesminThe reason for this phenomenon is not known

To analyse the relationship between microphaseproportions and the force curve profile there was thetime from the catch until the moment when the forceVh achieved 10 per cent 20 per cent 100 per centof its maximal value The following statistically sig-nificant relationships were found (Fig 8)

1 The D1 period had the strongest positive correla-tion with the time for achieving 30 per cent of themaximal force (rfrac14 091 plt 0001)

2 D3 had the strongest negative correlation with thetime for achieving 70 per cent (rfrac14 ndash045 plt 005)and 80 per cent (rfrac14 ndash046 plt 005) of the max-imal force

3 D6 had a negative correlation with the timetaken to achieve 30 per cent of the maximal force(rfrac14 ndash046 plt 005)

34 Case study the importance of theD3 microphase

To discuss the influence of different rowing techni-que factors on the temporal structure of the drive

Fig 6 Time structures of the drive phase in small (1middot single scull 2pair) medium (2middotdouble scull4 four) and large (4middotquadruple scull 8thorn eight) boat types

Fig 5 Time structures of the drive phase in sweep and sculling crews

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 69

phase two crews of international level were selectedThe first crew consisted of Olympic Champions andwas considered by experts to be highly technicallyeffective The second crew was in the finals of theWorld Championship had very good physiologicalwork capacity but could not achieve excellent per-formance on the water Both samples were taken at astroke rate close to 32 strokesmin The first crewproduced a much quicker rise in the handle force(Fig 9(a)) achieving 70 per cent of their maximalforce at 231 per cent of the drive time The secondcrew had a relatively flatter force curve (70 per cent ofthe maximal force at 291 per cent of the drive time)but higher maximal and average forces

The faster rise in the force of the first crew coin-cides with a faster increase in the velocity Vlegs of thelegs (Fig 9(d)) (the maximal velocity of the legsoccurred at 224 per cent of the drive time) andwas accompanied by a steady increase in the trunkvelocity Vtrunk (Fig 9(f)) The second crew had aslower velocity of the legs in the first half of the drive

(a)

(b)

(c)

10

15

20

25

30

15 25 35

Rate (minndash1)

0

5

10

15

20

25

16 20 24 28 32 36 40

Sweep

Scull

Rate (minndash1)

)( e

mit evird ni 3D fo erah

S

0

5

10

15

20

15 25 35Rate (minndash1)

45

45

)( e

mit evird ni 6D f o erah

S )

( emit evird ni 1

D fo erahS

Fig 7 Dependences of the microphase time shares in thedrive time on the stroke rate (a) (b) linear trends(c) second-order polynomial trend

Fig 8 Dependences of the proportion of microphasewithin the drive phase on the stroke rate with lineartrend lines

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

70 V Kleshnev

(the maximal velocity of the legs at 322 per cent ofthe drive time) and a double-peaked trunk speedprofile Consequently the handle velocity Vh of thefirst crew increased more rapidly during the first halfof the drive (Fig 9(b))

The boat acceleration Aboat of the first crew had adeeper and later negative peak at the catch (84ms2

compared with 73ms2 in the second crew) but amuch quicker increase afterwards (Fig 9(c)) The firstpeak of Aboat was 27ms2 in the first crew but only03ms2 in the second crew which had Aboat dropbelow zero after the first peak

The first crew had a higher acceleration Arow ofthe rowerrsquos CM at the catch but the second crew

Fig 9 Biomechanical parameters and temporal structures of the drive phase for two examples of rowingtechnique

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 71

accelerated their body mass more intensively duringthe middle of the drive (Fig 9(e)) As a result thesecond crew did not have microphase D3 (initial boatacceleration) at all but it had a much longer D4(main rowersrsquo acceleration)

4 DISCUSSION AND IMPLICATIONS TEMPORALANALYSIS AND THE EFFECTIVENESS OF THEROWING TECHNIQUE

The main measure of rowing performance is theaverage speed of the rowerndashboat system Rowers can-not increase the system velocity during the recoveryphase because its deceleration is affected by suchenvironmental factors as water and air resistanceThe drive phase is the only time when rowers canincrease the system velocity (more precisely onlythat part of the drive when the propulsive force ishigher than the drag force) The higher accelera-tion of the system during the drive phase meansa higher average speed and hence better perfor-mance The system accumulates kinetic energy dur-ing the drive phase and loses it during the recoveryphase

The gain dE in kinetic energy of any body can bedefined using the equation

dE frac14 mv222

mv212

frac14 m

2ethv22 v21THORN eth13THORN

wherem is the mass of the body and v1 and v2 are thestart and end velocities respectively of the bodyrsquos CM

The mass of the rower is significantly greater thanthat of the boat (the ratio ranges from 400 per cent inlightweight women to 600 per cent in heavyweightmen) To maximize the average speed the rowersmust maximize the gain in kinetic energy which canbe achieved most effectively by means of accelerationof the heaviest part of the system (ie the rowerrsquosCM) Therefore the effectiveness of the rowing tech-

nique is related to the magnitude and timing of theacceleration of the rowerrsquos CM

The acceleration of any body depends on the forceacting on it The force Frow applied to the rowerrsquos CM(Fig 10) is equal to the sum of the foot-stretcherreaction force Frfoot which is positive ie pushesthe athletersquos CM forwards and the handle reactionforce Frhandle which is negative ie pulls the rowerbackwards according to

F row frac14 F rfoot thorn F rhandle eth14THORNBoth these reaction forces have the same magnitudeand the opposite direction to the action forces Ffootand Fh

The force Fboat applied to the boat hull is equal tothe sum of the gate force Fgate the stretcher forceFfoot and the drag force Fdrag (ignoring the small seatfrictional force) according to

Fboat frac14 Fgate thorn F foot thorn Fdrag eth15THORNTherefore emphasis on the boat accelerationrequires rowers to produce a higher gate force Fgate(and a correspondingly higher handle force Fh) andto push on the stretcher less Vice versa emphasison accelerating the rowerrsquos CM requires a higherstretcher force Ffoot and lower handle force FhGraphs of the gate and stretcher forces in Fig 4 con-firm that this switching of the emphasis happenstwice during the drive

The stretcher force Ffoot is the primary focus duringthe drive because it increases the acceleration Arow ofthe rowerrsquos CM which provides the most significantaccumulation of kinetic energy and hence increasesthe systemrsquos average speed

However the effectiveness of the rowerrsquos push onthe stretcher is affected by the boat velocity at thattime If the stretcher (boat) has a higher velocity thenthe rowers can accelerate their CM more efficientlythan if the boat is moving more slowly It can bespeculated that using the elastic energy stored in the

Frower

Fhandle

F foot

Fgate

Fbl

M

mFb

F drag

Fig 10 The main forces in the rowerndashboatndashoar system

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

72 V Kleshnev

oar handle and in the rowerrsquos tendons and musclescan play a significant role at the beginning of thedrive However this topic is beyond the scope of thispaper and can be the subject of further research Theuse of elastic energy is similar to jumping on atrampoline the point of support moves in the direc-tion of acceleration of the athletersquos CM and adds thevelocity of the support point to that of the athletersquosCM which makes the acceleration more effectiveThe presence of D3 could be called a lsquotrampoliningeffectrsquo and can be regarded as an important indicatorof the effectiveness of the rowing technique [11]In addition earlier force application could help toutilize the hydrolift effect better [12]

The practical implications of the microphases ofthe drive phase are as follows

During microphases D1 and D2 the rowers empha-size push with legs because they have to changethe direction of their movement from the sternto the bow at the catch This accelerates their bodymass and decelerates the boat At the same timethe blade obtains a grip on the water

During D3 the rowers emphasize pulling thehandle to accelerate the boat This creates a faster-moving support on the stretcher This microphase isimportant for performing an effective drive phase Itwas found that in 43 per cent of the crews thismicrophase was absent but a longer duration of D3does not mean better technique because D3 was thelongest in some low-level crews (according to anexpertrsquos opinion)

During D4 the rowers emphasize the stretcherpush again to accelerate themselves and hence toaccumulate kinetic energy The effectiveness of thismicrophase depends on the amount of boat velocitygained during the previous microphase D3 and on afast powerful leg drive

The microphases D5 and D6 occur after the legs aresubstantially extended and use trunk and arm workinstead The force levels and the total system accel-eration decrease but the stretcher force decreasesmore quickly and becomes negative (pull on thestretcher) This puts emphasis on the boat accelera-tion again The rowerrsquos CM acceleration becomesnegative which means transferring kinetic energyfrom the rowerrsquos mass to the boat

5 CONCLUSIONS

The phenomenon of a double peak in the boatacceleration during the drive phase can be explainedby double switching of the emphasis from pushingthe stretcher (more rowersrsquo CM acceleration) topulling the handle (more boat hull acceleration)Acceleration of the boat and of the rowersrsquo CM andthe velocity of the oar handle were used to define the

temporal structure of the stroke cycle Six micro-phases were defined during the drive phase and threemicrophases were defined during the recovery phaseThe presence of the microphase D3 (initial boatacceleration) was an important indicator of the eff-ectiveness of a rowerrsquos technique The initial boatacceleration creates a faster-moving platform for thestretcher to allow better acceleration of the rowerrsquosCM

ACKNOWLEDGEMENT

This research project was kindly supported by theAustralian Institute of Sport and Rowing Australia

Author 2010

REFERENCES

1 Bartlett R Sport biomechanics preventing injury andimproving performance 1999 pp 11ndash21 (RoutledgeNew York)

2 Toussaint H M and Beek P J Biomechanics ofcompetitive front crawl swimming Sports Med 199213 8ndash24

3 James S L and Brubaker C E Biomechanical andneuromuscular aspects of running Exercise Sport SciRev 1973 1 189ndash216

4 Dal-Monte A and Komor A Rowing and scullingmechanics In Biomechanics of sport (Ed C L Vaughan)1989 pp 53ndash119 (CRC Press Boca Raton Florida)

5 Rodriguez R J Rogriguez R P Cook S D andSandborn P M Electromyographic analysis of rowingstroke biomechanics J Sports Med Phys Fitness 199030 103ndash108

6 Shimoda M Kawakami Y and Fukunaga T Anapplication of acceleration analysis to evaluation of arowing technique In Proceedings of the 13th Interna-tional Symposium on Biomechanics in sports (EdT Bauer) Thunder Bay Ontario Canada 18ndash22 July1995 pp 95ndash100 (Lakehead University Thunder BayOntario)

7 Federation Internationale des Societes drsquoAviron (Inter-national Rowing Federation) FISA rules of racingand related bye-laws Part IV Boats and construction2005ndash2006 pp 60ndash66 available from httpwwwworldrowingcommediasdocsmedia_352459pdf

8 Geoscience Australia Applying geoscience to Australiarsquosmost important challenges 2009 available from httpwwwgeogovau

9 Kleshnev V Technology for technique improvementIn Rowing faster (Ed V Nolte) 2004 pp 209ndash228(Human Kinetics Champaign Illinois)

10 Kleshnev V Facts Did you know that Rowing Bio-mech Newslett 2007 7(72) available from httpwwwbiorowcomRBN_en_2007_files2007RowBiomNews03pdf

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 73

oar angles were calibrated at four or more pointsusing a protractor

The normal component of the force Fh applied tothe handle of the oar was measured using a custom-made strain-gauge-based transducer attached to theoar shaft (accuracy ndash 05 per cent) Each instru-mented oar was dynamically calibrated before eachsession using a precision load cell to apply a load tothe handle The oar was supported 002m from thecollar (at the point corresponding to the centre ofcontact with the gate) the blade was fixed and theforce applied perpendicular to the oar shaft at themiddle of the handle (015m in from the end for asweep oar 006m for a sculling oar)

The seat position Ls was measured using a custom-made device which consisted of a spring-loadedten-turn potentiometer connected to the seat with alow-stretch line (accuracy ndash 01 per cent) The posi-tion Lt of the top of the trunk was measured in smallboats using a device similar to the seat positionsensor The device was attached to the boat deck andthe line was passed up through a pulley mounted ona mast and attached to the trunk of the athlete levelwith the sternum and clavicle (C7) joint

23 Data processing

The data collected during one test effort were con-verted into a form representing one typical strokecycle for the sample [9] In this lsquonormalizationrsquo pro-cess the average cycle time was calculated over allthe strokes in the sample and then each cycle of the

sample was extended or compressed to have thesame average cycle time The cycle start time wasdefined as the moment when the oar (right oar insculling) of the stroke rower crosses the zero angle(perpendicular to the boat) during the recoveryphase This moment was used as the start of thestroke cycle for the whole crew The data were thenprocessed (averaged) into 50-point data arrays foreach measured variable (oar angle force etc)

Discrete data values (extremes or microphases)were then derived using second-order polynomialinterpolation based on the four nearest points ofthe arrays The normalization algorithm was checkedfor validity by means of a comparison between thederived values (such as catch and release anglesmaximum force work and power) calculated usingnormalized data and an average of those values fromeach cycle in the sample using raw data The differ-ences ranged from 002 per cent to 085 per centwhich was considered satisfactory

24 Data analysis

A new method was developed to define the accel-eration of the rowerrsquos CM The total blade force Fbl ofthe crew was derived from the sum of the measuredhandle forces Fh multiplied by the ratio of the actualinboard length Lin-a to the outboard length Lout-a(Fig 1)

Neglecting oar inertia (which is very small duringthe drive) the total blade force is given by

Fgate

Fh

Fbl

X

OVboat Aboat

Ah

Lin-a

Lout-a

Fdrag

Lin

Y

Lout

Fig 1 Coordinate system and forces acting on the boat and oar

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 65

Fbl frac14 FhLin-aLout-a

eth1THORN

The effective inboard length was calculated from

Lin-a frac14 Lin Lh

2thornWg

2eth2THORN

where Lh is the length of handle grip (012m for scullsand 030m for sweep oars) Lin is the measuredinboard length and Wgfrac14 004m is the gate widthThe effective outboard length was calculated from

Lout-a frac14 ethLoar LinTHORN Lsp

2Wg

2eth3THORN

where Loar is the overall oar length and Lsp is thelength of the blade spoon It was assumed that thispoint is the centre of water pressure on the bladewhich is a limitation of the method

The drag force Fdrag acting on the boat shell wasderived as [4]

Fdrag frac14 K dragV2boat eth4THORN

where the drag factor Kdrag was derived as the ratio ofthe integral of the blade propulsive force to the inte-gral of the square of boat speed over the stroke cycleinterval T according to

K drag frac14ETH tfrac14Ttfrac140 Fbl cos u dtETH tfrac14T

tfrac140 V 2boat dt

eth5THORN

It was assumed that the blade force acts perpendi-cularly to the oar shaft which is a limitation of themethod The propulsive force Fsys acting on therowerndashboat system was defined as

F sys frac14 Fbl cosAh thorn Fdrag eth6THORNThe acceleration Asys of the CM of the system wascalculated as

Asys frac14 F sys

msysfrac14 F sys

mboat thornmroweth7THORN

where msys mboat and mrow are the masses of thesystem boat and rower respectively The accelera-tion Arow of the rowerrsquos CM was calculated as

Arow frac14 F row

mroweth8THORN

The boat acceleration Aboat and propulsive force Fboatare related as

Aboat frac14 Fboat

mboateth9THORN

Using the boat propulsive force Fboat derived inthe detailed measurements and the measured boatacceleration Aboat it was found that the acting boatmass mboat was higher than the measured boat mass

The reason for this could be the hydrodynamic addedmass or other factors (aerodynamics dynamics ofoars force on seat etc) which were difficult to esti-mate Therefore a correction factor was used andmboat was assumed to be 30 per cent heavier than wasmeasured

The force Frow applied to the rower was derived as

F row frac14 F sys Fboat

frac14 F sys Aboatmboat eth10THORN

The deviations Vd of the instantaneous velocities ofthe system boat and rowerrsquos CM from the averagevelocity of the system (and each component) duringthe stroke cycle were calculated as

V d frac14ethA dt eth11THORN

where A is the corresponding acceleration (Asys Aboator Arow) In other words a frame of reference wasused which moved with a constant velocity equal tothe average velocity of the boat during the strokecycle The X axis was parallel to the boatrsquos long-itudinal axis and directed towards the bow of theboat

The velocities of the body segments were derivedfor use in qualitative analyses of microphases Thevelocity Vlegs of the legs was derived as the rate ofvariation in the measured seat position Ls The velo-city Vtrunkb of the trunk relative to the boat wasderived in the same way using the trunk position LtThe velocity Vtrunk of the trunk relative to the seat wascalculated by subtracting Vlegs from Vtrunkb Thevelocity Varms of the arms was derived as the differ-ence between the handle velocity Vh and Vtrunkbwhere Vh was derived from the measured horizontaloar angle and actual inboard according to

V h frac14 dAh

dtLin-a

p

180eth12THORN

25 Statistical analysis

An independent-samples t test was used to deter-mine whether any differences between sampleswere significant Significance was tested at the 005level Pearson correlation was used to define therelationships of the measured variables in the wholesample

3 RESULTS AND DISCUSSION

31 Boat acceleration patterns

The magnitude of the boat acceleration Aboat washighly dependent on the stroke rate Typical patterns

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

66 V Kleshnev

(Fig 2(a)) usually had one negative peak at the catchand two positive peaks during the drive The accel-eration Arow of the rowersrsquo CM (Fig 2(b)) had twopositive peaks one just after the catch and one in themiddle of the drive The system acceleration Asys

reflected the force application pattern and followedthe shape of the force curve

The negative peak of the boat acceleration Aboat

significantly increased in magnitude with increasingstroke rate (Fig 3(a)) (correlation with stroke raterfrac14 ndash085plt 001) This can be explained by thesubstantial increase in inertial forces at higher strokerates In contrast the first positive peak of Aboat

increased less at higher stroke rates (Fig 3(b)) Thesystem acceleration Asys usually increased its positivepeak at higher stroke rates and decreased its negativevalues during recovery This reflected the increaseddrag resistance at higher boat speeds

32 Definition of the microphases of thestroke cycle

The accelerations of the CM of the boat rower andthe whole system (Aboat Arow and Asys) were used to

define the microphases of the stroke cycle Thesevariables were chosen because they are related to theaccumulation of kinetic energy in the two majorcomponents of the rowerndashboat system and hence tothe effectiveness of rowing technique The wordlsquoeffectivenessrsquo here means the capability of produ-cing an effect ie the manner of application of therowerrsquos efforts is such that it provides the maximalaverage speed of the boatndashrower system The effec-tiveness differs from the lsquoefficiencyrsquo which is a ratioof the output effect (boat speed) to the input energy(rowerrsquos power) and is not discussed in this paperWhile the efficiency is easy to measure the effec-tiveness is difficult to quantify because the boatspeed depends on many factors which are beyondour control namely weather conditions rowerrsquosefforts and capabilities Therefore in this article theterm effectiveness is used in a qualitative way and itsevaluation is based on the overall performance of thecrew

The handle velocity Vh was used to define theoverall drive phase boundaries Figure 4 shows typi-cal biomechanical variables of a single sculler

Fig 3 Dependences of (a) the negative peak of the boat acceleration at catch and (b) the first positivepeak of the boat acceleration during the drive (b) on the stroke rate

Fig 2 Patterns of acceleration of (a) the boat and (b) the rowersrsquo CM at different stroke rates for anOlympic Champion menrsquos pair

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 67

Six microphases of the drive phase (D1 to D6) andthree microphases of the recovery phase (R1 to R3)were defined Table 2 presents the microphase iden-tifications descriptions and key events which definetheir start and end points

33 Dependence of the microphase structureon the boat type rower gender strokerate and force profile

Microphases D1 D2 and D3 were slightly longer insweep boats than in sculling boats and microphases

Fig 4 Typical biomechanical parameters and microphases of the stroke cycle (M1x menrsquos single scullrate 32 strokesmin) Key events are indicated as open circles

Table 2 Characteristics of the microphases of the stroke cycle

Identification Microphase Start event Description

D1 Blade immersion Catch beginning of the drive Vh changes sign topositive

Asys and Aboat are negative but Arow is positiveFast increases in Vh and Vlegs

D2 Initial rowersrsquoacceleration

Asys becomes positive The centre of the bladecrosses the water level downwards

Fh and Aboat increase but Aboat is still negativeand lower than Arow

D3 Initial boatacceleration

Aboat becomes higher than Arow First positive peak of Aboat which becomeshigher than Arow Maximum of Vlegs

D4 Rowersrsquoacceleration

Aboat decreases and becomes lower than therowerrsquos acceleration

Fh Arow and Asys increase slowly Vlegs decreases

D5 Boat acceleration Aboat again becomes higher than Arow Fh Ffoot Arow and Asys decrease but Ffootdecreases faster than Fh which produces thehighest Aboat

D6 Blade removal Asys becomes negative The centre of the bladecrosses the water level upwards

Arow is negative and Aboat is close to zero Vh isstill positive Varms is maximal

R1 Arms and trunkreturn

Finish end of the drive Vh changes sign tonegative

A quick positive peak of Aboat and negative Arowcaused by transfer of the moment of inertiafrom the rower to the boat

R2 Legs return Seat starts to move towards the stern Increase inAboat and decrease in Arow

Aboat is positive but Arow and Asys are negativeVlegs increases towards the stern

R3 Catch preparation Aboat becomes negative and Arow becomespositive

Ffoot increases which causes a decrease in Vlegsdeceleration of the boat and acceleration of therower

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

68 V Kleshnev

D4 D5 and D6 were slightly shorter (plt 005)(Fig 5) The longer duration of D1 in sweep rowingmight be due to the greater width of the sweep oarblade which requires more time to insert into thewater The shorter duration of D2 and D3 in scullingmight be due to the longer catch angles which cre-ates heavier effective gearing [9 10] A heavier gear-ing produces a lower ratio of handle to gate forceswhich creates higher acceleration of the boat com-pared with the rowerrsquos CM

There were no significant differences in the dura-tions of the D1 D2 D3 and D6 microphases amongboats of different sizes (Fig 6) D4 was significantlylonger in large boats (eights and quads) and D5 waslonger in small boats (singles and pairs plt 001) butthe reasons for these differences are not known

Male crews had a longer D3 microphase thanfemale crews did (plt 001) which is probably due totheir more rapid generation of forces Differencesbetween other microphases for the two genders werenot statistically significant

The absolute duration of all microphases exceptfor D3 tended to decrease at higher stroke rates Thecorrelation coefficients ranged from ndash010 for D4 tondash070 for D1 This decrease in duration was stronglycorrelated with a decrease in total drive time athigher stroke rates (rfrac14 ndash087)

The relative share of the D1 microphase within thedrive phase tended to decrease at higher stroke rates

(rfrac14 ndash049) whereas the relative share of the D6microphase tended to increase (Fig 7) Sculling crewstended to exhibit an absolute maximum in the rela-tive share of the D3 microphase at a stroke rate of30 strokesmin and sweep crews at 36 strokesminThe reason for this phenomenon is not known

To analyse the relationship between microphaseproportions and the force curve profile there was thetime from the catch until the moment when the forceVh achieved 10 per cent 20 per cent 100 per centof its maximal value The following statistically sig-nificant relationships were found (Fig 8)

1 The D1 period had the strongest positive correla-tion with the time for achieving 30 per cent of themaximal force (rfrac14 091 plt 0001)

2 D3 had the strongest negative correlation with thetime for achieving 70 per cent (rfrac14 ndash045 plt 005)and 80 per cent (rfrac14 ndash046 plt 005) of the max-imal force

3 D6 had a negative correlation with the timetaken to achieve 30 per cent of the maximal force(rfrac14 ndash046 plt 005)

34 Case study the importance of theD3 microphase

To discuss the influence of different rowing techni-que factors on the temporal structure of the drive

Fig 6 Time structures of the drive phase in small (1middot single scull 2pair) medium (2middotdouble scull4 four) and large (4middotquadruple scull 8thorn eight) boat types

Fig 5 Time structures of the drive phase in sweep and sculling crews

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 69

phase two crews of international level were selectedThe first crew consisted of Olympic Champions andwas considered by experts to be highly technicallyeffective The second crew was in the finals of theWorld Championship had very good physiologicalwork capacity but could not achieve excellent per-formance on the water Both samples were taken at astroke rate close to 32 strokesmin The first crewproduced a much quicker rise in the handle force(Fig 9(a)) achieving 70 per cent of their maximalforce at 231 per cent of the drive time The secondcrew had a relatively flatter force curve (70 per cent ofthe maximal force at 291 per cent of the drive time)but higher maximal and average forces

The faster rise in the force of the first crew coin-cides with a faster increase in the velocity Vlegs of thelegs (Fig 9(d)) (the maximal velocity of the legsoccurred at 224 per cent of the drive time) andwas accompanied by a steady increase in the trunkvelocity Vtrunk (Fig 9(f)) The second crew had aslower velocity of the legs in the first half of the drive

(a)

(b)

(c)

10

15

20

25

30

15 25 35

Rate (minndash1)

0

5

10

15

20

25

16 20 24 28 32 36 40

Sweep

Scull

Rate (minndash1)

)( e

mit evird ni 3D fo erah

S

0

5

10

15

20

15 25 35Rate (minndash1)

45

45

)( e

mit evird ni 6D f o erah

S )

( emit evird ni 1

D fo erahS

Fig 7 Dependences of the microphase time shares in thedrive time on the stroke rate (a) (b) linear trends(c) second-order polynomial trend

Fig 8 Dependences of the proportion of microphasewithin the drive phase on the stroke rate with lineartrend lines

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

70 V Kleshnev

(the maximal velocity of the legs at 322 per cent ofthe drive time) and a double-peaked trunk speedprofile Consequently the handle velocity Vh of thefirst crew increased more rapidly during the first halfof the drive (Fig 9(b))

The boat acceleration Aboat of the first crew had adeeper and later negative peak at the catch (84ms2

compared with 73ms2 in the second crew) but amuch quicker increase afterwards (Fig 9(c)) The firstpeak of Aboat was 27ms2 in the first crew but only03ms2 in the second crew which had Aboat dropbelow zero after the first peak

The first crew had a higher acceleration Arow ofthe rowerrsquos CM at the catch but the second crew

Fig 9 Biomechanical parameters and temporal structures of the drive phase for two examples of rowingtechnique

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 71

accelerated their body mass more intensively duringthe middle of the drive (Fig 9(e)) As a result thesecond crew did not have microphase D3 (initial boatacceleration) at all but it had a much longer D4(main rowersrsquo acceleration)

4 DISCUSSION AND IMPLICATIONS TEMPORALANALYSIS AND THE EFFECTIVENESS OF THEROWING TECHNIQUE

The main measure of rowing performance is theaverage speed of the rowerndashboat system Rowers can-not increase the system velocity during the recoveryphase because its deceleration is affected by suchenvironmental factors as water and air resistanceThe drive phase is the only time when rowers canincrease the system velocity (more precisely onlythat part of the drive when the propulsive force ishigher than the drag force) The higher accelera-tion of the system during the drive phase meansa higher average speed and hence better perfor-mance The system accumulates kinetic energy dur-ing the drive phase and loses it during the recoveryphase

The gain dE in kinetic energy of any body can bedefined using the equation

dE frac14 mv222

mv212

frac14 m

2ethv22 v21THORN eth13THORN

wherem is the mass of the body and v1 and v2 are thestart and end velocities respectively of the bodyrsquos CM

The mass of the rower is significantly greater thanthat of the boat (the ratio ranges from 400 per cent inlightweight women to 600 per cent in heavyweightmen) To maximize the average speed the rowersmust maximize the gain in kinetic energy which canbe achieved most effectively by means of accelerationof the heaviest part of the system (ie the rowerrsquosCM) Therefore the effectiveness of the rowing tech-

nique is related to the magnitude and timing of theacceleration of the rowerrsquos CM

The acceleration of any body depends on the forceacting on it The force Frow applied to the rowerrsquos CM(Fig 10) is equal to the sum of the foot-stretcherreaction force Frfoot which is positive ie pushesthe athletersquos CM forwards and the handle reactionforce Frhandle which is negative ie pulls the rowerbackwards according to

F row frac14 F rfoot thorn F rhandle eth14THORNBoth these reaction forces have the same magnitudeand the opposite direction to the action forces Ffootand Fh

The force Fboat applied to the boat hull is equal tothe sum of the gate force Fgate the stretcher forceFfoot and the drag force Fdrag (ignoring the small seatfrictional force) according to

Fboat frac14 Fgate thorn F foot thorn Fdrag eth15THORNTherefore emphasis on the boat accelerationrequires rowers to produce a higher gate force Fgate(and a correspondingly higher handle force Fh) andto push on the stretcher less Vice versa emphasison accelerating the rowerrsquos CM requires a higherstretcher force Ffoot and lower handle force FhGraphs of the gate and stretcher forces in Fig 4 con-firm that this switching of the emphasis happenstwice during the drive

The stretcher force Ffoot is the primary focus duringthe drive because it increases the acceleration Arow ofthe rowerrsquos CM which provides the most significantaccumulation of kinetic energy and hence increasesthe systemrsquos average speed

However the effectiveness of the rowerrsquos push onthe stretcher is affected by the boat velocity at thattime If the stretcher (boat) has a higher velocity thenthe rowers can accelerate their CM more efficientlythan if the boat is moving more slowly It can bespeculated that using the elastic energy stored in the

Frower

Fhandle

F foot

Fgate

Fbl

M

mFb

F drag

Fig 10 The main forces in the rowerndashboatndashoar system

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

72 V Kleshnev

oar handle and in the rowerrsquos tendons and musclescan play a significant role at the beginning of thedrive However this topic is beyond the scope of thispaper and can be the subject of further research Theuse of elastic energy is similar to jumping on atrampoline the point of support moves in the direc-tion of acceleration of the athletersquos CM and adds thevelocity of the support point to that of the athletersquosCM which makes the acceleration more effectiveThe presence of D3 could be called a lsquotrampoliningeffectrsquo and can be regarded as an important indicatorof the effectiveness of the rowing technique [11]In addition earlier force application could help toutilize the hydrolift effect better [12]

The practical implications of the microphases ofthe drive phase are as follows

During microphases D1 and D2 the rowers empha-size push with legs because they have to changethe direction of their movement from the sternto the bow at the catch This accelerates their bodymass and decelerates the boat At the same timethe blade obtains a grip on the water

During D3 the rowers emphasize pulling thehandle to accelerate the boat This creates a faster-moving support on the stretcher This microphase isimportant for performing an effective drive phase Itwas found that in 43 per cent of the crews thismicrophase was absent but a longer duration of D3does not mean better technique because D3 was thelongest in some low-level crews (according to anexpertrsquos opinion)

During D4 the rowers emphasize the stretcherpush again to accelerate themselves and hence toaccumulate kinetic energy The effectiveness of thismicrophase depends on the amount of boat velocitygained during the previous microphase D3 and on afast powerful leg drive

The microphases D5 and D6 occur after the legs aresubstantially extended and use trunk and arm workinstead The force levels and the total system accel-eration decrease but the stretcher force decreasesmore quickly and becomes negative (pull on thestretcher) This puts emphasis on the boat accelera-tion again The rowerrsquos CM acceleration becomesnegative which means transferring kinetic energyfrom the rowerrsquos mass to the boat

5 CONCLUSIONS

The phenomenon of a double peak in the boatacceleration during the drive phase can be explainedby double switching of the emphasis from pushingthe stretcher (more rowersrsquo CM acceleration) topulling the handle (more boat hull acceleration)Acceleration of the boat and of the rowersrsquo CM andthe velocity of the oar handle were used to define the

temporal structure of the stroke cycle Six micro-phases were defined during the drive phase and threemicrophases were defined during the recovery phaseThe presence of the microphase D3 (initial boatacceleration) was an important indicator of the eff-ectiveness of a rowerrsquos technique The initial boatacceleration creates a faster-moving platform for thestretcher to allow better acceleration of the rowerrsquosCM

ACKNOWLEDGEMENT

This research project was kindly supported by theAustralian Institute of Sport and Rowing Australia

Author 2010

REFERENCES

1 Bartlett R Sport biomechanics preventing injury andimproving performance 1999 pp 11ndash21 (RoutledgeNew York)

2 Toussaint H M and Beek P J Biomechanics ofcompetitive front crawl swimming Sports Med 199213 8ndash24

3 James S L and Brubaker C E Biomechanical andneuromuscular aspects of running Exercise Sport SciRev 1973 1 189ndash216

4 Dal-Monte A and Komor A Rowing and scullingmechanics In Biomechanics of sport (Ed C L Vaughan)1989 pp 53ndash119 (CRC Press Boca Raton Florida)

5 Rodriguez R J Rogriguez R P Cook S D andSandborn P M Electromyographic analysis of rowingstroke biomechanics J Sports Med Phys Fitness 199030 103ndash108

6 Shimoda M Kawakami Y and Fukunaga T Anapplication of acceleration analysis to evaluation of arowing technique In Proceedings of the 13th Interna-tional Symposium on Biomechanics in sports (EdT Bauer) Thunder Bay Ontario Canada 18ndash22 July1995 pp 95ndash100 (Lakehead University Thunder BayOntario)

7 Federation Internationale des Societes drsquoAviron (Inter-national Rowing Federation) FISA rules of racingand related bye-laws Part IV Boats and construction2005ndash2006 pp 60ndash66 available from httpwwwworldrowingcommediasdocsmedia_352459pdf

8 Geoscience Australia Applying geoscience to Australiarsquosmost important challenges 2009 available from httpwwwgeogovau

9 Kleshnev V Technology for technique improvementIn Rowing faster (Ed V Nolte) 2004 pp 209ndash228(Human Kinetics Champaign Illinois)

10 Kleshnev V Facts Did you know that Rowing Bio-mech Newslett 2007 7(72) available from httpwwwbiorowcomRBN_en_2007_files2007RowBiomNews03pdf

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 73

Fbl frac14 FhLin-aLout-a

eth1THORN

The effective inboard length was calculated from

Lin-a frac14 Lin Lh

2thornWg

2eth2THORN

where Lh is the length of handle grip (012m for scullsand 030m for sweep oars) Lin is the measuredinboard length and Wgfrac14 004m is the gate widthThe effective outboard length was calculated from

Lout-a frac14 ethLoar LinTHORN Lsp

2Wg

2eth3THORN

where Loar is the overall oar length and Lsp is thelength of the blade spoon It was assumed that thispoint is the centre of water pressure on the bladewhich is a limitation of the method

The drag force Fdrag acting on the boat shell wasderived as [4]

Fdrag frac14 K dragV2boat eth4THORN

where the drag factor Kdrag was derived as the ratio ofthe integral of the blade propulsive force to the inte-gral of the square of boat speed over the stroke cycleinterval T according to

K drag frac14ETH tfrac14Ttfrac140 Fbl cos u dtETH tfrac14T

tfrac140 V 2boat dt

eth5THORN

It was assumed that the blade force acts perpendi-cularly to the oar shaft which is a limitation of themethod The propulsive force Fsys acting on therowerndashboat system was defined as

F sys frac14 Fbl cosAh thorn Fdrag eth6THORNThe acceleration Asys of the CM of the system wascalculated as

Asys frac14 F sys

msysfrac14 F sys

mboat thornmroweth7THORN

where msys mboat and mrow are the masses of thesystem boat and rower respectively The accelera-tion Arow of the rowerrsquos CM was calculated as

Arow frac14 F row

mroweth8THORN

The boat acceleration Aboat and propulsive force Fboatare related as

Aboat frac14 Fboat

mboateth9THORN

Using the boat propulsive force Fboat derived inthe detailed measurements and the measured boatacceleration Aboat it was found that the acting boatmass mboat was higher than the measured boat mass

The reason for this could be the hydrodynamic addedmass or other factors (aerodynamics dynamics ofoars force on seat etc) which were difficult to esti-mate Therefore a correction factor was used andmboat was assumed to be 30 per cent heavier than wasmeasured

The force Frow applied to the rower was derived as

F row frac14 F sys Fboat

frac14 F sys Aboatmboat eth10THORN

The deviations Vd of the instantaneous velocities ofthe system boat and rowerrsquos CM from the averagevelocity of the system (and each component) duringthe stroke cycle were calculated as

V d frac14ethA dt eth11THORN

where A is the corresponding acceleration (Asys Aboator Arow) In other words a frame of reference wasused which moved with a constant velocity equal tothe average velocity of the boat during the strokecycle The X axis was parallel to the boatrsquos long-itudinal axis and directed towards the bow of theboat

The velocities of the body segments were derivedfor use in qualitative analyses of microphases Thevelocity Vlegs of the legs was derived as the rate ofvariation in the measured seat position Ls The velo-city Vtrunkb of the trunk relative to the boat wasderived in the same way using the trunk position LtThe velocity Vtrunk of the trunk relative to the seat wascalculated by subtracting Vlegs from Vtrunkb Thevelocity Varms of the arms was derived as the differ-ence between the handle velocity Vh and Vtrunkbwhere Vh was derived from the measured horizontaloar angle and actual inboard according to

V h frac14 dAh

dtLin-a

p

180eth12THORN

25 Statistical analysis

An independent-samples t test was used to deter-mine whether any differences between sampleswere significant Significance was tested at the 005level Pearson correlation was used to define therelationships of the measured variables in the wholesample

3 RESULTS AND DISCUSSION

31 Boat acceleration patterns

The magnitude of the boat acceleration Aboat washighly dependent on the stroke rate Typical patterns

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

66 V Kleshnev

(Fig 2(a)) usually had one negative peak at the catchand two positive peaks during the drive The accel-eration Arow of the rowersrsquo CM (Fig 2(b)) had twopositive peaks one just after the catch and one in themiddle of the drive The system acceleration Asys

reflected the force application pattern and followedthe shape of the force curve

The negative peak of the boat acceleration Aboat

significantly increased in magnitude with increasingstroke rate (Fig 3(a)) (correlation with stroke raterfrac14 ndash085plt 001) This can be explained by thesubstantial increase in inertial forces at higher strokerates In contrast the first positive peak of Aboat

increased less at higher stroke rates (Fig 3(b)) Thesystem acceleration Asys usually increased its positivepeak at higher stroke rates and decreased its negativevalues during recovery This reflected the increaseddrag resistance at higher boat speeds

32 Definition of the microphases of thestroke cycle

The accelerations of the CM of the boat rower andthe whole system (Aboat Arow and Asys) were used to

define the microphases of the stroke cycle Thesevariables were chosen because they are related to theaccumulation of kinetic energy in the two majorcomponents of the rowerndashboat system and hence tothe effectiveness of rowing technique The wordlsquoeffectivenessrsquo here means the capability of produ-cing an effect ie the manner of application of therowerrsquos efforts is such that it provides the maximalaverage speed of the boatndashrower system The effec-tiveness differs from the lsquoefficiencyrsquo which is a ratioof the output effect (boat speed) to the input energy(rowerrsquos power) and is not discussed in this paperWhile the efficiency is easy to measure the effec-tiveness is difficult to quantify because the boatspeed depends on many factors which are beyondour control namely weather conditions rowerrsquosefforts and capabilities Therefore in this article theterm effectiveness is used in a qualitative way and itsevaluation is based on the overall performance of thecrew

The handle velocity Vh was used to define theoverall drive phase boundaries Figure 4 shows typi-cal biomechanical variables of a single sculler

Fig 3 Dependences of (a) the negative peak of the boat acceleration at catch and (b) the first positivepeak of the boat acceleration during the drive (b) on the stroke rate

Fig 2 Patterns of acceleration of (a) the boat and (b) the rowersrsquo CM at different stroke rates for anOlympic Champion menrsquos pair

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 67

Six microphases of the drive phase (D1 to D6) andthree microphases of the recovery phase (R1 to R3)were defined Table 2 presents the microphase iden-tifications descriptions and key events which definetheir start and end points

33 Dependence of the microphase structureon the boat type rower gender strokerate and force profile

Microphases D1 D2 and D3 were slightly longer insweep boats than in sculling boats and microphases

Fig 4 Typical biomechanical parameters and microphases of the stroke cycle (M1x menrsquos single scullrate 32 strokesmin) Key events are indicated as open circles

Table 2 Characteristics of the microphases of the stroke cycle

Identification Microphase Start event Description

D1 Blade immersion Catch beginning of the drive Vh changes sign topositive

Asys and Aboat are negative but Arow is positiveFast increases in Vh and Vlegs

D2 Initial rowersrsquoacceleration

Asys becomes positive The centre of the bladecrosses the water level downwards

Fh and Aboat increase but Aboat is still negativeand lower than Arow

D3 Initial boatacceleration

Aboat becomes higher than Arow First positive peak of Aboat which becomeshigher than Arow Maximum of Vlegs

D4 Rowersrsquoacceleration

Aboat decreases and becomes lower than therowerrsquos acceleration

Fh Arow and Asys increase slowly Vlegs decreases

D5 Boat acceleration Aboat again becomes higher than Arow Fh Ffoot Arow and Asys decrease but Ffootdecreases faster than Fh which produces thehighest Aboat

D6 Blade removal Asys becomes negative The centre of the bladecrosses the water level upwards

Arow is negative and Aboat is close to zero Vh isstill positive Varms is maximal

R1 Arms and trunkreturn

Finish end of the drive Vh changes sign tonegative

A quick positive peak of Aboat and negative Arowcaused by transfer of the moment of inertiafrom the rower to the boat

R2 Legs return Seat starts to move towards the stern Increase inAboat and decrease in Arow

Aboat is positive but Arow and Asys are negativeVlegs increases towards the stern

R3 Catch preparation Aboat becomes negative and Arow becomespositive

Ffoot increases which causes a decrease in Vlegsdeceleration of the boat and acceleration of therower

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

68 V Kleshnev

D4 D5 and D6 were slightly shorter (plt 005)(Fig 5) The longer duration of D1 in sweep rowingmight be due to the greater width of the sweep oarblade which requires more time to insert into thewater The shorter duration of D2 and D3 in scullingmight be due to the longer catch angles which cre-ates heavier effective gearing [9 10] A heavier gear-ing produces a lower ratio of handle to gate forceswhich creates higher acceleration of the boat com-pared with the rowerrsquos CM

There were no significant differences in the dura-tions of the D1 D2 D3 and D6 microphases amongboats of different sizes (Fig 6) D4 was significantlylonger in large boats (eights and quads) and D5 waslonger in small boats (singles and pairs plt 001) butthe reasons for these differences are not known

Male crews had a longer D3 microphase thanfemale crews did (plt 001) which is probably due totheir more rapid generation of forces Differencesbetween other microphases for the two genders werenot statistically significant

The absolute duration of all microphases exceptfor D3 tended to decrease at higher stroke rates Thecorrelation coefficients ranged from ndash010 for D4 tondash070 for D1 This decrease in duration was stronglycorrelated with a decrease in total drive time athigher stroke rates (rfrac14 ndash087)

The relative share of the D1 microphase within thedrive phase tended to decrease at higher stroke rates

(rfrac14 ndash049) whereas the relative share of the D6microphase tended to increase (Fig 7) Sculling crewstended to exhibit an absolute maximum in the rela-tive share of the D3 microphase at a stroke rate of30 strokesmin and sweep crews at 36 strokesminThe reason for this phenomenon is not known

To analyse the relationship between microphaseproportions and the force curve profile there was thetime from the catch until the moment when the forceVh achieved 10 per cent 20 per cent 100 per centof its maximal value The following statistically sig-nificant relationships were found (Fig 8)

1 The D1 period had the strongest positive correla-tion with the time for achieving 30 per cent of themaximal force (rfrac14 091 plt 0001)

2 D3 had the strongest negative correlation with thetime for achieving 70 per cent (rfrac14 ndash045 plt 005)and 80 per cent (rfrac14 ndash046 plt 005) of the max-imal force

3 D6 had a negative correlation with the timetaken to achieve 30 per cent of the maximal force(rfrac14 ndash046 plt 005)

34 Case study the importance of theD3 microphase

To discuss the influence of different rowing techni-que factors on the temporal structure of the drive

Fig 6 Time structures of the drive phase in small (1middot single scull 2pair) medium (2middotdouble scull4 four) and large (4middotquadruple scull 8thorn eight) boat types

Fig 5 Time structures of the drive phase in sweep and sculling crews

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 69

phase two crews of international level were selectedThe first crew consisted of Olympic Champions andwas considered by experts to be highly technicallyeffective The second crew was in the finals of theWorld Championship had very good physiologicalwork capacity but could not achieve excellent per-formance on the water Both samples were taken at astroke rate close to 32 strokesmin The first crewproduced a much quicker rise in the handle force(Fig 9(a)) achieving 70 per cent of their maximalforce at 231 per cent of the drive time The secondcrew had a relatively flatter force curve (70 per cent ofthe maximal force at 291 per cent of the drive time)but higher maximal and average forces

The faster rise in the force of the first crew coin-cides with a faster increase in the velocity Vlegs of thelegs (Fig 9(d)) (the maximal velocity of the legsoccurred at 224 per cent of the drive time) andwas accompanied by a steady increase in the trunkvelocity Vtrunk (Fig 9(f)) The second crew had aslower velocity of the legs in the first half of the drive

(a)

(b)

(c)

10

15

20

25

30

15 25 35

Rate (minndash1)

0

5

10

15

20

25

16 20 24 28 32 36 40

Sweep

Scull

Rate (minndash1)

)( e

mit evird ni 3D fo erah

S

0

5

10

15

20

15 25 35Rate (minndash1)

45

45

)( e

mit evird ni 6D f o erah

S )

( emit evird ni 1

D fo erahS

Fig 7 Dependences of the microphase time shares in thedrive time on the stroke rate (a) (b) linear trends(c) second-order polynomial trend

Fig 8 Dependences of the proportion of microphasewithin the drive phase on the stroke rate with lineartrend lines

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

70 V Kleshnev

(the maximal velocity of the legs at 322 per cent ofthe drive time) and a double-peaked trunk speedprofile Consequently the handle velocity Vh of thefirst crew increased more rapidly during the first halfof the drive (Fig 9(b))

The boat acceleration Aboat of the first crew had adeeper and later negative peak at the catch (84ms2

compared with 73ms2 in the second crew) but amuch quicker increase afterwards (Fig 9(c)) The firstpeak of Aboat was 27ms2 in the first crew but only03ms2 in the second crew which had Aboat dropbelow zero after the first peak

The first crew had a higher acceleration Arow ofthe rowerrsquos CM at the catch but the second crew

Fig 9 Biomechanical parameters and temporal structures of the drive phase for two examples of rowingtechnique

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 71

accelerated their body mass more intensively duringthe middle of the drive (Fig 9(e)) As a result thesecond crew did not have microphase D3 (initial boatacceleration) at all but it had a much longer D4(main rowersrsquo acceleration)

4 DISCUSSION AND IMPLICATIONS TEMPORALANALYSIS AND THE EFFECTIVENESS OF THEROWING TECHNIQUE

The main measure of rowing performance is theaverage speed of the rowerndashboat system Rowers can-not increase the system velocity during the recoveryphase because its deceleration is affected by suchenvironmental factors as water and air resistanceThe drive phase is the only time when rowers canincrease the system velocity (more precisely onlythat part of the drive when the propulsive force ishigher than the drag force) The higher accelera-tion of the system during the drive phase meansa higher average speed and hence better perfor-mance The system accumulates kinetic energy dur-ing the drive phase and loses it during the recoveryphase

The gain dE in kinetic energy of any body can bedefined using the equation

dE frac14 mv222

mv212

frac14 m

2ethv22 v21THORN eth13THORN

wherem is the mass of the body and v1 and v2 are thestart and end velocities respectively of the bodyrsquos CM

The mass of the rower is significantly greater thanthat of the boat (the ratio ranges from 400 per cent inlightweight women to 600 per cent in heavyweightmen) To maximize the average speed the rowersmust maximize the gain in kinetic energy which canbe achieved most effectively by means of accelerationof the heaviest part of the system (ie the rowerrsquosCM) Therefore the effectiveness of the rowing tech-

nique is related to the magnitude and timing of theacceleration of the rowerrsquos CM

The acceleration of any body depends on the forceacting on it The force Frow applied to the rowerrsquos CM(Fig 10) is equal to the sum of the foot-stretcherreaction force Frfoot which is positive ie pushesthe athletersquos CM forwards and the handle reactionforce Frhandle which is negative ie pulls the rowerbackwards according to

F row frac14 F rfoot thorn F rhandle eth14THORNBoth these reaction forces have the same magnitudeand the opposite direction to the action forces Ffootand Fh

The force Fboat applied to the boat hull is equal tothe sum of the gate force Fgate the stretcher forceFfoot and the drag force Fdrag (ignoring the small seatfrictional force) according to

Fboat frac14 Fgate thorn F foot thorn Fdrag eth15THORNTherefore emphasis on the boat accelerationrequires rowers to produce a higher gate force Fgate(and a correspondingly higher handle force Fh) andto push on the stretcher less Vice versa emphasison accelerating the rowerrsquos CM requires a higherstretcher force Ffoot and lower handle force FhGraphs of the gate and stretcher forces in Fig 4 con-firm that this switching of the emphasis happenstwice during the drive

The stretcher force Ffoot is the primary focus duringthe drive because it increases the acceleration Arow ofthe rowerrsquos CM which provides the most significantaccumulation of kinetic energy and hence increasesthe systemrsquos average speed

However the effectiveness of the rowerrsquos push onthe stretcher is affected by the boat velocity at thattime If the stretcher (boat) has a higher velocity thenthe rowers can accelerate their CM more efficientlythan if the boat is moving more slowly It can bespeculated that using the elastic energy stored in the

Frower

Fhandle

F foot

Fgate

Fbl

M

mFb

F drag

Fig 10 The main forces in the rowerndashboatndashoar system

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

72 V Kleshnev

oar handle and in the rowerrsquos tendons and musclescan play a significant role at the beginning of thedrive However this topic is beyond the scope of thispaper and can be the subject of further research Theuse of elastic energy is similar to jumping on atrampoline the point of support moves in the direc-tion of acceleration of the athletersquos CM and adds thevelocity of the support point to that of the athletersquosCM which makes the acceleration more effectiveThe presence of D3 could be called a lsquotrampoliningeffectrsquo and can be regarded as an important indicatorof the effectiveness of the rowing technique [11]In addition earlier force application could help toutilize the hydrolift effect better [12]

The practical implications of the microphases ofthe drive phase are as follows

During microphases D1 and D2 the rowers empha-size push with legs because they have to changethe direction of their movement from the sternto the bow at the catch This accelerates their bodymass and decelerates the boat At the same timethe blade obtains a grip on the water

During D3 the rowers emphasize pulling thehandle to accelerate the boat This creates a faster-moving support on the stretcher This microphase isimportant for performing an effective drive phase Itwas found that in 43 per cent of the crews thismicrophase was absent but a longer duration of D3does not mean better technique because D3 was thelongest in some low-level crews (according to anexpertrsquos opinion)

During D4 the rowers emphasize the stretcherpush again to accelerate themselves and hence toaccumulate kinetic energy The effectiveness of thismicrophase depends on the amount of boat velocitygained during the previous microphase D3 and on afast powerful leg drive

The microphases D5 and D6 occur after the legs aresubstantially extended and use trunk and arm workinstead The force levels and the total system accel-eration decrease but the stretcher force decreasesmore quickly and becomes negative (pull on thestretcher) This puts emphasis on the boat accelera-tion again The rowerrsquos CM acceleration becomesnegative which means transferring kinetic energyfrom the rowerrsquos mass to the boat

5 CONCLUSIONS

The phenomenon of a double peak in the boatacceleration during the drive phase can be explainedby double switching of the emphasis from pushingthe stretcher (more rowersrsquo CM acceleration) topulling the handle (more boat hull acceleration)Acceleration of the boat and of the rowersrsquo CM andthe velocity of the oar handle were used to define the

temporal structure of the stroke cycle Six micro-phases were defined during the drive phase and threemicrophases were defined during the recovery phaseThe presence of the microphase D3 (initial boatacceleration) was an important indicator of the eff-ectiveness of a rowerrsquos technique The initial boatacceleration creates a faster-moving platform for thestretcher to allow better acceleration of the rowerrsquosCM

ACKNOWLEDGEMENT

This research project was kindly supported by theAustralian Institute of Sport and Rowing Australia

Author 2010

REFERENCES

1 Bartlett R Sport biomechanics preventing injury andimproving performance 1999 pp 11ndash21 (RoutledgeNew York)

2 Toussaint H M and Beek P J Biomechanics ofcompetitive front crawl swimming Sports Med 199213 8ndash24

3 James S L and Brubaker C E Biomechanical andneuromuscular aspects of running Exercise Sport SciRev 1973 1 189ndash216

4 Dal-Monte A and Komor A Rowing and scullingmechanics In Biomechanics of sport (Ed C L Vaughan)1989 pp 53ndash119 (CRC Press Boca Raton Florida)

5 Rodriguez R J Rogriguez R P Cook S D andSandborn P M Electromyographic analysis of rowingstroke biomechanics J Sports Med Phys Fitness 199030 103ndash108

6 Shimoda M Kawakami Y and Fukunaga T Anapplication of acceleration analysis to evaluation of arowing technique In Proceedings of the 13th Interna-tional Symposium on Biomechanics in sports (EdT Bauer) Thunder Bay Ontario Canada 18ndash22 July1995 pp 95ndash100 (Lakehead University Thunder BayOntario)

7 Federation Internationale des Societes drsquoAviron (Inter-national Rowing Federation) FISA rules of racingand related bye-laws Part IV Boats and construction2005ndash2006 pp 60ndash66 available from httpwwwworldrowingcommediasdocsmedia_352459pdf

8 Geoscience Australia Applying geoscience to Australiarsquosmost important challenges 2009 available from httpwwwgeogovau

9 Kleshnev V Technology for technique improvementIn Rowing faster (Ed V Nolte) 2004 pp 209ndash228(Human Kinetics Champaign Illinois)

10 Kleshnev V Facts Did you know that Rowing Bio-mech Newslett 2007 7(72) available from httpwwwbiorowcomRBN_en_2007_files2007RowBiomNews03pdf

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 73

(Fig 2(a)) usually had one negative peak at the catchand two positive peaks during the drive The accel-eration Arow of the rowersrsquo CM (Fig 2(b)) had twopositive peaks one just after the catch and one in themiddle of the drive The system acceleration Asys

reflected the force application pattern and followedthe shape of the force curve

The negative peak of the boat acceleration Aboat

significantly increased in magnitude with increasingstroke rate (Fig 3(a)) (correlation with stroke raterfrac14 ndash085plt 001) This can be explained by thesubstantial increase in inertial forces at higher strokerates In contrast the first positive peak of Aboat

increased less at higher stroke rates (Fig 3(b)) Thesystem acceleration Asys usually increased its positivepeak at higher stroke rates and decreased its negativevalues during recovery This reflected the increaseddrag resistance at higher boat speeds

32 Definition of the microphases of thestroke cycle

The accelerations of the CM of the boat rower andthe whole system (Aboat Arow and Asys) were used to

define the microphases of the stroke cycle Thesevariables were chosen because they are related to theaccumulation of kinetic energy in the two majorcomponents of the rowerndashboat system and hence tothe effectiveness of rowing technique The wordlsquoeffectivenessrsquo here means the capability of produ-cing an effect ie the manner of application of therowerrsquos efforts is such that it provides the maximalaverage speed of the boatndashrower system The effec-tiveness differs from the lsquoefficiencyrsquo which is a ratioof the output effect (boat speed) to the input energy(rowerrsquos power) and is not discussed in this paperWhile the efficiency is easy to measure the effec-tiveness is difficult to quantify because the boatspeed depends on many factors which are beyondour control namely weather conditions rowerrsquosefforts and capabilities Therefore in this article theterm effectiveness is used in a qualitative way and itsevaluation is based on the overall performance of thecrew

The handle velocity Vh was used to define theoverall drive phase boundaries Figure 4 shows typi-cal biomechanical variables of a single sculler

Fig 3 Dependences of (a) the negative peak of the boat acceleration at catch and (b) the first positivepeak of the boat acceleration during the drive (b) on the stroke rate

Fig 2 Patterns of acceleration of (a) the boat and (b) the rowersrsquo CM at different stroke rates for anOlympic Champion menrsquos pair

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 67

Six microphases of the drive phase (D1 to D6) andthree microphases of the recovery phase (R1 to R3)were defined Table 2 presents the microphase iden-tifications descriptions and key events which definetheir start and end points

33 Dependence of the microphase structureon the boat type rower gender strokerate and force profile

Microphases D1 D2 and D3 were slightly longer insweep boats than in sculling boats and microphases

Fig 4 Typical biomechanical parameters and microphases of the stroke cycle (M1x menrsquos single scullrate 32 strokesmin) Key events are indicated as open circles

Table 2 Characteristics of the microphases of the stroke cycle

Identification Microphase Start event Description

D1 Blade immersion Catch beginning of the drive Vh changes sign topositive

Asys and Aboat are negative but Arow is positiveFast increases in Vh and Vlegs

D2 Initial rowersrsquoacceleration

Asys becomes positive The centre of the bladecrosses the water level downwards

Fh and Aboat increase but Aboat is still negativeand lower than Arow

D3 Initial boatacceleration

Aboat becomes higher than Arow First positive peak of Aboat which becomeshigher than Arow Maximum of Vlegs

D4 Rowersrsquoacceleration

Aboat decreases and becomes lower than therowerrsquos acceleration

Fh Arow and Asys increase slowly Vlegs decreases

D5 Boat acceleration Aboat again becomes higher than Arow Fh Ffoot Arow and Asys decrease but Ffootdecreases faster than Fh which produces thehighest Aboat

D6 Blade removal Asys becomes negative The centre of the bladecrosses the water level upwards

Arow is negative and Aboat is close to zero Vh isstill positive Varms is maximal

R1 Arms and trunkreturn

Finish end of the drive Vh changes sign tonegative

A quick positive peak of Aboat and negative Arowcaused by transfer of the moment of inertiafrom the rower to the boat

R2 Legs return Seat starts to move towards the stern Increase inAboat and decrease in Arow

Aboat is positive but Arow and Asys are negativeVlegs increases towards the stern

R3 Catch preparation Aboat becomes negative and Arow becomespositive

Ffoot increases which causes a decrease in Vlegsdeceleration of the boat and acceleration of therower

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

68 V Kleshnev

D4 D5 and D6 were slightly shorter (plt 005)(Fig 5) The longer duration of D1 in sweep rowingmight be due to the greater width of the sweep oarblade which requires more time to insert into thewater The shorter duration of D2 and D3 in scullingmight be due to the longer catch angles which cre-ates heavier effective gearing [9 10] A heavier gear-ing produces a lower ratio of handle to gate forceswhich creates higher acceleration of the boat com-pared with the rowerrsquos CM

There were no significant differences in the dura-tions of the D1 D2 D3 and D6 microphases amongboats of different sizes (Fig 6) D4 was significantlylonger in large boats (eights and quads) and D5 waslonger in small boats (singles and pairs plt 001) butthe reasons for these differences are not known

Male crews had a longer D3 microphase thanfemale crews did (plt 001) which is probably due totheir more rapid generation of forces Differencesbetween other microphases for the two genders werenot statistically significant

The absolute duration of all microphases exceptfor D3 tended to decrease at higher stroke rates Thecorrelation coefficients ranged from ndash010 for D4 tondash070 for D1 This decrease in duration was stronglycorrelated with a decrease in total drive time athigher stroke rates (rfrac14 ndash087)

The relative share of the D1 microphase within thedrive phase tended to decrease at higher stroke rates

(rfrac14 ndash049) whereas the relative share of the D6microphase tended to increase (Fig 7) Sculling crewstended to exhibit an absolute maximum in the rela-tive share of the D3 microphase at a stroke rate of30 strokesmin and sweep crews at 36 strokesminThe reason for this phenomenon is not known

To analyse the relationship between microphaseproportions and the force curve profile there was thetime from the catch until the moment when the forceVh achieved 10 per cent 20 per cent 100 per centof its maximal value The following statistically sig-nificant relationships were found (Fig 8)

1 The D1 period had the strongest positive correla-tion with the time for achieving 30 per cent of themaximal force (rfrac14 091 plt 0001)

2 D3 had the strongest negative correlation with thetime for achieving 70 per cent (rfrac14 ndash045 plt 005)and 80 per cent (rfrac14 ndash046 plt 005) of the max-imal force

3 D6 had a negative correlation with the timetaken to achieve 30 per cent of the maximal force(rfrac14 ndash046 plt 005)

34 Case study the importance of theD3 microphase

To discuss the influence of different rowing techni-que factors on the temporal structure of the drive

Fig 6 Time structures of the drive phase in small (1middot single scull 2pair) medium (2middotdouble scull4 four) and large (4middotquadruple scull 8thorn eight) boat types

Fig 5 Time structures of the drive phase in sweep and sculling crews

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 69

phase two crews of international level were selectedThe first crew consisted of Olympic Champions andwas considered by experts to be highly technicallyeffective The second crew was in the finals of theWorld Championship had very good physiologicalwork capacity but could not achieve excellent per-formance on the water Both samples were taken at astroke rate close to 32 strokesmin The first crewproduced a much quicker rise in the handle force(Fig 9(a)) achieving 70 per cent of their maximalforce at 231 per cent of the drive time The secondcrew had a relatively flatter force curve (70 per cent ofthe maximal force at 291 per cent of the drive time)but higher maximal and average forces

The faster rise in the force of the first crew coin-cides with a faster increase in the velocity Vlegs of thelegs (Fig 9(d)) (the maximal velocity of the legsoccurred at 224 per cent of the drive time) andwas accompanied by a steady increase in the trunkvelocity Vtrunk (Fig 9(f)) The second crew had aslower velocity of the legs in the first half of the drive

(a)

(b)

(c)

10

15

20

25

30

15 25 35

Rate (minndash1)

0

5

10

15

20

25

16 20 24 28 32 36 40

Sweep

Scull

Rate (minndash1)

)( e

mit evird ni 3D fo erah

S

0

5

10

15

20

15 25 35Rate (minndash1)

45

45

)( e

mit evird ni 6D f o erah

S )

( emit evird ni 1

D fo erahS

Fig 7 Dependences of the microphase time shares in thedrive time on the stroke rate (a) (b) linear trends(c) second-order polynomial trend

Fig 8 Dependences of the proportion of microphasewithin the drive phase on the stroke rate with lineartrend lines

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

70 V Kleshnev

(the maximal velocity of the legs at 322 per cent ofthe drive time) and a double-peaked trunk speedprofile Consequently the handle velocity Vh of thefirst crew increased more rapidly during the first halfof the drive (Fig 9(b))

The boat acceleration Aboat of the first crew had adeeper and later negative peak at the catch (84ms2

compared with 73ms2 in the second crew) but amuch quicker increase afterwards (Fig 9(c)) The firstpeak of Aboat was 27ms2 in the first crew but only03ms2 in the second crew which had Aboat dropbelow zero after the first peak

The first crew had a higher acceleration Arow ofthe rowerrsquos CM at the catch but the second crew

Fig 9 Biomechanical parameters and temporal structures of the drive phase for two examples of rowingtechnique

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 71

accelerated their body mass more intensively duringthe middle of the drive (Fig 9(e)) As a result thesecond crew did not have microphase D3 (initial boatacceleration) at all but it had a much longer D4(main rowersrsquo acceleration)

4 DISCUSSION AND IMPLICATIONS TEMPORALANALYSIS AND THE EFFECTIVENESS OF THEROWING TECHNIQUE

The main measure of rowing performance is theaverage speed of the rowerndashboat system Rowers can-not increase the system velocity during the recoveryphase because its deceleration is affected by suchenvironmental factors as water and air resistanceThe drive phase is the only time when rowers canincrease the system velocity (more precisely onlythat part of the drive when the propulsive force ishigher than the drag force) The higher accelera-tion of the system during the drive phase meansa higher average speed and hence better perfor-mance The system accumulates kinetic energy dur-ing the drive phase and loses it during the recoveryphase

The gain dE in kinetic energy of any body can bedefined using the equation

dE frac14 mv222

mv212

frac14 m

2ethv22 v21THORN eth13THORN

wherem is the mass of the body and v1 and v2 are thestart and end velocities respectively of the bodyrsquos CM

The mass of the rower is significantly greater thanthat of the boat (the ratio ranges from 400 per cent inlightweight women to 600 per cent in heavyweightmen) To maximize the average speed the rowersmust maximize the gain in kinetic energy which canbe achieved most effectively by means of accelerationof the heaviest part of the system (ie the rowerrsquosCM) Therefore the effectiveness of the rowing tech-

nique is related to the magnitude and timing of theacceleration of the rowerrsquos CM

The acceleration of any body depends on the forceacting on it The force Frow applied to the rowerrsquos CM(Fig 10) is equal to the sum of the foot-stretcherreaction force Frfoot which is positive ie pushesthe athletersquos CM forwards and the handle reactionforce Frhandle which is negative ie pulls the rowerbackwards according to

F row frac14 F rfoot thorn F rhandle eth14THORNBoth these reaction forces have the same magnitudeand the opposite direction to the action forces Ffootand Fh

The force Fboat applied to the boat hull is equal tothe sum of the gate force Fgate the stretcher forceFfoot and the drag force Fdrag (ignoring the small seatfrictional force) according to

Fboat frac14 Fgate thorn F foot thorn Fdrag eth15THORNTherefore emphasis on the boat accelerationrequires rowers to produce a higher gate force Fgate(and a correspondingly higher handle force Fh) andto push on the stretcher less Vice versa emphasison accelerating the rowerrsquos CM requires a higherstretcher force Ffoot and lower handle force FhGraphs of the gate and stretcher forces in Fig 4 con-firm that this switching of the emphasis happenstwice during the drive

The stretcher force Ffoot is the primary focus duringthe drive because it increases the acceleration Arow ofthe rowerrsquos CM which provides the most significantaccumulation of kinetic energy and hence increasesthe systemrsquos average speed

However the effectiveness of the rowerrsquos push onthe stretcher is affected by the boat velocity at thattime If the stretcher (boat) has a higher velocity thenthe rowers can accelerate their CM more efficientlythan if the boat is moving more slowly It can bespeculated that using the elastic energy stored in the

Frower

Fhandle

F foot

Fgate

Fbl

M

mFb

F drag

Fig 10 The main forces in the rowerndashboatndashoar system

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

72 V Kleshnev

oar handle and in the rowerrsquos tendons and musclescan play a significant role at the beginning of thedrive However this topic is beyond the scope of thispaper and can be the subject of further research Theuse of elastic energy is similar to jumping on atrampoline the point of support moves in the direc-tion of acceleration of the athletersquos CM and adds thevelocity of the support point to that of the athletersquosCM which makes the acceleration more effectiveThe presence of D3 could be called a lsquotrampoliningeffectrsquo and can be regarded as an important indicatorof the effectiveness of the rowing technique [11]In addition earlier force application could help toutilize the hydrolift effect better [12]

The practical implications of the microphases ofthe drive phase are as follows

During microphases D1 and D2 the rowers empha-size push with legs because they have to changethe direction of their movement from the sternto the bow at the catch This accelerates their bodymass and decelerates the boat At the same timethe blade obtains a grip on the water

During D3 the rowers emphasize pulling thehandle to accelerate the boat This creates a faster-moving support on the stretcher This microphase isimportant for performing an effective drive phase Itwas found that in 43 per cent of the crews thismicrophase was absent but a longer duration of D3does not mean better technique because D3 was thelongest in some low-level crews (according to anexpertrsquos opinion)

During D4 the rowers emphasize the stretcherpush again to accelerate themselves and hence toaccumulate kinetic energy The effectiveness of thismicrophase depends on the amount of boat velocitygained during the previous microphase D3 and on afast powerful leg drive

The microphases D5 and D6 occur after the legs aresubstantially extended and use trunk and arm workinstead The force levels and the total system accel-eration decrease but the stretcher force decreasesmore quickly and becomes negative (pull on thestretcher) This puts emphasis on the boat accelera-tion again The rowerrsquos CM acceleration becomesnegative which means transferring kinetic energyfrom the rowerrsquos mass to the boat

5 CONCLUSIONS

The phenomenon of a double peak in the boatacceleration during the drive phase can be explainedby double switching of the emphasis from pushingthe stretcher (more rowersrsquo CM acceleration) topulling the handle (more boat hull acceleration)Acceleration of the boat and of the rowersrsquo CM andthe velocity of the oar handle were used to define the

temporal structure of the stroke cycle Six micro-phases were defined during the drive phase and threemicrophases were defined during the recovery phaseThe presence of the microphase D3 (initial boatacceleration) was an important indicator of the eff-ectiveness of a rowerrsquos technique The initial boatacceleration creates a faster-moving platform for thestretcher to allow better acceleration of the rowerrsquosCM

ACKNOWLEDGEMENT

This research project was kindly supported by theAustralian Institute of Sport and Rowing Australia

Author 2010

REFERENCES

1 Bartlett R Sport biomechanics preventing injury andimproving performance 1999 pp 11ndash21 (RoutledgeNew York)

2 Toussaint H M and Beek P J Biomechanics ofcompetitive front crawl swimming Sports Med 199213 8ndash24

3 James S L and Brubaker C E Biomechanical andneuromuscular aspects of running Exercise Sport SciRev 1973 1 189ndash216

4 Dal-Monte A and Komor A Rowing and scullingmechanics In Biomechanics of sport (Ed C L Vaughan)1989 pp 53ndash119 (CRC Press Boca Raton Florida)

5 Rodriguez R J Rogriguez R P Cook S D andSandborn P M Electromyographic analysis of rowingstroke biomechanics J Sports Med Phys Fitness 199030 103ndash108

6 Shimoda M Kawakami Y and Fukunaga T Anapplication of acceleration analysis to evaluation of arowing technique In Proceedings of the 13th Interna-tional Symposium on Biomechanics in sports (EdT Bauer) Thunder Bay Ontario Canada 18ndash22 July1995 pp 95ndash100 (Lakehead University Thunder BayOntario)

7 Federation Internationale des Societes drsquoAviron (Inter-national Rowing Federation) FISA rules of racingand related bye-laws Part IV Boats and construction2005ndash2006 pp 60ndash66 available from httpwwwworldrowingcommediasdocsmedia_352459pdf

8 Geoscience Australia Applying geoscience to Australiarsquosmost important challenges 2009 available from httpwwwgeogovau

9 Kleshnev V Technology for technique improvementIn Rowing faster (Ed V Nolte) 2004 pp 209ndash228(Human Kinetics Champaign Illinois)

10 Kleshnev V Facts Did you know that Rowing Bio-mech Newslett 2007 7(72) available from httpwwwbiorowcomRBN_en_2007_files2007RowBiomNews03pdf

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 73

Six microphases of the drive phase (D1 to D6) andthree microphases of the recovery phase (R1 to R3)were defined Table 2 presents the microphase iden-tifications descriptions and key events which definetheir start and end points

33 Dependence of the microphase structureon the boat type rower gender strokerate and force profile

Microphases D1 D2 and D3 were slightly longer insweep boats than in sculling boats and microphases

Fig 4 Typical biomechanical parameters and microphases of the stroke cycle (M1x menrsquos single scullrate 32 strokesmin) Key events are indicated as open circles

Table 2 Characteristics of the microphases of the stroke cycle

Identification Microphase Start event Description

D1 Blade immersion Catch beginning of the drive Vh changes sign topositive

Asys and Aboat are negative but Arow is positiveFast increases in Vh and Vlegs

D2 Initial rowersrsquoacceleration

Asys becomes positive The centre of the bladecrosses the water level downwards

Fh and Aboat increase but Aboat is still negativeand lower than Arow

D3 Initial boatacceleration

Aboat becomes higher than Arow First positive peak of Aboat which becomeshigher than Arow Maximum of Vlegs

D4 Rowersrsquoacceleration

Aboat decreases and becomes lower than therowerrsquos acceleration

Fh Arow and Asys increase slowly Vlegs decreases

D5 Boat acceleration Aboat again becomes higher than Arow Fh Ffoot Arow and Asys decrease but Ffootdecreases faster than Fh which produces thehighest Aboat

D6 Blade removal Asys becomes negative The centre of the bladecrosses the water level upwards

Arow is negative and Aboat is close to zero Vh isstill positive Varms is maximal

R1 Arms and trunkreturn

Finish end of the drive Vh changes sign tonegative

A quick positive peak of Aboat and negative Arowcaused by transfer of the moment of inertiafrom the rower to the boat

R2 Legs return Seat starts to move towards the stern Increase inAboat and decrease in Arow

Aboat is positive but Arow and Asys are negativeVlegs increases towards the stern

R3 Catch preparation Aboat becomes negative and Arow becomespositive

Ffoot increases which causes a decrease in Vlegsdeceleration of the boat and acceleration of therower

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

68 V Kleshnev

D4 D5 and D6 were slightly shorter (plt 005)(Fig 5) The longer duration of D1 in sweep rowingmight be due to the greater width of the sweep oarblade which requires more time to insert into thewater The shorter duration of D2 and D3 in scullingmight be due to the longer catch angles which cre-ates heavier effective gearing [9 10] A heavier gear-ing produces a lower ratio of handle to gate forceswhich creates higher acceleration of the boat com-pared with the rowerrsquos CM

There were no significant differences in the dura-tions of the D1 D2 D3 and D6 microphases amongboats of different sizes (Fig 6) D4 was significantlylonger in large boats (eights and quads) and D5 waslonger in small boats (singles and pairs plt 001) butthe reasons for these differences are not known

Male crews had a longer D3 microphase thanfemale crews did (plt 001) which is probably due totheir more rapid generation of forces Differencesbetween other microphases for the two genders werenot statistically significant

The absolute duration of all microphases exceptfor D3 tended to decrease at higher stroke rates Thecorrelation coefficients ranged from ndash010 for D4 tondash070 for D1 This decrease in duration was stronglycorrelated with a decrease in total drive time athigher stroke rates (rfrac14 ndash087)

The relative share of the D1 microphase within thedrive phase tended to decrease at higher stroke rates

(rfrac14 ndash049) whereas the relative share of the D6microphase tended to increase (Fig 7) Sculling crewstended to exhibit an absolute maximum in the rela-tive share of the D3 microphase at a stroke rate of30 strokesmin and sweep crews at 36 strokesminThe reason for this phenomenon is not known

To analyse the relationship between microphaseproportions and the force curve profile there was thetime from the catch until the moment when the forceVh achieved 10 per cent 20 per cent 100 per centof its maximal value The following statistically sig-nificant relationships were found (Fig 8)

1 The D1 period had the strongest positive correla-tion with the time for achieving 30 per cent of themaximal force (rfrac14 091 plt 0001)

2 D3 had the strongest negative correlation with thetime for achieving 70 per cent (rfrac14 ndash045 plt 005)and 80 per cent (rfrac14 ndash046 plt 005) of the max-imal force

3 D6 had a negative correlation with the timetaken to achieve 30 per cent of the maximal force(rfrac14 ndash046 plt 005)

34 Case study the importance of theD3 microphase

To discuss the influence of different rowing techni-que factors on the temporal structure of the drive

Fig 6 Time structures of the drive phase in small (1middot single scull 2pair) medium (2middotdouble scull4 four) and large (4middotquadruple scull 8thorn eight) boat types

Fig 5 Time structures of the drive phase in sweep and sculling crews

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 69

phase two crews of international level were selectedThe first crew consisted of Olympic Champions andwas considered by experts to be highly technicallyeffective The second crew was in the finals of theWorld Championship had very good physiologicalwork capacity but could not achieve excellent per-formance on the water Both samples were taken at astroke rate close to 32 strokesmin The first crewproduced a much quicker rise in the handle force(Fig 9(a)) achieving 70 per cent of their maximalforce at 231 per cent of the drive time The secondcrew had a relatively flatter force curve (70 per cent ofthe maximal force at 291 per cent of the drive time)but higher maximal and average forces

The faster rise in the force of the first crew coin-cides with a faster increase in the velocity Vlegs of thelegs (Fig 9(d)) (the maximal velocity of the legsoccurred at 224 per cent of the drive time) andwas accompanied by a steady increase in the trunkvelocity Vtrunk (Fig 9(f)) The second crew had aslower velocity of the legs in the first half of the drive

(a)

(b)

(c)

10

15

20

25

30

15 25 35

Rate (minndash1)

0

5

10

15

20

25

16 20 24 28 32 36 40

Sweep

Scull

Rate (minndash1)

)( e

mit evird ni 3D fo erah

S

0

5

10

15

20

15 25 35Rate (minndash1)

45

45

)( e

mit evird ni 6D f o erah

S )

( emit evird ni 1

D fo erahS

Fig 7 Dependences of the microphase time shares in thedrive time on the stroke rate (a) (b) linear trends(c) second-order polynomial trend

Fig 8 Dependences of the proportion of microphasewithin the drive phase on the stroke rate with lineartrend lines

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

70 V Kleshnev

(the maximal velocity of the legs at 322 per cent ofthe drive time) and a double-peaked trunk speedprofile Consequently the handle velocity Vh of thefirst crew increased more rapidly during the first halfof the drive (Fig 9(b))

The boat acceleration Aboat of the first crew had adeeper and later negative peak at the catch (84ms2

compared with 73ms2 in the second crew) but amuch quicker increase afterwards (Fig 9(c)) The firstpeak of Aboat was 27ms2 in the first crew but only03ms2 in the second crew which had Aboat dropbelow zero after the first peak

The first crew had a higher acceleration Arow ofthe rowerrsquos CM at the catch but the second crew

Fig 9 Biomechanical parameters and temporal structures of the drive phase for two examples of rowingtechnique

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 71

accelerated their body mass more intensively duringthe middle of the drive (Fig 9(e)) As a result thesecond crew did not have microphase D3 (initial boatacceleration) at all but it had a much longer D4(main rowersrsquo acceleration)

4 DISCUSSION AND IMPLICATIONS TEMPORALANALYSIS AND THE EFFECTIVENESS OF THEROWING TECHNIQUE

The main measure of rowing performance is theaverage speed of the rowerndashboat system Rowers can-not increase the system velocity during the recoveryphase because its deceleration is affected by suchenvironmental factors as water and air resistanceThe drive phase is the only time when rowers canincrease the system velocity (more precisely onlythat part of the drive when the propulsive force ishigher than the drag force) The higher accelera-tion of the system during the drive phase meansa higher average speed and hence better perfor-mance The system accumulates kinetic energy dur-ing the drive phase and loses it during the recoveryphase

The gain dE in kinetic energy of any body can bedefined using the equation

dE frac14 mv222

mv212

frac14 m

2ethv22 v21THORN eth13THORN

wherem is the mass of the body and v1 and v2 are thestart and end velocities respectively of the bodyrsquos CM

The mass of the rower is significantly greater thanthat of the boat (the ratio ranges from 400 per cent inlightweight women to 600 per cent in heavyweightmen) To maximize the average speed the rowersmust maximize the gain in kinetic energy which canbe achieved most effectively by means of accelerationof the heaviest part of the system (ie the rowerrsquosCM) Therefore the effectiveness of the rowing tech-

nique is related to the magnitude and timing of theacceleration of the rowerrsquos CM

The acceleration of any body depends on the forceacting on it The force Frow applied to the rowerrsquos CM(Fig 10) is equal to the sum of the foot-stretcherreaction force Frfoot which is positive ie pushesthe athletersquos CM forwards and the handle reactionforce Frhandle which is negative ie pulls the rowerbackwards according to

F row frac14 F rfoot thorn F rhandle eth14THORNBoth these reaction forces have the same magnitudeand the opposite direction to the action forces Ffootand Fh

The force Fboat applied to the boat hull is equal tothe sum of the gate force Fgate the stretcher forceFfoot and the drag force Fdrag (ignoring the small seatfrictional force) according to

Fboat frac14 Fgate thorn F foot thorn Fdrag eth15THORNTherefore emphasis on the boat accelerationrequires rowers to produce a higher gate force Fgate(and a correspondingly higher handle force Fh) andto push on the stretcher less Vice versa emphasison accelerating the rowerrsquos CM requires a higherstretcher force Ffoot and lower handle force FhGraphs of the gate and stretcher forces in Fig 4 con-firm that this switching of the emphasis happenstwice during the drive

The stretcher force Ffoot is the primary focus duringthe drive because it increases the acceleration Arow ofthe rowerrsquos CM which provides the most significantaccumulation of kinetic energy and hence increasesthe systemrsquos average speed

However the effectiveness of the rowerrsquos push onthe stretcher is affected by the boat velocity at thattime If the stretcher (boat) has a higher velocity thenthe rowers can accelerate their CM more efficientlythan if the boat is moving more slowly It can bespeculated that using the elastic energy stored in the

Frower

Fhandle

F foot

Fgate

Fbl

M

mFb

F drag

Fig 10 The main forces in the rowerndashboatndashoar system

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

72 V Kleshnev

oar handle and in the rowerrsquos tendons and musclescan play a significant role at the beginning of thedrive However this topic is beyond the scope of thispaper and can be the subject of further research Theuse of elastic energy is similar to jumping on atrampoline the point of support moves in the direc-tion of acceleration of the athletersquos CM and adds thevelocity of the support point to that of the athletersquosCM which makes the acceleration more effectiveThe presence of D3 could be called a lsquotrampoliningeffectrsquo and can be regarded as an important indicatorof the effectiveness of the rowing technique [11]In addition earlier force application could help toutilize the hydrolift effect better [12]

The practical implications of the microphases ofthe drive phase are as follows

During microphases D1 and D2 the rowers empha-size push with legs because they have to changethe direction of their movement from the sternto the bow at the catch This accelerates their bodymass and decelerates the boat At the same timethe blade obtains a grip on the water

During D3 the rowers emphasize pulling thehandle to accelerate the boat This creates a faster-moving support on the stretcher This microphase isimportant for performing an effective drive phase Itwas found that in 43 per cent of the crews thismicrophase was absent but a longer duration of D3does not mean better technique because D3 was thelongest in some low-level crews (according to anexpertrsquos opinion)

During D4 the rowers emphasize the stretcherpush again to accelerate themselves and hence toaccumulate kinetic energy The effectiveness of thismicrophase depends on the amount of boat velocitygained during the previous microphase D3 and on afast powerful leg drive

The microphases D5 and D6 occur after the legs aresubstantially extended and use trunk and arm workinstead The force levels and the total system accel-eration decrease but the stretcher force decreasesmore quickly and becomes negative (pull on thestretcher) This puts emphasis on the boat accelera-tion again The rowerrsquos CM acceleration becomesnegative which means transferring kinetic energyfrom the rowerrsquos mass to the boat

5 CONCLUSIONS

The phenomenon of a double peak in the boatacceleration during the drive phase can be explainedby double switching of the emphasis from pushingthe stretcher (more rowersrsquo CM acceleration) topulling the handle (more boat hull acceleration)Acceleration of the boat and of the rowersrsquo CM andthe velocity of the oar handle were used to define the

temporal structure of the stroke cycle Six micro-phases were defined during the drive phase and threemicrophases were defined during the recovery phaseThe presence of the microphase D3 (initial boatacceleration) was an important indicator of the eff-ectiveness of a rowerrsquos technique The initial boatacceleration creates a faster-moving platform for thestretcher to allow better acceleration of the rowerrsquosCM

ACKNOWLEDGEMENT

This research project was kindly supported by theAustralian Institute of Sport and Rowing Australia

Author 2010

REFERENCES

1 Bartlett R Sport biomechanics preventing injury andimproving performance 1999 pp 11ndash21 (RoutledgeNew York)

2 Toussaint H M and Beek P J Biomechanics ofcompetitive front crawl swimming Sports Med 199213 8ndash24

3 James S L and Brubaker C E Biomechanical andneuromuscular aspects of running Exercise Sport SciRev 1973 1 189ndash216

4 Dal-Monte A and Komor A Rowing and scullingmechanics In Biomechanics of sport (Ed C L Vaughan)1989 pp 53ndash119 (CRC Press Boca Raton Florida)

5 Rodriguez R J Rogriguez R P Cook S D andSandborn P M Electromyographic analysis of rowingstroke biomechanics J Sports Med Phys Fitness 199030 103ndash108

6 Shimoda M Kawakami Y and Fukunaga T Anapplication of acceleration analysis to evaluation of arowing technique In Proceedings of the 13th Interna-tional Symposium on Biomechanics in sports (EdT Bauer) Thunder Bay Ontario Canada 18ndash22 July1995 pp 95ndash100 (Lakehead University Thunder BayOntario)

7 Federation Internationale des Societes drsquoAviron (Inter-national Rowing Federation) FISA rules of racingand related bye-laws Part IV Boats and construction2005ndash2006 pp 60ndash66 available from httpwwwworldrowingcommediasdocsmedia_352459pdf

8 Geoscience Australia Applying geoscience to Australiarsquosmost important challenges 2009 available from httpwwwgeogovau

9 Kleshnev V Technology for technique improvementIn Rowing faster (Ed V Nolte) 2004 pp 209ndash228(Human Kinetics Champaign Illinois)

10 Kleshnev V Facts Did you know that Rowing Bio-mech Newslett 2007 7(72) available from httpwwwbiorowcomRBN_en_2007_files2007RowBiomNews03pdf

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 73

D4 D5 and D6 were slightly shorter (plt 005)(Fig 5) The longer duration of D1 in sweep rowingmight be due to the greater width of the sweep oarblade which requires more time to insert into thewater The shorter duration of D2 and D3 in scullingmight be due to the longer catch angles which cre-ates heavier effective gearing [9 10] A heavier gear-ing produces a lower ratio of handle to gate forceswhich creates higher acceleration of the boat com-pared with the rowerrsquos CM

There were no significant differences in the dura-tions of the D1 D2 D3 and D6 microphases amongboats of different sizes (Fig 6) D4 was significantlylonger in large boats (eights and quads) and D5 waslonger in small boats (singles and pairs plt 001) butthe reasons for these differences are not known

Male crews had a longer D3 microphase thanfemale crews did (plt 001) which is probably due totheir more rapid generation of forces Differencesbetween other microphases for the two genders werenot statistically significant

The absolute duration of all microphases exceptfor D3 tended to decrease at higher stroke rates Thecorrelation coefficients ranged from ndash010 for D4 tondash070 for D1 This decrease in duration was stronglycorrelated with a decrease in total drive time athigher stroke rates (rfrac14 ndash087)

The relative share of the D1 microphase within thedrive phase tended to decrease at higher stroke rates

(rfrac14 ndash049) whereas the relative share of the D6microphase tended to increase (Fig 7) Sculling crewstended to exhibit an absolute maximum in the rela-tive share of the D3 microphase at a stroke rate of30 strokesmin and sweep crews at 36 strokesminThe reason for this phenomenon is not known

To analyse the relationship between microphaseproportions and the force curve profile there was thetime from the catch until the moment when the forceVh achieved 10 per cent 20 per cent 100 per centof its maximal value The following statistically sig-nificant relationships were found (Fig 8)

1 The D1 period had the strongest positive correla-tion with the time for achieving 30 per cent of themaximal force (rfrac14 091 plt 0001)

2 D3 had the strongest negative correlation with thetime for achieving 70 per cent (rfrac14 ndash045 plt 005)and 80 per cent (rfrac14 ndash046 plt 005) of the max-imal force

3 D6 had a negative correlation with the timetaken to achieve 30 per cent of the maximal force(rfrac14 ndash046 plt 005)

34 Case study the importance of theD3 microphase

To discuss the influence of different rowing techni-que factors on the temporal structure of the drive

Fig 6 Time structures of the drive phase in small (1middot single scull 2pair) medium (2middotdouble scull4 four) and large (4middotquadruple scull 8thorn eight) boat types

Fig 5 Time structures of the drive phase in sweep and sculling crews

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 69

phase two crews of international level were selectedThe first crew consisted of Olympic Champions andwas considered by experts to be highly technicallyeffective The second crew was in the finals of theWorld Championship had very good physiologicalwork capacity but could not achieve excellent per-formance on the water Both samples were taken at astroke rate close to 32 strokesmin The first crewproduced a much quicker rise in the handle force(Fig 9(a)) achieving 70 per cent of their maximalforce at 231 per cent of the drive time The secondcrew had a relatively flatter force curve (70 per cent ofthe maximal force at 291 per cent of the drive time)but higher maximal and average forces

The faster rise in the force of the first crew coin-cides with a faster increase in the velocity Vlegs of thelegs (Fig 9(d)) (the maximal velocity of the legsoccurred at 224 per cent of the drive time) andwas accompanied by a steady increase in the trunkvelocity Vtrunk (Fig 9(f)) The second crew had aslower velocity of the legs in the first half of the drive

(a)

(b)

(c)

10

15

20

25

30

15 25 35

Rate (minndash1)

0

5

10

15

20

25

16 20 24 28 32 36 40

Sweep

Scull

Rate (minndash1)

)( e

mit evird ni 3D fo erah

S

0

5

10

15

20

15 25 35Rate (minndash1)

45

45

)( e

mit evird ni 6D f o erah

S )

( emit evird ni 1

D fo erahS

Fig 7 Dependences of the microphase time shares in thedrive time on the stroke rate (a) (b) linear trends(c) second-order polynomial trend

Fig 8 Dependences of the proportion of microphasewithin the drive phase on the stroke rate with lineartrend lines

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

70 V Kleshnev

(the maximal velocity of the legs at 322 per cent ofthe drive time) and a double-peaked trunk speedprofile Consequently the handle velocity Vh of thefirst crew increased more rapidly during the first halfof the drive (Fig 9(b))

The boat acceleration Aboat of the first crew had adeeper and later negative peak at the catch (84ms2

compared with 73ms2 in the second crew) but amuch quicker increase afterwards (Fig 9(c)) The firstpeak of Aboat was 27ms2 in the first crew but only03ms2 in the second crew which had Aboat dropbelow zero after the first peak

The first crew had a higher acceleration Arow ofthe rowerrsquos CM at the catch but the second crew

Fig 9 Biomechanical parameters and temporal structures of the drive phase for two examples of rowingtechnique

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 71

accelerated their body mass more intensively duringthe middle of the drive (Fig 9(e)) As a result thesecond crew did not have microphase D3 (initial boatacceleration) at all but it had a much longer D4(main rowersrsquo acceleration)

4 DISCUSSION AND IMPLICATIONS TEMPORALANALYSIS AND THE EFFECTIVENESS OF THEROWING TECHNIQUE

The main measure of rowing performance is theaverage speed of the rowerndashboat system Rowers can-not increase the system velocity during the recoveryphase because its deceleration is affected by suchenvironmental factors as water and air resistanceThe drive phase is the only time when rowers canincrease the system velocity (more precisely onlythat part of the drive when the propulsive force ishigher than the drag force) The higher accelera-tion of the system during the drive phase meansa higher average speed and hence better perfor-mance The system accumulates kinetic energy dur-ing the drive phase and loses it during the recoveryphase

The gain dE in kinetic energy of any body can bedefined using the equation

dE frac14 mv222

mv212

frac14 m

2ethv22 v21THORN eth13THORN

wherem is the mass of the body and v1 and v2 are thestart and end velocities respectively of the bodyrsquos CM

The mass of the rower is significantly greater thanthat of the boat (the ratio ranges from 400 per cent inlightweight women to 600 per cent in heavyweightmen) To maximize the average speed the rowersmust maximize the gain in kinetic energy which canbe achieved most effectively by means of accelerationof the heaviest part of the system (ie the rowerrsquosCM) Therefore the effectiveness of the rowing tech-

nique is related to the magnitude and timing of theacceleration of the rowerrsquos CM

The acceleration of any body depends on the forceacting on it The force Frow applied to the rowerrsquos CM(Fig 10) is equal to the sum of the foot-stretcherreaction force Frfoot which is positive ie pushesthe athletersquos CM forwards and the handle reactionforce Frhandle which is negative ie pulls the rowerbackwards according to

F row frac14 F rfoot thorn F rhandle eth14THORNBoth these reaction forces have the same magnitudeand the opposite direction to the action forces Ffootand Fh

The force Fboat applied to the boat hull is equal tothe sum of the gate force Fgate the stretcher forceFfoot and the drag force Fdrag (ignoring the small seatfrictional force) according to

Fboat frac14 Fgate thorn F foot thorn Fdrag eth15THORNTherefore emphasis on the boat accelerationrequires rowers to produce a higher gate force Fgate(and a correspondingly higher handle force Fh) andto push on the stretcher less Vice versa emphasison accelerating the rowerrsquos CM requires a higherstretcher force Ffoot and lower handle force FhGraphs of the gate and stretcher forces in Fig 4 con-firm that this switching of the emphasis happenstwice during the drive

The stretcher force Ffoot is the primary focus duringthe drive because it increases the acceleration Arow ofthe rowerrsquos CM which provides the most significantaccumulation of kinetic energy and hence increasesthe systemrsquos average speed

However the effectiveness of the rowerrsquos push onthe stretcher is affected by the boat velocity at thattime If the stretcher (boat) has a higher velocity thenthe rowers can accelerate their CM more efficientlythan if the boat is moving more slowly It can bespeculated that using the elastic energy stored in the

Frower

Fhandle

F foot

Fgate

Fbl

M

mFb

F drag

Fig 10 The main forces in the rowerndashboatndashoar system

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

72 V Kleshnev

oar handle and in the rowerrsquos tendons and musclescan play a significant role at the beginning of thedrive However this topic is beyond the scope of thispaper and can be the subject of further research Theuse of elastic energy is similar to jumping on atrampoline the point of support moves in the direc-tion of acceleration of the athletersquos CM and adds thevelocity of the support point to that of the athletersquosCM which makes the acceleration more effectiveThe presence of D3 could be called a lsquotrampoliningeffectrsquo and can be regarded as an important indicatorof the effectiveness of the rowing technique [11]In addition earlier force application could help toutilize the hydrolift effect better [12]

The practical implications of the microphases ofthe drive phase are as follows

During microphases D1 and D2 the rowers empha-size push with legs because they have to changethe direction of their movement from the sternto the bow at the catch This accelerates their bodymass and decelerates the boat At the same timethe blade obtains a grip on the water

During D3 the rowers emphasize pulling thehandle to accelerate the boat This creates a faster-moving support on the stretcher This microphase isimportant for performing an effective drive phase Itwas found that in 43 per cent of the crews thismicrophase was absent but a longer duration of D3does not mean better technique because D3 was thelongest in some low-level crews (according to anexpertrsquos opinion)

During D4 the rowers emphasize the stretcherpush again to accelerate themselves and hence toaccumulate kinetic energy The effectiveness of thismicrophase depends on the amount of boat velocitygained during the previous microphase D3 and on afast powerful leg drive

The microphases D5 and D6 occur after the legs aresubstantially extended and use trunk and arm workinstead The force levels and the total system accel-eration decrease but the stretcher force decreasesmore quickly and becomes negative (pull on thestretcher) This puts emphasis on the boat accelera-tion again The rowerrsquos CM acceleration becomesnegative which means transferring kinetic energyfrom the rowerrsquos mass to the boat

5 CONCLUSIONS

The phenomenon of a double peak in the boatacceleration during the drive phase can be explainedby double switching of the emphasis from pushingthe stretcher (more rowersrsquo CM acceleration) topulling the handle (more boat hull acceleration)Acceleration of the boat and of the rowersrsquo CM andthe velocity of the oar handle were used to define the

temporal structure of the stroke cycle Six micro-phases were defined during the drive phase and threemicrophases were defined during the recovery phaseThe presence of the microphase D3 (initial boatacceleration) was an important indicator of the eff-ectiveness of a rowerrsquos technique The initial boatacceleration creates a faster-moving platform for thestretcher to allow better acceleration of the rowerrsquosCM

ACKNOWLEDGEMENT

This research project was kindly supported by theAustralian Institute of Sport and Rowing Australia

Author 2010

REFERENCES

1 Bartlett R Sport biomechanics preventing injury andimproving performance 1999 pp 11ndash21 (RoutledgeNew York)

2 Toussaint H M and Beek P J Biomechanics ofcompetitive front crawl swimming Sports Med 199213 8ndash24

3 James S L and Brubaker C E Biomechanical andneuromuscular aspects of running Exercise Sport SciRev 1973 1 189ndash216

4 Dal-Monte A and Komor A Rowing and scullingmechanics In Biomechanics of sport (Ed C L Vaughan)1989 pp 53ndash119 (CRC Press Boca Raton Florida)

5 Rodriguez R J Rogriguez R P Cook S D andSandborn P M Electromyographic analysis of rowingstroke biomechanics J Sports Med Phys Fitness 199030 103ndash108

6 Shimoda M Kawakami Y and Fukunaga T Anapplication of acceleration analysis to evaluation of arowing technique In Proceedings of the 13th Interna-tional Symposium on Biomechanics in sports (EdT Bauer) Thunder Bay Ontario Canada 18ndash22 July1995 pp 95ndash100 (Lakehead University Thunder BayOntario)

7 Federation Internationale des Societes drsquoAviron (Inter-national Rowing Federation) FISA rules of racingand related bye-laws Part IV Boats and construction2005ndash2006 pp 60ndash66 available from httpwwwworldrowingcommediasdocsmedia_352459pdf

8 Geoscience Australia Applying geoscience to Australiarsquosmost important challenges 2009 available from httpwwwgeogovau

9 Kleshnev V Technology for technique improvementIn Rowing faster (Ed V Nolte) 2004 pp 209ndash228(Human Kinetics Champaign Illinois)

10 Kleshnev V Facts Did you know that Rowing Bio-mech Newslett 2007 7(72) available from httpwwwbiorowcomRBN_en_2007_files2007RowBiomNews03pdf

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 73

phase two crews of international level were selectedThe first crew consisted of Olympic Champions andwas considered by experts to be highly technicallyeffective The second crew was in the finals of theWorld Championship had very good physiologicalwork capacity but could not achieve excellent per-formance on the water Both samples were taken at astroke rate close to 32 strokesmin The first crewproduced a much quicker rise in the handle force(Fig 9(a)) achieving 70 per cent of their maximalforce at 231 per cent of the drive time The secondcrew had a relatively flatter force curve (70 per cent ofthe maximal force at 291 per cent of the drive time)but higher maximal and average forces

The faster rise in the force of the first crew coin-cides with a faster increase in the velocity Vlegs of thelegs (Fig 9(d)) (the maximal velocity of the legsoccurred at 224 per cent of the drive time) andwas accompanied by a steady increase in the trunkvelocity Vtrunk (Fig 9(f)) The second crew had aslower velocity of the legs in the first half of the drive

(a)

(b)

(c)

10

15

20

25

30

15 25 35

Rate (minndash1)

0

5

10

15

20

25

16 20 24 28 32 36 40

Sweep

Scull

Rate (minndash1)

)( e

mit evird ni 3D fo erah

S

0

5

10

15

20

15 25 35Rate (minndash1)

45

45

)( e

mit evird ni 6D f o erah

S )

( emit evird ni 1

D fo erahS

Fig 7 Dependences of the microphase time shares in thedrive time on the stroke rate (a) (b) linear trends(c) second-order polynomial trend

Fig 8 Dependences of the proportion of microphasewithin the drive phase on the stroke rate with lineartrend lines

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

70 V Kleshnev

(the maximal velocity of the legs at 322 per cent ofthe drive time) and a double-peaked trunk speedprofile Consequently the handle velocity Vh of thefirst crew increased more rapidly during the first halfof the drive (Fig 9(b))

The boat acceleration Aboat of the first crew had adeeper and later negative peak at the catch (84ms2

compared with 73ms2 in the second crew) but amuch quicker increase afterwards (Fig 9(c)) The firstpeak of Aboat was 27ms2 in the first crew but only03ms2 in the second crew which had Aboat dropbelow zero after the first peak

The first crew had a higher acceleration Arow ofthe rowerrsquos CM at the catch but the second crew

Fig 9 Biomechanical parameters and temporal structures of the drive phase for two examples of rowingtechnique

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 71

accelerated their body mass more intensively duringthe middle of the drive (Fig 9(e)) As a result thesecond crew did not have microphase D3 (initial boatacceleration) at all but it had a much longer D4(main rowersrsquo acceleration)

4 DISCUSSION AND IMPLICATIONS TEMPORALANALYSIS AND THE EFFECTIVENESS OF THEROWING TECHNIQUE

The main measure of rowing performance is theaverage speed of the rowerndashboat system Rowers can-not increase the system velocity during the recoveryphase because its deceleration is affected by suchenvironmental factors as water and air resistanceThe drive phase is the only time when rowers canincrease the system velocity (more precisely onlythat part of the drive when the propulsive force ishigher than the drag force) The higher accelera-tion of the system during the drive phase meansa higher average speed and hence better perfor-mance The system accumulates kinetic energy dur-ing the drive phase and loses it during the recoveryphase

The gain dE in kinetic energy of any body can bedefined using the equation

dE frac14 mv222

mv212

frac14 m

2ethv22 v21THORN eth13THORN

wherem is the mass of the body and v1 and v2 are thestart and end velocities respectively of the bodyrsquos CM

The mass of the rower is significantly greater thanthat of the boat (the ratio ranges from 400 per cent inlightweight women to 600 per cent in heavyweightmen) To maximize the average speed the rowersmust maximize the gain in kinetic energy which canbe achieved most effectively by means of accelerationof the heaviest part of the system (ie the rowerrsquosCM) Therefore the effectiveness of the rowing tech-

nique is related to the magnitude and timing of theacceleration of the rowerrsquos CM

The acceleration of any body depends on the forceacting on it The force Frow applied to the rowerrsquos CM(Fig 10) is equal to the sum of the foot-stretcherreaction force Frfoot which is positive ie pushesthe athletersquos CM forwards and the handle reactionforce Frhandle which is negative ie pulls the rowerbackwards according to

F row frac14 F rfoot thorn F rhandle eth14THORNBoth these reaction forces have the same magnitudeand the opposite direction to the action forces Ffootand Fh

The force Fboat applied to the boat hull is equal tothe sum of the gate force Fgate the stretcher forceFfoot and the drag force Fdrag (ignoring the small seatfrictional force) according to

Fboat frac14 Fgate thorn F foot thorn Fdrag eth15THORNTherefore emphasis on the boat accelerationrequires rowers to produce a higher gate force Fgate(and a correspondingly higher handle force Fh) andto push on the stretcher less Vice versa emphasison accelerating the rowerrsquos CM requires a higherstretcher force Ffoot and lower handle force FhGraphs of the gate and stretcher forces in Fig 4 con-firm that this switching of the emphasis happenstwice during the drive

The stretcher force Ffoot is the primary focus duringthe drive because it increases the acceleration Arow ofthe rowerrsquos CM which provides the most significantaccumulation of kinetic energy and hence increasesthe systemrsquos average speed

However the effectiveness of the rowerrsquos push onthe stretcher is affected by the boat velocity at thattime If the stretcher (boat) has a higher velocity thenthe rowers can accelerate their CM more efficientlythan if the boat is moving more slowly It can bespeculated that using the elastic energy stored in the

Frower

Fhandle

F foot

Fgate

Fbl

M

mFb

F drag

Fig 10 The main forces in the rowerndashboatndashoar system

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

72 V Kleshnev

oar handle and in the rowerrsquos tendons and musclescan play a significant role at the beginning of thedrive However this topic is beyond the scope of thispaper and can be the subject of further research Theuse of elastic energy is similar to jumping on atrampoline the point of support moves in the direc-tion of acceleration of the athletersquos CM and adds thevelocity of the support point to that of the athletersquosCM which makes the acceleration more effectiveThe presence of D3 could be called a lsquotrampoliningeffectrsquo and can be regarded as an important indicatorof the effectiveness of the rowing technique [11]In addition earlier force application could help toutilize the hydrolift effect better [12]

The practical implications of the microphases ofthe drive phase are as follows

During microphases D1 and D2 the rowers empha-size push with legs because they have to changethe direction of their movement from the sternto the bow at the catch This accelerates their bodymass and decelerates the boat At the same timethe blade obtains a grip on the water

During D3 the rowers emphasize pulling thehandle to accelerate the boat This creates a faster-moving support on the stretcher This microphase isimportant for performing an effective drive phase Itwas found that in 43 per cent of the crews thismicrophase was absent but a longer duration of D3does not mean better technique because D3 was thelongest in some low-level crews (according to anexpertrsquos opinion)

During D4 the rowers emphasize the stretcherpush again to accelerate themselves and hence toaccumulate kinetic energy The effectiveness of thismicrophase depends on the amount of boat velocitygained during the previous microphase D3 and on afast powerful leg drive

The microphases D5 and D6 occur after the legs aresubstantially extended and use trunk and arm workinstead The force levels and the total system accel-eration decrease but the stretcher force decreasesmore quickly and becomes negative (pull on thestretcher) This puts emphasis on the boat accelera-tion again The rowerrsquos CM acceleration becomesnegative which means transferring kinetic energyfrom the rowerrsquos mass to the boat

5 CONCLUSIONS

The phenomenon of a double peak in the boatacceleration during the drive phase can be explainedby double switching of the emphasis from pushingthe stretcher (more rowersrsquo CM acceleration) topulling the handle (more boat hull acceleration)Acceleration of the boat and of the rowersrsquo CM andthe velocity of the oar handle were used to define the

temporal structure of the stroke cycle Six micro-phases were defined during the drive phase and threemicrophases were defined during the recovery phaseThe presence of the microphase D3 (initial boatacceleration) was an important indicator of the eff-ectiveness of a rowerrsquos technique The initial boatacceleration creates a faster-moving platform for thestretcher to allow better acceleration of the rowerrsquosCM

ACKNOWLEDGEMENT

This research project was kindly supported by theAustralian Institute of Sport and Rowing Australia

Author 2010

REFERENCES

1 Bartlett R Sport biomechanics preventing injury andimproving performance 1999 pp 11ndash21 (RoutledgeNew York)

2 Toussaint H M and Beek P J Biomechanics ofcompetitive front crawl swimming Sports Med 199213 8ndash24

3 James S L and Brubaker C E Biomechanical andneuromuscular aspects of running Exercise Sport SciRev 1973 1 189ndash216

4 Dal-Monte A and Komor A Rowing and scullingmechanics In Biomechanics of sport (Ed C L Vaughan)1989 pp 53ndash119 (CRC Press Boca Raton Florida)

5 Rodriguez R J Rogriguez R P Cook S D andSandborn P M Electromyographic analysis of rowingstroke biomechanics J Sports Med Phys Fitness 199030 103ndash108

6 Shimoda M Kawakami Y and Fukunaga T Anapplication of acceleration analysis to evaluation of arowing technique In Proceedings of the 13th Interna-tional Symposium on Biomechanics in sports (EdT Bauer) Thunder Bay Ontario Canada 18ndash22 July1995 pp 95ndash100 (Lakehead University Thunder BayOntario)

7 Federation Internationale des Societes drsquoAviron (Inter-national Rowing Federation) FISA rules of racingand related bye-laws Part IV Boats and construction2005ndash2006 pp 60ndash66 available from httpwwwworldrowingcommediasdocsmedia_352459pdf

8 Geoscience Australia Applying geoscience to Australiarsquosmost important challenges 2009 available from httpwwwgeogovau

9 Kleshnev V Technology for technique improvementIn Rowing faster (Ed V Nolte) 2004 pp 209ndash228(Human Kinetics Champaign Illinois)

10 Kleshnev V Facts Did you know that Rowing Bio-mech Newslett 2007 7(72) available from httpwwwbiorowcomRBN_en_2007_files2007RowBiomNews03pdf

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 73

(the maximal velocity of the legs at 322 per cent ofthe drive time) and a double-peaked trunk speedprofile Consequently the handle velocity Vh of thefirst crew increased more rapidly during the first halfof the drive (Fig 9(b))

The boat acceleration Aboat of the first crew had adeeper and later negative peak at the catch (84ms2

compared with 73ms2 in the second crew) but amuch quicker increase afterwards (Fig 9(c)) The firstpeak of Aboat was 27ms2 in the first crew but only03ms2 in the second crew which had Aboat dropbelow zero after the first peak

The first crew had a higher acceleration Arow ofthe rowerrsquos CM at the catch but the second crew

Fig 9 Biomechanical parameters and temporal structures of the drive phase for two examples of rowingtechnique

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 71

accelerated their body mass more intensively duringthe middle of the drive (Fig 9(e)) As a result thesecond crew did not have microphase D3 (initial boatacceleration) at all but it had a much longer D4(main rowersrsquo acceleration)

4 DISCUSSION AND IMPLICATIONS TEMPORALANALYSIS AND THE EFFECTIVENESS OF THEROWING TECHNIQUE

The main measure of rowing performance is theaverage speed of the rowerndashboat system Rowers can-not increase the system velocity during the recoveryphase because its deceleration is affected by suchenvironmental factors as water and air resistanceThe drive phase is the only time when rowers canincrease the system velocity (more precisely onlythat part of the drive when the propulsive force ishigher than the drag force) The higher accelera-tion of the system during the drive phase meansa higher average speed and hence better perfor-mance The system accumulates kinetic energy dur-ing the drive phase and loses it during the recoveryphase

The gain dE in kinetic energy of any body can bedefined using the equation

dE frac14 mv222

mv212

frac14 m

2ethv22 v21THORN eth13THORN

wherem is the mass of the body and v1 and v2 are thestart and end velocities respectively of the bodyrsquos CM

The mass of the rower is significantly greater thanthat of the boat (the ratio ranges from 400 per cent inlightweight women to 600 per cent in heavyweightmen) To maximize the average speed the rowersmust maximize the gain in kinetic energy which canbe achieved most effectively by means of accelerationof the heaviest part of the system (ie the rowerrsquosCM) Therefore the effectiveness of the rowing tech-

nique is related to the magnitude and timing of theacceleration of the rowerrsquos CM

The acceleration of any body depends on the forceacting on it The force Frow applied to the rowerrsquos CM(Fig 10) is equal to the sum of the foot-stretcherreaction force Frfoot which is positive ie pushesthe athletersquos CM forwards and the handle reactionforce Frhandle which is negative ie pulls the rowerbackwards according to

F row frac14 F rfoot thorn F rhandle eth14THORNBoth these reaction forces have the same magnitudeand the opposite direction to the action forces Ffootand Fh

The force Fboat applied to the boat hull is equal tothe sum of the gate force Fgate the stretcher forceFfoot and the drag force Fdrag (ignoring the small seatfrictional force) according to

Fboat frac14 Fgate thorn F foot thorn Fdrag eth15THORNTherefore emphasis on the boat accelerationrequires rowers to produce a higher gate force Fgate(and a correspondingly higher handle force Fh) andto push on the stretcher less Vice versa emphasison accelerating the rowerrsquos CM requires a higherstretcher force Ffoot and lower handle force FhGraphs of the gate and stretcher forces in Fig 4 con-firm that this switching of the emphasis happenstwice during the drive

The stretcher force Ffoot is the primary focus duringthe drive because it increases the acceleration Arow ofthe rowerrsquos CM which provides the most significantaccumulation of kinetic energy and hence increasesthe systemrsquos average speed

However the effectiveness of the rowerrsquos push onthe stretcher is affected by the boat velocity at thattime If the stretcher (boat) has a higher velocity thenthe rowers can accelerate their CM more efficientlythan if the boat is moving more slowly It can bespeculated that using the elastic energy stored in the

Frower

Fhandle

F foot

Fgate

Fbl

M

mFb

F drag

Fig 10 The main forces in the rowerndashboatndashoar system

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

72 V Kleshnev

oar handle and in the rowerrsquos tendons and musclescan play a significant role at the beginning of thedrive However this topic is beyond the scope of thispaper and can be the subject of further research Theuse of elastic energy is similar to jumping on atrampoline the point of support moves in the direc-tion of acceleration of the athletersquos CM and adds thevelocity of the support point to that of the athletersquosCM which makes the acceleration more effectiveThe presence of D3 could be called a lsquotrampoliningeffectrsquo and can be regarded as an important indicatorof the effectiveness of the rowing technique [11]In addition earlier force application could help toutilize the hydrolift effect better [12]

The practical implications of the microphases ofthe drive phase are as follows

During microphases D1 and D2 the rowers empha-size push with legs because they have to changethe direction of their movement from the sternto the bow at the catch This accelerates their bodymass and decelerates the boat At the same timethe blade obtains a grip on the water

During D3 the rowers emphasize pulling thehandle to accelerate the boat This creates a faster-moving support on the stretcher This microphase isimportant for performing an effective drive phase Itwas found that in 43 per cent of the crews thismicrophase was absent but a longer duration of D3does not mean better technique because D3 was thelongest in some low-level crews (according to anexpertrsquos opinion)

During D4 the rowers emphasize the stretcherpush again to accelerate themselves and hence toaccumulate kinetic energy The effectiveness of thismicrophase depends on the amount of boat velocitygained during the previous microphase D3 and on afast powerful leg drive

The microphases D5 and D6 occur after the legs aresubstantially extended and use trunk and arm workinstead The force levels and the total system accel-eration decrease but the stretcher force decreasesmore quickly and becomes negative (pull on thestretcher) This puts emphasis on the boat accelera-tion again The rowerrsquos CM acceleration becomesnegative which means transferring kinetic energyfrom the rowerrsquos mass to the boat

5 CONCLUSIONS

The phenomenon of a double peak in the boatacceleration during the drive phase can be explainedby double switching of the emphasis from pushingthe stretcher (more rowersrsquo CM acceleration) topulling the handle (more boat hull acceleration)Acceleration of the boat and of the rowersrsquo CM andthe velocity of the oar handle were used to define the

temporal structure of the stroke cycle Six micro-phases were defined during the drive phase and threemicrophases were defined during the recovery phaseThe presence of the microphase D3 (initial boatacceleration) was an important indicator of the eff-ectiveness of a rowerrsquos technique The initial boatacceleration creates a faster-moving platform for thestretcher to allow better acceleration of the rowerrsquosCM

ACKNOWLEDGEMENT

This research project was kindly supported by theAustralian Institute of Sport and Rowing Australia

Author 2010

REFERENCES

1 Bartlett R Sport biomechanics preventing injury andimproving performance 1999 pp 11ndash21 (RoutledgeNew York)

2 Toussaint H M and Beek P J Biomechanics ofcompetitive front crawl swimming Sports Med 199213 8ndash24

3 James S L and Brubaker C E Biomechanical andneuromuscular aspects of running Exercise Sport SciRev 1973 1 189ndash216

4 Dal-Monte A and Komor A Rowing and scullingmechanics In Biomechanics of sport (Ed C L Vaughan)1989 pp 53ndash119 (CRC Press Boca Raton Florida)

5 Rodriguez R J Rogriguez R P Cook S D andSandborn P M Electromyographic analysis of rowingstroke biomechanics J Sports Med Phys Fitness 199030 103ndash108

6 Shimoda M Kawakami Y and Fukunaga T Anapplication of acceleration analysis to evaluation of arowing technique In Proceedings of the 13th Interna-tional Symposium on Biomechanics in sports (EdT Bauer) Thunder Bay Ontario Canada 18ndash22 July1995 pp 95ndash100 (Lakehead University Thunder BayOntario)

7 Federation Internationale des Societes drsquoAviron (Inter-national Rowing Federation) FISA rules of racingand related bye-laws Part IV Boats and construction2005ndash2006 pp 60ndash66 available from httpwwwworldrowingcommediasdocsmedia_352459pdf

8 Geoscience Australia Applying geoscience to Australiarsquosmost important challenges 2009 available from httpwwwgeogovau

9 Kleshnev V Technology for technique improvementIn Rowing faster (Ed V Nolte) 2004 pp 209ndash228(Human Kinetics Champaign Illinois)

10 Kleshnev V Facts Did you know that Rowing Bio-mech Newslett 2007 7(72) available from httpwwwbiorowcomRBN_en_2007_files2007RowBiomNews03pdf

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 73

accelerated their body mass more intensively duringthe middle of the drive (Fig 9(e)) As a result thesecond crew did not have microphase D3 (initial boatacceleration) at all but it had a much longer D4(main rowersrsquo acceleration)

4 DISCUSSION AND IMPLICATIONS TEMPORALANALYSIS AND THE EFFECTIVENESS OF THEROWING TECHNIQUE

The main measure of rowing performance is theaverage speed of the rowerndashboat system Rowers can-not increase the system velocity during the recoveryphase because its deceleration is affected by suchenvironmental factors as water and air resistanceThe drive phase is the only time when rowers canincrease the system velocity (more precisely onlythat part of the drive when the propulsive force ishigher than the drag force) The higher accelera-tion of the system during the drive phase meansa higher average speed and hence better perfor-mance The system accumulates kinetic energy dur-ing the drive phase and loses it during the recoveryphase

The gain dE in kinetic energy of any body can bedefined using the equation

dE frac14 mv222

mv212

frac14 m

2ethv22 v21THORN eth13THORN

wherem is the mass of the body and v1 and v2 are thestart and end velocities respectively of the bodyrsquos CM

The mass of the rower is significantly greater thanthat of the boat (the ratio ranges from 400 per cent inlightweight women to 600 per cent in heavyweightmen) To maximize the average speed the rowersmust maximize the gain in kinetic energy which canbe achieved most effectively by means of accelerationof the heaviest part of the system (ie the rowerrsquosCM) Therefore the effectiveness of the rowing tech-

nique is related to the magnitude and timing of theacceleration of the rowerrsquos CM

The acceleration of any body depends on the forceacting on it The force Frow applied to the rowerrsquos CM(Fig 10) is equal to the sum of the foot-stretcherreaction force Frfoot which is positive ie pushesthe athletersquos CM forwards and the handle reactionforce Frhandle which is negative ie pulls the rowerbackwards according to

F row frac14 F rfoot thorn F rhandle eth14THORNBoth these reaction forces have the same magnitudeand the opposite direction to the action forces Ffootand Fh

The force Fboat applied to the boat hull is equal tothe sum of the gate force Fgate the stretcher forceFfoot and the drag force Fdrag (ignoring the small seatfrictional force) according to

Fboat frac14 Fgate thorn F foot thorn Fdrag eth15THORNTherefore emphasis on the boat accelerationrequires rowers to produce a higher gate force Fgate(and a correspondingly higher handle force Fh) andto push on the stretcher less Vice versa emphasison accelerating the rowerrsquos CM requires a higherstretcher force Ffoot and lower handle force FhGraphs of the gate and stretcher forces in Fig 4 con-firm that this switching of the emphasis happenstwice during the drive

The stretcher force Ffoot is the primary focus duringthe drive because it increases the acceleration Arow ofthe rowerrsquos CM which provides the most significantaccumulation of kinetic energy and hence increasesthe systemrsquos average speed

However the effectiveness of the rowerrsquos push onthe stretcher is affected by the boat velocity at thattime If the stretcher (boat) has a higher velocity thenthe rowers can accelerate their CM more efficientlythan if the boat is moving more slowly It can bespeculated that using the elastic energy stored in the

Frower

Fhandle

F foot

Fgate

Fbl

M

mFb

F drag

Fig 10 The main forces in the rowerndashboatndashoar system

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

72 V Kleshnev

oar handle and in the rowerrsquos tendons and musclescan play a significant role at the beginning of thedrive However this topic is beyond the scope of thispaper and can be the subject of further research Theuse of elastic energy is similar to jumping on atrampoline the point of support moves in the direc-tion of acceleration of the athletersquos CM and adds thevelocity of the support point to that of the athletersquosCM which makes the acceleration more effectiveThe presence of D3 could be called a lsquotrampoliningeffectrsquo and can be regarded as an important indicatorof the effectiveness of the rowing technique [11]In addition earlier force application could help toutilize the hydrolift effect better [12]

The practical implications of the microphases ofthe drive phase are as follows

During microphases D1 and D2 the rowers empha-size push with legs because they have to changethe direction of their movement from the sternto the bow at the catch This accelerates their bodymass and decelerates the boat At the same timethe blade obtains a grip on the water

During D3 the rowers emphasize pulling thehandle to accelerate the boat This creates a faster-moving support on the stretcher This microphase isimportant for performing an effective drive phase Itwas found that in 43 per cent of the crews thismicrophase was absent but a longer duration of D3does not mean better technique because D3 was thelongest in some low-level crews (according to anexpertrsquos opinion)

During D4 the rowers emphasize the stretcherpush again to accelerate themselves and hence toaccumulate kinetic energy The effectiveness of thismicrophase depends on the amount of boat velocitygained during the previous microphase D3 and on afast powerful leg drive

The microphases D5 and D6 occur after the legs aresubstantially extended and use trunk and arm workinstead The force levels and the total system accel-eration decrease but the stretcher force decreasesmore quickly and becomes negative (pull on thestretcher) This puts emphasis on the boat accelera-tion again The rowerrsquos CM acceleration becomesnegative which means transferring kinetic energyfrom the rowerrsquos mass to the boat

5 CONCLUSIONS

The phenomenon of a double peak in the boatacceleration during the drive phase can be explainedby double switching of the emphasis from pushingthe stretcher (more rowersrsquo CM acceleration) topulling the handle (more boat hull acceleration)Acceleration of the boat and of the rowersrsquo CM andthe velocity of the oar handle were used to define the

temporal structure of the stroke cycle Six micro-phases were defined during the drive phase and threemicrophases were defined during the recovery phaseThe presence of the microphase D3 (initial boatacceleration) was an important indicator of the eff-ectiveness of a rowerrsquos technique The initial boatacceleration creates a faster-moving platform for thestretcher to allow better acceleration of the rowerrsquosCM

ACKNOWLEDGEMENT

This research project was kindly supported by theAustralian Institute of Sport and Rowing Australia

Author 2010

REFERENCES

1 Bartlett R Sport biomechanics preventing injury andimproving performance 1999 pp 11ndash21 (RoutledgeNew York)

2 Toussaint H M and Beek P J Biomechanics ofcompetitive front crawl swimming Sports Med 199213 8ndash24

3 James S L and Brubaker C E Biomechanical andneuromuscular aspects of running Exercise Sport SciRev 1973 1 189ndash216

4 Dal-Monte A and Komor A Rowing and scullingmechanics In Biomechanics of sport (Ed C L Vaughan)1989 pp 53ndash119 (CRC Press Boca Raton Florida)

5 Rodriguez R J Rogriguez R P Cook S D andSandborn P M Electromyographic analysis of rowingstroke biomechanics J Sports Med Phys Fitness 199030 103ndash108

6 Shimoda M Kawakami Y and Fukunaga T Anapplication of acceleration analysis to evaluation of arowing technique In Proceedings of the 13th Interna-tional Symposium on Biomechanics in sports (EdT Bauer) Thunder Bay Ontario Canada 18ndash22 July1995 pp 95ndash100 (Lakehead University Thunder BayOntario)

7 Federation Internationale des Societes drsquoAviron (Inter-national Rowing Federation) FISA rules of racingand related bye-laws Part IV Boats and construction2005ndash2006 pp 60ndash66 available from httpwwwworldrowingcommediasdocsmedia_352459pdf

8 Geoscience Australia Applying geoscience to Australiarsquosmost important challenges 2009 available from httpwwwgeogovau

9 Kleshnev V Technology for technique improvementIn Rowing faster (Ed V Nolte) 2004 pp 209ndash228(Human Kinetics Champaign Illinois)

10 Kleshnev V Facts Did you know that Rowing Bio-mech Newslett 2007 7(72) available from httpwwwbiorowcomRBN_en_2007_files2007RowBiomNews03pdf

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 73

oar handle and in the rowerrsquos tendons and musclescan play a significant role at the beginning of thedrive However this topic is beyond the scope of thispaper and can be the subject of further research Theuse of elastic energy is similar to jumping on atrampoline the point of support moves in the direc-tion of acceleration of the athletersquos CM and adds thevelocity of the support point to that of the athletersquosCM which makes the acceleration more effectiveThe presence of D3 could be called a lsquotrampoliningeffectrsquo and can be regarded as an important indicatorof the effectiveness of the rowing technique [11]In addition earlier force application could help toutilize the hydrolift effect better [12]

The practical implications of the microphases ofthe drive phase are as follows

During microphases D1 and D2 the rowers empha-size push with legs because they have to changethe direction of their movement from the sternto the bow at the catch This accelerates their bodymass and decelerates the boat At the same timethe blade obtains a grip on the water

During D3 the rowers emphasize pulling thehandle to accelerate the boat This creates a faster-moving support on the stretcher This microphase isimportant for performing an effective drive phase Itwas found that in 43 per cent of the crews thismicrophase was absent but a longer duration of D3does not mean better technique because D3 was thelongest in some low-level crews (according to anexpertrsquos opinion)

During D4 the rowers emphasize the stretcherpush again to accelerate themselves and hence toaccumulate kinetic energy The effectiveness of thismicrophase depends on the amount of boat velocitygained during the previous microphase D3 and on afast powerful leg drive

The microphases D5 and D6 occur after the legs aresubstantially extended and use trunk and arm workinstead The force levels and the total system accel-eration decrease but the stretcher force decreasesmore quickly and becomes negative (pull on thestretcher) This puts emphasis on the boat accelera-tion again The rowerrsquos CM acceleration becomesnegative which means transferring kinetic energyfrom the rowerrsquos mass to the boat

5 CONCLUSIONS

The phenomenon of a double peak in the boatacceleration during the drive phase can be explainedby double switching of the emphasis from pushingthe stretcher (more rowersrsquo CM acceleration) topulling the handle (more boat hull acceleration)Acceleration of the boat and of the rowersrsquo CM andthe velocity of the oar handle were used to define the

temporal structure of the stroke cycle Six micro-phases were defined during the drive phase and threemicrophases were defined during the recovery phaseThe presence of the microphase D3 (initial boatacceleration) was an important indicator of the eff-ectiveness of a rowerrsquos technique The initial boatacceleration creates a faster-moving platform for thestretcher to allow better acceleration of the rowerrsquosCM

ACKNOWLEDGEMENT

This research project was kindly supported by theAustralian Institute of Sport and Rowing Australia

Author 2010

REFERENCES

1 Bartlett R Sport biomechanics preventing injury andimproving performance 1999 pp 11ndash21 (RoutledgeNew York)

2 Toussaint H M and Beek P J Biomechanics ofcompetitive front crawl swimming Sports Med 199213 8ndash24

3 James S L and Brubaker C E Biomechanical andneuromuscular aspects of running Exercise Sport SciRev 1973 1 189ndash216

4 Dal-Monte A and Komor A Rowing and scullingmechanics In Biomechanics of sport (Ed C L Vaughan)1989 pp 53ndash119 (CRC Press Boca Raton Florida)

5 Rodriguez R J Rogriguez R P Cook S D andSandborn P M Electromyographic analysis of rowingstroke biomechanics J Sports Med Phys Fitness 199030 103ndash108

6 Shimoda M Kawakami Y and Fukunaga T Anapplication of acceleration analysis to evaluation of arowing technique In Proceedings of the 13th Interna-tional Symposium on Biomechanics in sports (EdT Bauer) Thunder Bay Ontario Canada 18ndash22 July1995 pp 95ndash100 (Lakehead University Thunder BayOntario)

7 Federation Internationale des Societes drsquoAviron (Inter-national Rowing Federation) FISA rules of racingand related bye-laws Part IV Boats and construction2005ndash2006 pp 60ndash66 available from httpwwwworldrowingcommediasdocsmedia_352459pdf

8 Geoscience Australia Applying geoscience to Australiarsquosmost important challenges 2009 available from httpwwwgeogovau

9 Kleshnev V Technology for technique improvementIn Rowing faster (Ed V Nolte) 2004 pp 209ndash228(Human Kinetics Champaign Illinois)

10 Kleshnev V Facts Did you know that Rowing Bio-mech Newslett 2007 7(72) available from httpwwwbiorowcomRBN_en_2007_files2007RowBiomNews03pdf

JSET40 Proc IMechE Vol 224 Part P J Sports Engineering and Technology

Boat acceleration temporal structure of the stroke cycle and effectiveness in rowing 73

11 Kleshnev V Ideas Rowing Biomech Newslett 20066(2) available from httpwwwbiorowcomRBN_en_2006_filesRowBiomNews03pdf

12 Pulman C The physics of rowing 2008 available fromhttpwwwatmoxacukrowingphysicsrowingpdf

APPENDIX

Notation

Ah horizontal oar angleAv vertical oar angleAboat boat accelerationArow acceleration of the centre of mass of the

rowerAsys acceleration of the centre of mass of the

rowerndashboat systemCM centre of massdE gain in kinetic energy of any bodyFbl blade forceFboat force acting on the boatFdrag drag forceFfoot foot-stretcher forceFgate gate forceFh normal component of the force applied to

the handle of the oarFrfoot foot-stretcher reaction forceFrhandle handle reaction force

Frow force acting on the centre of mass of therower

Fsys force acting on the centre of mass of therowerndashboat system

Kdrag drag factorLh length of the hand gripLin measured inboard lengthLin-a actual inboard lengthLoar overall oar lengthLout-a actual outboard lengthLs seat positionLsp length of the blade spoonLt position of the top of the trunkm mass of the bodymboat mass of the boatmrow mass of the rowermsys mass of the rowerndashboat systemv1 start velocityv2 end velocityVarms velocity of the armsVboat velocity of the boatVd deviation of the instantaneous velocity from

the average velocityVlegs velocity of the legsVtrunk velocity of the trunkVtrunkb velocity of the trunk relative to the boatWg width of the gate

Proc IMechE Vol 224 Part P J Sports Engineering and Technology JSET40

74 V Kleshnev