c1 c2 c3 c4 c5 c6

6
RIVET I 2012 Data Collections Coastal Imaging Lab Oregon State University April 27 to May 20, 2012 CIL RIVET Collections RIVET required installation of a new Argus station in late April on the top of a 106’ high jackup tower installed on the SW side of NRI. This document describes the Argus collection system and the data collected as well as how and where to access the data. Argus Cameras The Argus cameras are located in the jackup tower at roughly xyz = [169, 596, 34] in FRF coordinates. Six cameras allowed a roughly 200° field of view from alongshore to the SW to slightly into the inlet to the NE. Numbers and orientations of cameras are shown in Figure 1 below on a background image that is a merged, rectified time exposure image of the region in the RIVET coordinate system. Images can either be used individually or as merged products such as Figure 1. Figure 1. Merged snapshot showing the fields of view of the six primary cameras. Cameras geometry control was established based on a set of ground control points (GCPs) whose locations were surveyed. Geometries were established at the beginning of the experiment and were checked qualitatively, especially to determine if the aim points changed when the tower was lowered then raised again in C1 C2 C3 C4 C5 C6

Upload: trantram

Post on 10-Feb-2017

298 views

Category:

Documents


5 download

TRANSCRIPT

Page 1: C1 C2 C3 C4 C5 C6

RIVET  I  2012  Data  Collections  Coastal  Imaging  Lab  

Oregon  State  University  April  27  to  May  20,  2012  

 CIL  RIVET  Collections  RIVET  required  installation  of  a  new  Argus  station  in  late  April  on  the  top  of  a  106’  high  jack-­‐up  tower  installed  on  the  SW  side  of  NRI.    This  document  describes  the  Argus  collection  system  and  the  data  collected  as  well  as  how  and  where  to  access  the  data.        Argus  Cameras  The  Argus  cameras  are  located  in  the  jack-­‐up  tower  at  roughly  xyz  =  [-­‐169,  -­‐596,  34]  in  FRF  coordinates.    Six  cameras  allowed  a  roughly  200°  field  of  view  from  alongshore  to  the  SW  to  slightly  into  the  inlet  to  the  NE.    Numbers  and  orientations  of  cameras  are  shown  in  Figure  1  below  on  a  background  image  that  is  a  merged,  rectified  time  exposure  image  of  the  region  in  the  RIVET  coordinate  system.    Images  can  either  be  used  individually  or  as  merged  products  such  as  Figure  1.    

 Figure  1.    Merged  snapshot  showing  the  fields  of  view  of  the  six  primary  cameras.        Cameras  geometry  control  was  established  based  on  a  set  of  ground  control  points  (GCPs)  whose  locations  were  surveyed.    Geometries  were  established  at  the  beginning  of  the  experiment  and  were  checked  qualitatively,  especially  to  determine  if  the  aim  points  changed  when  the  tower  was  lowered  then  raised  again  in  

C1

C2 C3

C4

C5

C6

Page 2: C1 C2 C3 C4 C5 C6

response  to  impending  storms  or  for  camera  maintenance.    Camera  resolution  varies  with  distance  from  the  camera  as  show  in  Figure  2.    Time  is  synched  to  an  NTP  server  continuously.      

   Figure  2.    Local  resolution  maps  in  x  and  y.    Data  Locations  and  Naming  Conventions  Data  are  stored  on  the  CIL  ftp  site,  cil-­‐ftp.coas.oregonstate.edu  and  can  be  downloaded  using  normal  anonymous  ftp  (with  a  valid  RFC822  format  email  address  as  the  password).    Data  are  sorted  by  site/year/camera/day/file.    For  example,  images  for  C3  of  April  29,  2012  would  be  located  in  /ftp/pub/rosie/2012/c3/120_Apr.29.        Argus  filenames  using  a  long  (and  useful)  naming  convention.    For  example:  

1335740400.Sun.Apr.29_23_00_00.GMT.2012.rosie.c3.timex.jpg    is  a  time  exposure  image  for  C3  collected  on  April  29,  2012  at  23:00:00  GMT.    All  Argus  data  are  saved  in  GMT  (because  we  have  stations  in  many  time  zones).    Note  that  images  also  have  timestamp  and  other  information  imprinted  on  the  top  and  bottom  borders.    These  are  written  in  local  standard  time.    Be  careful  to  not  confuse  these  (image  stamp  in  EST,  filename  in  GMT  =  EST+5).    Also  note  that  we  never  use  daylight  time,  so  you  need  to  correct  from  GMT  to  daylight  by  subtracting  4  hours.    Our  primary  time  reference  is  epoch  time,  a  computer  standard  that  is  the  number  of  seconds  since  January  1,  1970  (GMT).    This  is  the  10-­‐digit  leading  number  in  the  

Page 3: C1 C2 C3 C4 C5 C6

filename.    Routines  to  convert  between  epoch  to  matlab’s  datenum  are  located  in  /ftp/pub/Experiments/SZO2010/commonMatlabCode.        Five  types  of  images  are  collected:  

• Snapshot    • Timex  –  average  of  2Hz  frames  collected  over  ten-­‐minute  period  • Var     –  standard  deviation  of  same  image  sample    • Bright    –  brightest  that  each  pixel  gets  over  the  same  period  • Dark   -­‐  darkest  that  each  pixel  gets  over  the  same  period  

In  addition,  daytimex  images  average  all  of  the  timex  images  for  a  day.    Bright  images  can  be  useful  for  showing  the  map  of  where  any  wave  broke  during  the  10  minute  collect.    Merged  images  like  Figure  1,  are  automatically  processed  and  are  stored  as  camera  cx  using  the  same  naming  conventions  but  with  an  additional  suffix  ‘merge’.    For  example,       1335740402.Sun.Apr.29_23_00_02.GMT.2012.rosie.c3.bright.merge.png    is  a  merged  brightest  image  from  April  29,  2300  GMT  (1800  EST  or  1900  EDT).    These  are  png  images  and  are  saved  in  two  sizes:  merge  images  that  span  from  x  =  -­‐500:  1000,  and  y  =  -­‐1300:  1000,  so  cover  the  entire  inlet  but  at  fuzzy  resolution  at  large  distances,  and  mergeSmall  images  that  span  from  x  =  -­‐500:500  and  y  =  -­‐1200:+100  at  much  better  resolution.    smallMerge  images  cover  the  south  ebb  tidal  delta,  western  beach  and  part  of  the  inlet.    Note  that  the  merged  image  data  are  also  stored  as  .mat  files  so  you  can  create  your  own  figures.    Pixel  Time  Series  While  image  data  synthesize  data  from  a  10-­‐minute  collect  into  single  images,  pixel  time  series  data  sample  and  save  the  2Hz  time  variability  for  2048  samples  (17+  minutes),  but  only  for  specific  requested  pixels.    Data  are  generally  called  time  stacks  and  files  are  saved  under  each  camera  as  raster  files  with  the  suffix  ‘.stack.ras’  (usually  gzip  compressed,  so  with  an  additional  .gz  suffix).    Retrieval  of  data  is  somewhat  complicated,  but  has  been  automated  for  RIVET  I  and  results  are  stored  under  cx.    For  example,  the  file       1337471940.Sat,May.19_23_59:00.GMT.2012.rosie.cx.vbar2.mat    is  a  matlab  data  file  from  May  19  that  contains  the  pixel  instrument  vbar2  (a  close-­‐packed  array  of  pixels  from  the  middle  of  the  inlet  designed  to  allow  estimation  of  currents  –  originally  designed  for  longshore  currents,  hence  the  instrument  name).        All  pixel  instrument  mat  files  contain  the  following  variables  

• T     -­‐  epoch  time  for  each  of  the  N  samples  in  the  time  series  • XYZ     -­‐  RIVET  coordinates  for  each  of  the  M  pixels  (z  usually  =0)  • RAW     -­‐  N  by  M  time  stack  of  raw  camera  intensities  • CAM     -­‐  camera  number  associated  with  each  pixel  (length  M)  • GAIN     -­‐  N  gain  values  (for  each  frame)  (usually  constant  for  run)  • SHUTTER   -­‐  N  shutter  exposure  times  

Page 4: C1 C2 C3 C4 C5 C6

• CORRECTED   -­‐  image  brightness  corrected  for  gain  and  shutter.  Corrected  intensities  are  RAW  divided  by  10^(gain/20)  *  shutter,  reversing  the  camera  compensation  for  changing  lighting  conditions  during  the  day.    Pixel  instrument  collections  during  RIVET  were  of  two  kinds,  cBathy  arrays  and  vBar  vector  collections.    Sampling  evolved  through  the  experiment  as  we  tried  different  concepts,  so  some  instruments  were  only  collected  during  the  latter  part  of  sampling.    cBathy  arrays  use  the  instrument  name  mBWx,  where  x  =  1:4  (mBW  is  a  historical  name  indicating  a  matrix  collect  (the  ‘m’)  of  type  Beach  Wizard  (BW)).    Figure  3  shows  the  cBathy  pixel  array  with  dot  symbols,  hard  to  distinguish  because  they  are  so  dense.    The  blue  mBW1  pixels  closest  to  the  camera  (the  white  partial  circle  at  the  top  of  the  figure)  have  a  spacing  of  2.5  x  2.5  m,  commensurate  with  the  better  pixel  resolution  at  those  short  ranges,  whereas  the  more  distant  pixels  (mBW2-­‐  red,  mBW3-­‐green,  mBW4-­‐black)  have  a  5  x  5  pixels  spacing.    

 Figure  3.    Pixel  arrays  for  the  cBathy  array  showing  mBW1  (blue  dots,  closest  to  the  cameras  at  the  center  top),  mBW2  (red,  toward  the  inlet  channel),  mBW  3  (green,  the  offshore  pixels,  and  mBW4  (black  to  the  SW  covering  the  part  of  the  open  beach).      

Page 5: C1 C2 C3 C4 C5 C6

There  are  eventually  four  vbar  arrays,  one  along  the  channel  immediately  offshore  of  the  cameras  (the  site  of  the  triPole  and  Jim  Thomson’s  met  and  in-­‐situ  fixed  instruments)  and  three  parallel  transects  down  the  throat  of  the  inlet  (Figure  4).    

 Figure  4.    Merged  snapshot  image  (camera  boundaries  apparent  due  to  gain  differences  between  cameras)  and  the  analysis  locations  of  vbar  pixel  time  stacks  along  the  main  inlet  channel.    The  red,  green  and  blue  transects  correspond  to  instrument  names  vbar  2,  3  and  4,  respectively  while  the  experimental  instrument  in  front  of  the  cameras  is  vbar1.    Note  that  the  vbar  pixel  instruments  have  a  dense  spacing  along  transect  (as  close  as  is  possible)  while  the  cBathy  array  is  more  sparsely  spaced  but  is  usually  dense  enough  for  most  purposes.    Time  series  data  from  either  array  can  be  used  as  optical  instruments  to  compare  to  any  other  data  source.    Both  cBathy  and  vbar  arrays  were  collected  for  research  purposes  to  investigate  new  methods  (under  non-­‐traditional  conditions).    Thus,  production  analyses  are  not  current  available  for  these  instruments.    

Page 6: C1 C2 C3 C4 C5 C6

Collection  Schedules  Argus  data  were  collected  every  half  hour  during  RIVET  I.    Collection  start  times  are  staggered  slightly  due  to  a  constraint  in  our  archiving  protocols.    For  each  daylight  hour,  collections  were  as  follows:  

Minute  after  hour   Collection  00,  30   image  products  (snap,  timex,  etc)  59,  29   pixel  time  stack  data,  mBW  and  vbar  

The  collection  clock  is  synched  continuously  with  an  NTP  time  server.    All  collections  should  start  at  the  top  of  the  minute,  but  make  have  order  ½  second  delays.    However,  epoch  times  listed  in  files  will  always  be  correct.    Matlab  Files  Several  useful  matlab  files  are  located  on  /ftp/pub/Experiments/SZO2010/commonMatlabCode.    These  include  epoch  to  matlab  conversion  and  a  short  example  routine  to  load  and  use  an  instrument.    Problems:  Contact  Rob  Holman,  [email protected].