dust storm approaching stratford, texas dust bowl surveying in texas

Dust storm approaching Stratford, Texas Dust bowl surveying in Texas Image ID: theb1365, Historic C&GS Collection

Location: Stratford, Texas Photo Date: April 18, 1935

Credit: NOAA George E. Marsh Album

Drought

Soil moisture percentile (-)

SW Monitor: Regional to global

VIC Soil Moisture Percentiles (wrt 1960-2008)20131201

Challenge: Meteorological data

http://www.ncdc.noaa.gov/; Global Summary of the Day

Three requirements:- long term data record (climatology)- near real-time reporting- consistency

http://trmm.gsfc.nasa.gov/

TRMM instruments:

• Precipitation Radar (PR)

• TRMM Microwave Imager (TMI)

• Visible and InfraRed Scanner (VIRS)

• Cloud and Earth Radiant Energy Sensor (CERES)

• Lightning Imaging Sensor (LIS)

http://trmm.gsfc.nasa.gov/

http://www.emc.ncep.noaa.gov/

Forcing data

For the purpose of this presentation, current is 12/01/2012. However, the RT and GFS4 datasets can be updated daily.

Precipitation data sets used:Sheffield et al: 1948-2008 (T and P) (SH)TRMM V7:

Research product RP: 1/1/1998-6/30/2012Real-time product RT: 3/1/2000-current

available in near real-time

Temperature data sets used:Sheffield et al: 1948-2008NCEP/NCAR Reanalysis: 1979-2011GFS 4: 2009-current

available in near real-time

Station-based: climatology

Station-based: climatology

Satellite + stations: spinup

Satellite only: real-time

Model + stations: spinup

Model + stations: real-time

The challenge• The climatological period, used to determine the soil moisture

percentiles for each grid cell for a given date, is based on the period 1960-2008 (Sheffield data set)

• Because soil moisture conditions in the realtime simulations are compared with those from the climatological period, systematic differences between the forcing datasets must be avoided

• Correct the precipitation and temperature data sets used for the post-2008 period so that they match the Sheffield dataset (in a statistical sense) for any overlapping period

Forcing data• Wind speed values are climatological values based on a

reanalysis• All other meteorological forcings needed to drive the models

are estimated from the daily temperature range and precipitation– downward shortwave radiation– downwelling longwave radiation– humidity

Precipitation data available: • Sheffield (1/1/1948-12/31/2008) (SH)• TMPAV7 3B42 research product (1/1/1998-6/30/2012) (RP)• TMPAV7 3B42RT (3/1/2000-current) (RT)

Goal: Bias correct RP or RT to SH

Precipitation Adjustment

Corrections:1. Match number of rainy days from RT and RP to Sheffield2. Match monthly mean (ratio of long-term mean)

Annual RP precipitation

Annual corrected RP precipitation

TRMM V7 RP Correction

Annual RT precipitation

Annual corrected RT precipitation

TRMM V7 RT Correction

SH/RP ratio (annual)

SH/RT ratio (annual)

Ratios between RT and SH are much larger (than RP and SH), because the RT data does not include station observations

Temperature Data Adjustment• Data set used:

- Sheffield et al: 1948 - 2008- NCEP/NCAR Reanalysis: 1979 - 2011- GFS4: 2009 - Current

• Individual Bias Correction on daily Tmax and Tmin to ensure temperature range is preserved

• For period 2009/01 – 2012/06:- Interpolate Reanalysis data to 0.5 degrees- Calculate mean monthly Tmax and Tmin offsets from Reanalysis to Sheffield

data set- Impose calculated offsets for each month on interpolated Reanalysis for this

period • For RT (post 2012/06):

- Interpolate GFS4 analysis data to 0.5 degrees- Calculate mean monthly Tmax and Tmin offsets from GFS4 to Reanalysis

data set- Impose calculated offsets for each month on interpolated GFS4 for this

period

Temperature Adjustment Results• GFS4 to Reanalysis

- Before Adjustment- January Tmax and Tmin

GFS4 Reanalysis

Adjusted GFS4 Data

Raw Reanalysis Data

Adjusted GFS4 Data

Raw Reanalysis Data

Tmax

Tmin

Historical Drought Events

1. 1952 drought in the U.S.

2. 1965 drought in Australia

3. 1983 drought in Africa

1952 drought in the U.S. (GSWM upper vs SWM lower)

VIC Noah 2.8

1965 drought in Australia

VIC Noah 2.8

1983 drought in Africa

VIC Noah 2.8

Recent Drought: multimodel 2012/07/01

Note that the percentile method is very sensitive to small changes in moisture in dry regions such as the Sahara. The high percentiles (“wetness”) in that region is not realistic.

In the realtime version of the global surface water monitor the desert areas of the world will be grayed out and set to “no data”

VIC

Noah

SAC

Recent Drought: individual models 2012/07/01

2012 Monthly Drought Information

2012/01/01 - Multimodel

2012/02/01 - Multimodel

2012 Monthly Drought Information

2012/03/01 - Multimodel

2012 Monthly Drought Information

2012/04/01 - Multimodel

2012 Monthly Drought Information

2012/05/01 - Multimodel

2012 Monthly Drought Information

2012/06/01 - Multimodel

2012 Monthly Drought Information

2012/07/01 - Multimodel

2012 Monthly Drought Information

2012/08/01 - Multimodel

2012 Monthly Drought Information

2012/09/01 - Multimodel

2012 Monthly Drought Information

2012/10/01 - Multimodel

2012 Monthly Drought Information

2012/11/01 - Multimodel

2012 Monthly Drought Information

2012/12/01 - Multimodel

2012 Monthly Drought Information

Comparison with other Drought Monitors2012/07/01

http://ncc.cma.gov.cn/influ/en_hljc.phpUW Surface Water Monitor Chinese Drought and Flood Monitor

Comparison with other Drought Monitors2012/10/01

UW Surface Water Monitor http://ncc.cma.gov.cn/influ/en_hljc.phpChinese Drought and Flood Monitor

Comparison with other Drought Monitors2012/12/01

UW Surface Water Monitor http://ncc.cma.gov.cn/influ/en_hljc.phpChinese Drought and Flood Monitor

Conclusions

1) Global multimodel drought nowcasting is entirely feasible now

2) Consistent forcings are essential

3) Droughts only make sense when expressed relative to a

climatology – hence climatically consistent near real-time data

are essential

4) Skill in drought forecasting (not the focus of this talk) is critically

linked to getting the initial conditions (nowcast) right ... and still

a major challenge