Monthly ETact for California, 2002

The dynamics in the ETact becomes apparent when viewing the following animation. The significant temporal variability is caused by changes in weather conditions, leaf area development and the changing soil moisture and soil salinity status. The monthly ET and biomass production data is organized by county, but could be compiled for other geographic areas, such as watersheds.

Eta animation Bio animation

Monthly ETact and BIO values per county (Arc/View shape, 0.2MB)

An alternative option for organizing the data is to express the temporal variability per land use category. A high resolution 30m land use database from 1992 made by the USGS National Land Cover Characterization team has been used for this purpose (see http://edcwww.cr.usgs.gov/programs/lccp/natllandcover.html ). The 21 land use classes have been resampled to reduce the number of classes to 12. The MODIS ET images have been disaggregated using the fixed specific presence of certain land use classes in each MODIS pixel. The monthly ET results of the six dominant land use classes are demonstrated. This graph describes the aerially averaged monthly ET for each land use class (50% of the pixels in that class have lower and 50% have higher ETact values).

The variation within each land use class can be significant due to the soil moisture status during the summer season. The evaporative demand of the atmosphere during June and July reaches its maximum, and depending on the water availability to the roots, this demand may or may not be met. Although the average wetland has an ETact of 200mm/month in July, the very top values are 350mm/month (+95% confidence) and the lowest values of dried down natural ecosystems have no moisture left for evaporation.

Total Annual ETact for California, 2002

Total annual evapotranspiration values were computed from monthly MODIS images; 12 original images were used to create this composite map that contains the annual accumulated values. SEBAL computes actual evapotranspiration (ETact) and dry matter production (BIO) simultaneously. The CIMIS network of automatic weather stations has been used in conjunction with the DAYMET high resolution climate grid to fill out the days between consecutive MODIS images (Thornton, P.E., S.W. Running, and M.A. White, 1997. Generating surfaces of daily meteorological variables over large regions of complex terrain. Journal of Hydrology, 190: 214-251).

 

2002_ETa.jpg 2002_BIO.jpg
Total Evapotranspiration, 2002 Total Biomass production, 2002

Download with total evapotranspiration and biomass production (2002)

annual evapotranspiration 2002 (Erdas Imagine, 4.2MB)
annual evapotranspiration 2002 (Arc/Info GRID, 4.0MB)
annual biomass production 2002 (Erdas Imagine, 4.6MB)
annual biomass production 2002 (Arc/Info GRID, 4.4MB)

The USDA has measured the ET of alfalfa throughout 2002 using a weighing lysimeter in the vicinity of Parlier (the “West” lysimeter has been used to better comply with farmer management conditions; the “East” lysimeter was kept at all possible optimal conditions and is an upper limit rather than a reflection of grower management). Their curve of measured crop ET is compared to the predicted crop ET by SEBAL (data provided by Jim Ayers and Richard Soppe, USDA). To avoid incompatible land use types, a MODIS pixel that comprises a high density of irrigated alfalfa fields has been selected and compared against the lysimeter rather than the pixel on which the lysimeter itself is located, because the direct surroundings of the research lysimeter contained vineyards with some intended induced moisture stress. The difference in the annual crop ET between the lysimeter (1167mm) and SEBAL (1178mm) is 0.9%.

California

Using MODIS satellite data, SEBAL was used to compute actual evapotranspiration and the biomass production for the entire state of California for 2002. FREE samples of this coarse-resolution (1000m) data are available below for different time scales. Additionally, high-resolution (30m) data for Kesterson National Wildlife Refuge are available for downloading. These two data samples illustrate the power and flexibility of SEBAL in generating ET data sets with different spatial and temporal scales. This flexibility enables design of cost-effective solutions for virtually all problems where accurate ET mapping is needed.

For users who have no access to Arc/View or Erdas you can download free viewers:
Erdas viewer (19MB)
ESRI ArcExplorerTMGIS viewer for vector and raster data (13MB)
For reference purposes you can download an Arc/View shapefile with the counties of California here

This table summarizes the data that can be downloaded from these product pages.

Erdas
Imagine
Arc/Info
GRID
Arc/View
shape
other
poster with annual ETact* and BIO in 2002 0.6MB (JPG)
annual ETact in 2002 4.2MB 4.0MB
annual BIO** in 2002 4.6MB 4.4MB
monthly ETact animation (2002) 0.5MB (GIF)
monthly BIO animation (2002) 0.3MB (GIF)
monthly ETact and BIO values per county (2002) 0.2MB
30 meter ETact 31 July 1996 (Kesterson National Wildlife Refuge) 9MB 8MB
30 meter BIO 31 July 1996 (Kesterson National Wildlife Refuge) 11MB 10MB
* ETact = actual evapotranspiration
** BIO = biomass growth (dry matter)
*** ET0 = reference evapotranspiration
**** Kc = crop coefficient

Products and Pricing

The SEBAL model has during the course of years been extended with different output layers, making it more useful to a larger spectrum of analysts. Whereas a River Basin Commission might be interested in water depletions only (i.e. ETact), the irrigation engineer and farmers want to see a map of crop water stress (i.e. ETpot-ETact). The following products are offered by SNA:

SEBAL output

Product Acronym Principle inputs to computation
Actual ET ETact Surface temperature, albedo and vegetation index
Reference ET ET0 Penman-Monteith for reference crop in conjunction with weather station data
Potential ET ETpot Minimum resistance based on Leaf Area Index in conjunction with weather station data
Biomass production BIO Absorbed Photosynthetical Active Radiation (APAR) in combination with light use efficiency that varies with soil water potential
Crop yield Yact Accumulation of biomass and crop dependent harvest index
Soil moisture MOI Evaporative fraction
Soil salinity Salt Ratio of minimum and actual canopy resistance in relation to soil water potential and soil moisture


These data can be delivered as daily, weekly, monthly or annual totals. SNA applies a flexible pricing structure that depends on:

  • Type of product(s)
  • Area covered
  • Spatial resolution
  • Time integration of the data
  • Required accuracy

Depending on these variables, SNA will compile a selection of the following 4 satellite images that all fulfill the minimum requirement of having a Thermal InfraRed (TIR) spectral band to measure surface temperature. The total number of images to be processed and their areal coverage dictate the total costs of the SEBAL products.

Satellite Sensor Product*** Spatial resolution Minimum revisit
frequency
Landsat Thematic Mapper ET 120 m 16 days
Landsat Thematic Mapper BIO 30 m 16 days
Landsat Enhanced Thematic Mapper* ET 60 m 16 days
Landsat Enhanced Thematic Mapper* BIO 30 m 16 days
Terra ASTER ET 90 m variable
Terra ASTER BIO 15 m variable
Terra** MODIS ET 1000 m 0.5 day
Terra** MODIS BIO 250 m 0.5 day
NOAA AVHRR ET 1100 m 1 day
NOAA AVHRR BIO 1100 m 1 day
* The Enhanced Thematic Mapper sensor has technical failures after May 2003
** Terra has a mid-morning overpass, whereas the similar Aqua has a mid-afternoon overpass. MODIS is on both Terra and Aqua; ASTER is on Terra only
*** The yield data follows the spatial resolution of the biomass production; the moisture and salinity data have the same resolution as the ET products

For a firm price quote for SEBAL products that satisfy your specific needs, please contact:

Grant Davids
(+1)-530-757 9200
grant@sebal.us

Oakdale Irrigation District, San Joaquin Valley (CA)

According to its mission statement, the Oakdale Irrigation District (OID) in California’s San Joaquin Valley is committed to “providing dependable irrigation and domestic water service to its constituents at the lowest and most efficient cost possible. The District is committed to excellence in resource management and all aspects of its operation.”

SEBAL technology was utilized to document the actual evapotranspiration from the irrigated lands during the peak period of crop water use. The actual evapotranspiration from SEBAL was used to improve the accuracy of evapotranspiration estimates used for the OID water balance.

Crop ET using SEBAL
Utilizing the California Department of Water Resources GIS shape file of the 1996 crop survey for Stanislaus County, crop-specific actual ET statistics were developed based on 30m by 30m pixels.

Crop-specific actual ET statistics by 30m by 30m pixel

crop Actual ET, mm/day
Minimum Average Standard
Deviation
Maximum
Alfalfa 4.98 7.51 0.63 8.43
Almonds 1.58 6.78 1.27 9.33
Beans 7.45 8.08 0.39 8.88
Fruit Trees 2.28 7.30 1.30 9.08
Grapes 2.87 6.49 1.13 8.62
Nonirrigated 2.07 5.35 1.44 8.89
Other 2.18 6.04 1.51 8.70
Pasture 1.10 7.14 0.89 9.19
Rice 2.57 8.32 0.74 9.24
Riparian Vegetation 5.80 7.62 0.65 8.93
Small Grain 2.24 5.44 1.59 8.74
Tomatoes 1.62 4.88 1.33 8.76
Urban 0.90 5.92 1.35 9.33
Walnuts 2.37 7.24 1.29 9.50


As expected, the average values are not far from the maximum values ranging from 55% for tomatoes (many probably close to harvest) to 91% for beans and 90% for rice. This indicates that this area has minimal deficit irrigation, a reflection of the District’s adequate water supplies. The average actual ET value for each crop was used to revise the District’s water balance.


3-D map of Actual ET in the Northern San Joaquin Valley on July 31, 1996.
(Composite developed from 30m x 30m pixel values.)

Benton Irrigation District, Yakima River Basin (WA)

The Yakima River flows for over two hundred miles through south central Washington, and, with its tributaries, drains about 6,150 square miles or 4 million acres. The river originates in Kittitas County from Keechelus and Kachess Lakes on the east side of the Cascade Mountains near Snoqualmie Pass. The Yakima River flows southeast through the Kittitas and Yakima Valleys, ultimately discharging into the Columbia River near Richland. Tributaries include the Cle Elum, Teanaway, and Naches Rivers, as well as numerous creeks and irrigation return flows. The confluence of the Yakima and Naches Rivers at the city of Yakima divides the Yakima River into “upper” and “lower” portions. Much of the water is diverted for irrigation in the Yakima Valley, but some is recovered through surface and subsurface routes.

The Benton Irrigation District (BID) has historically received Yakima River water at end of a 60 mile canal. Water loss from the canal and from the District’s distribution system are large. Under the provisions of the the Yakima River Basin Water Enhancement Project (YRBWEP), the District prepared a Water Conservation Plan (WCP). The WCP was based on a “bottom-up” process whereby a historical water balance was developed to estimate losses, individual conservation measures were identified, and viable measures were combined to form integrated conservation programs. The District selected an integrated conservation program that would eliminate losses by moving the diversion to a Yakima River pumping plant located near the center of the District and by constructing a new pressurized pipeline distribution system.

SEBAL was utilized to refine the historical water balance by documenting actual evapotranspiration from the irrigated lands during the peak period of crop water use. The actual evapotranspiration from SEBAL was used to improve the accuracy of actual evapotranspiration estimates under both with- and without-project conditions.

SEBAL estimates of ET (in mm) for different ecosystems in the Columbia river basin (Washington State) on June 25, 2000. Part A consist of irrigated pivots (119o35’51” W – 46o03’13” N), part B is riparian vegetation at the confluence of the Yakima and the Colombia (119o14’14” W – 46o15’07” N), part C are the forests near Pendleton (118o53’30” W – 45o20’39” N) and part D consists of rangeland (119o33’25” W – 46o13’48” N).


Without-Project Crop ET using SEBAL
With a GIS tax parcel coverage of the area and parcel-specific crop records provided by the irrigation district, crop-specific actual ET statistics were developed based on 30m by 30m pixels.

Crop-specific actual ET statistics by 30m by 30m pixel (mm/d)

crop Actual ET, mm/day
Minimum Average Standard
Deviation
Maximum 10%
exceedance
Kc x ETo
July
Cherry 0.0 6.9 1.8 9.1 8.6 8.4
Apple 3.0 6.4 1.0 8.6 7.6 9.1
Alfalfa 0.0 5.8 1.8 9.1 7.6 6.9
Nut 0.8 5.6 1.8 8.1 7.1 9.1
Grass Hay 0.5 5.3 1.8 8.4 7.4 7.1
Pasture 0.0 4.6 2.3 9.4 7.1 7.1


As expected, the crop with the greatest financial return, cherries, had the highest average actual ET at 6.9 mm per day. Cherries were followed in order by largest average actual ET by apple, alfalfa, nut, grass hay and pasture. The average values are significantly lower than the maximum values ranging from less than 50% for pasture to only 75% for cherries indicating a significant amount of deficit irrigation. The maximum values can be viewed as achievable actual ET values given sufficient water supplies. The average actual ET value for each crop was assumed to represent average historical conditions within the District and was used to revise the without project water balance.

With-Project Crop ET using SEBAL
Under the with-project condition, the existing open channel distribution system would be replaced with a pressurized pipe distribution system. Thus, the conveyance system losses will be negligible with the project. In this situation, an achievable actual evapotranspiration can be used together with an on-farm efficiency to directly determine the district diversion requirements.

An achievable actual evapotranspiration was based on a crop-specific actual ET rate exceeded by 10 percent of pixels in the SEBAL image and the with-project irrigated crop areas. SEBAL provided information resulting in more accurate actual evapotranspiration values for both the with and with-out project water balances.

U.S. Projects

SEBAL has been employed on more than 150 projects in 15 countries, including nine projects in the United States.
The U.S. projects are summarized in the table below.

River Basin State(s) Year(s) of Study Year(s) of Images Client Use of ET maps
Bear River Idaho, Utah, Wyoming 2000 1985 IDWR Bear River Compact Compliance on Depletions from Irrigation
Middle Rio Grande New Mexico 2000, 2003 2000, 2002 NMT, USBR Endangered Species Issues, Impacts of water table on ET (of riparian vegetation)
Tampa Bay Florida 2000 2000 Tampa Bay Water Impacts of ground-water pumping on ET of natural vegetation
Boise River Idaho, Oregon 2002 1997, 2001 USBOR, IDWR Transition of agriculture to residential, planning reservoir operation
East Texas Texas 2002 2000 State of Texas Multi-basin water balance
San Juan New Mexico 2002 2002 NavajoAgricultural Products Industry (NAPI) Management of Center Pivot Irrigation
Snake River Idaho 2003 2003 IDWR Water rights administration
Yakima River Washington 2003 2001 Benton ID,USBR Quantification of ET for refinement of historical water balance
Stanislaus River California 2003 2000 Oakdale ID Quantification of ET for refinement of historical water balance

Literature

Digital version of the full papers will be send to you upon request ( admin@sebal.us )(antoabureyhan@gmail.com)

  1. Bastiaanssen, W.G.M., M. Menenti, R.A. Feddes and A.A.M. Holtslag, 1998. The Surface Energy Balance Algorithm for Land (SEBAL): Part 1 formulation, J. of Hydr. 212-213: 198-212
  2. Bastiaanssen, W.G.M., H. Pelgrum, J. Wang, Y. Ma, J. Moreno, G.J. Roerink and T. van der Wal, 1998. The Surface Energy Balance Algorithm for Land (SEBAL): Part 2 validation, J. Of Hydr. 212-213: 213-229
  3. Bastiaanssen, W.G.M. and M.G. Bos, 1999. Irrigation performance indicators based on remotely sensed data: a review of literature, Irrigation and Drainage Systems 13: 291-311
  4. Bastiaanssen, W.G.M., 2000. Sensible and latent heat fluxes in the irrigated Gediz Basin, Western Turkey, J. of Hydr. 229: 87-100
  5. Bastiaanssen, W.G.M., D.J. Molden and I.W. Makin, 2000. Remote sensing for irrigated agriculture: examples from research of possible applications, Agricultural Water Management, 46(2): 137-155
  6. Bastiaanssen, W.G.M., R.A.L. Brito, M.G. Bos, R. Souza, E.B. Cavalcanti and M.M. Bakker, 2001. Low cost satellite data applied to monthly irrigation performance monitoring; benchmarks of Nilo Coelho, Brazil, Irrigation and Drainage Systems 15: 53-79
  7. Bastiaanssen, W.G.M. and K.M.P.S. Bandara, 2001. Evaporative depletion assessments for irrigated watersheds in Sri Lanka, Irrigation Science, vol 21(1): 1-15
  8. Bastiaanssen, W.G.M., M. M.D. Ahmad and Y. Chemin, 2002. Satellite surveillance of evaporative depletion across the Indus Basin, Water Resources Research, vol. 38, no. 12: 1273-1282
  9. Bastiaanssen, W.G.M. and S. Ali, 2003. A new crop yield forecasting model based on satellite measurements applied across the Indus Basin, Pakistan, Agriculture, Ecology and Environment, 94(3): 321-340
  10. Bastiaanssen, W.G.M. and L. Chandrapala, 2003. Water balance variability across Sri Lanka for assessing agricultural and environmental water use, Agricultural Water Management 58(2): 171-192
  11. Bastiaanssen, W.G.M., E.J.M. Noordman, H. Pelgrum, G. Davids and R.G. Allen, 2004. SEBAL for spatially distributed ET under actual management and growing conditions, ASCE J. of Irrigation and Drainage Engineering
  12. Droogers, P. and W.G.M. Bastiaanssen, 2002. Evaporation in irrigation performance and water accounting frameworks: an assessment from combined hydrological and remote sensing modeling, ASCE Irrigation and Drainage Engineering vol. 128(1): 11-18
  13. Pelgrum, H. and W.G.M. Bastiaanssen, 1996. An intercomparison of techniques to determine the area-averaged latent heat flux from individual in-situ observations: A remote sensing approach using the European Field Experiment in a Desertification-Threatened Area data, Water Resources Research, vol. 32(9): 2775-2786
  14. Schuurmans, J.M., P.A. Troch, A.A. Veldhuizen, W.G.M. Bastiaanssen and M.F.P. Bierkens, 2003. Assimilation of remotely sensed latent heat fluxes in a distributed hydrological model, Adv. in Water Resources, vol. 26(2): 151-159
  15. Scott, C.A., W.G.M. Bastiaanssen and M.D. ud-Din Ahmad, 2003. Mapping root zone soil moisture using remotely sensed optical imagery, ASCE Irrigation and Drainage Engineering, 129(5): 326-335

Satellites

Landsat Enhanced Thematic Mapper
satt_landsat.gif
Temporal resolution  – 16 days
Spatial resolution  –

30m (RED, NIR, SWIR, b1-5,7)
60m (TIR, b6)
15m (VIS, b8)
Spatial coverage  – 180×180 km
Operational since  – 1985 (Landsat5)
1999 (Landsat7)
Cost  – US$600 – 1500 per image
More info  – www.eurimage.com
www.landsat.org

 

NOAA AVHRR (Advanced Very-High Resolution Radiometer)
satt_noaa-avhrr.gif
Temporal resolution  – 14 times per day
Spatial resolution  –
1000m (RED, NIR, b1-2)
1000m (TIR, b4-5)
Spatial coverage  – 2400km
Operational since  –

1994 (NOAA14)
1998 (NOAA15)
2000 (NOAA16)
Cost  – Free of charge
More info  – www.noaa.gov

 

Terra MODIS (Moderate-resolution Imaging Spectroradiometer)
satt_terra_MODIS.jpg
Temporal resolution  – daily
Spatial resolution  –

250m (RED, NIR, b1-2)
500m (SWIR, b3-7)
1000m (TIR, b8-36)
Spatial coverage  – 2330 km
Operational since  – 2000
Cost  – Free of charge  –
More info  – modis.gsfc.nasa.gov

 

Terra ASTER (Advanced Spaceborne Thermal Emission and Reflection Radiometer
satt_terra_ASTER.jpg
Temporal resolution  – daily
Spatial resolution  –

15m (VNIR, b1-3)
30m (SWIR, b4-9)
90m (TIR, b10-14)
Spatial coverage  – 60 km
Operational since  – 1999
Cost  – US$50,- per image
More info  – asterweb.jpl.nasa.gov

 

SPOT VEGETATION
satt_SPOT.gif
Temporal resolution  – daily
Spatial resolution  –
1150 m (VNIR, b1-3)
1150 m (SWIR, b4)
Spatial coverage  – continents
Operational since  – 1998
Cost  – free for images older than three months
More info  – free.vgt.vito.be

Validation

Irrigated crops


The Bear River Basin covers portions of Idaho, Utah, and Wyoming and contains about 470,000 acres of crop and pasture land. For the basin, a comparison was made between SEBAL-based ET and available lysimeter data for two growing seasons.

For 1985 four Landsat-TM images during the growing season (July 14, August 15, September 16, October 18, 1985) evapotanspiration was calculated using the SEBAL algorithm. To calculate evapotanspiration for longer periods the Kc method was used. The crop coefficients, Kc, are defined as ET/ET0 where ET0 is reference ET based on an alfalfa-referenced Penman equation. Kc‘s were computed for each pixel and used to extrapolate ET from the day of the satellite image to days between images. ET0 helped to account for changes in ET caused by weather variation from day to day.

Kc values from the four Landsat dates and weekly lysimeter measurement periods are plotted in the figure below.

val_irr_through_time.gif

The results compare very well to lysimeter data for the last three image dates. The earliest date, July 14, compares well too, also, when examined in context of the impact of precipitation preceding the Landsat image date and rapidly growing vegetation during that period.

Predicted monthly ET values averaged +/- 16% as compared to the lysimeter at Montepelier. However, seasonal differences between SEBAL and lysimeters were only 4% due to impacts of reduction in the randomness of the error.

Summary of SEBAL- and lysimeter-derived ET values for weekly and monthly periods and the associated error

SEBAL
Monthly
ET (inch)
Lysimeter
Monthly
ET (inch)
Monthly
Kc for
Lysimeter
Monthly ET
(SEBAL)-
Lysimeter (%)
Seasonal
Error (%)
Jul 7.8 6.6 0.83 19
Aug 4.7 5.7 0.72 -18
Sep 2.6 2.1 0.47 22
Oct 0.9 0.9 0.51 -5
Jul-Oct 15.9 15.3 0.69 4 4.3

val_irr_season_inch.gif

 

Riparian vegetation


Information on the regional distribution of evapotranspiration in riparian corridors is of paramount importance for the water balance of river systems and for the preservation and management of natural habitats. As part of the Sahra Project in the Middle Rio Grande, the SEBAL model was applied to assess the consumptive use of river water resources by riparian evapotranspiration (ET). Landsat satellite data from the following days was used:
14 October 1999
7 April 2000
20 April 2000
14 September 2000
3 March 2002
6 May 2002
16 June 2002

The SEBAL predictions of ET in the environmental ecosystems along the river corridor were compared with sophisticated eddy covariance flux measurements conducted by Dr. James Cleverly from the University of New Mexico and Dr. Eric Small from New Mexico Tech. The flux measurements were conducted over cottonwood (both flooded and non-flooded), salt cedar, grass, shrub and mixed vegetation. The overall agreement across a wide range of ET fluxes (range is 0 to 0.307 inch/d) for sparse vegetation is good. A slight underestimation of ET by 4% can be derived from the graph shown below. The average measured ET on all image dates was 0.089 inch/d and the SEBAL estimated value was 0.086 inch/d, being an error of 2.7% without any calibration. These errors fall within the noise of the eddy covariance measurements and SEBAL. The errors in the seasonal values of ET are expected to be as low as 3%, due to reduction in the random error components of individual image dates.

Validation of SEBAL-based ET estimates of riparian vegetation in the Middle Rio Grande under arid conditions in New Mexico for various image dates.

val_nat_veg.gif
 

Pastures


The Agricultural Research Station from the USDA in Greenbelt (MD) has organized a regional scale energy balance field experiment on the prairies of Oklahoma ( http://hydrolab.arsusda.gov/sgp97/ ). They have collected surface flux data from eddy-correlation and Bowen ratio systems in the Little Washita Watershed in the vicinity of El Reno (northwest of Oklahoma City). This experiment is also known as the Southern Great Plains (SGP) Hydrology Experiment project.

val_pastures_SGP.gif

The USDA is interested in the promotion of the ASTER satellite that has bands in the 15m (VNIR), 30m (NIR) and 90m (TIR) range. SEBAL was applied on an ASTER image of 10 June 2001. USDA scientists validated the results. The results are summarized below.

val_pastures.gif
 

val_pastures_overview.jpg
 

Playas


The very first prototype of SEBAL was developed in 1988 (SEBAL88) for estimating the evaporation from the playas in the Qattara Depression in Egypt. The Qattara Depression is a 5,000,000 acre sub-sea level depression with the deepest point having an elevation 440 feet below sea level. Evaporation rates from the bare and hyper-saline soils vary widely due to variability in soil types and depths to the shallow groundwater table. The spatial variations were measured simultaneously with 3 mobile Bowen ratio surface energy balance stations. Five different desert field campaigns were organized in different parts of the year to capture seasonal variation in evaporation

Surface type Water table (inch) Relative Occurrence (%)
Limestone >197 35.1
Clayey-shales 18-26 3.6
Gravel >197 4.0
Sandsheet >79 16.8
Sand >35 0.4
Sandcrust 17-43 18.9
Hummocky 8-28 6.0
Puffy 13-22 11.4
Salt crust 4-16 3.5
Salt floes 2-6 0.2
Brine 0 0.1
Total 100
“puffy” type of soil in the Qattara Depression (Egypt)
playa.jpg


The validation of SEBAL is demonstrated in the next graphs. SEBAL88 did not have an instantaneous to daytime conversion routine. For this reason, only Landsat overpass time evaporation fluxes are shown. The graphs demonstrates that there was one outlier (1 point out of 9 points, thus 11%) and that the remaining points are around the 1:1 line, which implies that there is no bias towards the wet or dry end.

Comparison of instantaneous ET as computed by SEBAL and as measured by Bowen ratio surface energy balance stations in the Qattara Depression of Egypt (June 1988)
val_playa_fig_inch.gif
SEBAL88 performance for the playa soils in the Qattara Depression (June 1988) with and without one outlier

Method Playa evaporation
SEBAL averaged for all 9 points 0.0100 inch/hr
Bowen ratio measurements for all 9 points 0.0095 inch/hr
Difference 6.3 %
SEBAL averaged for 8 points 0.0096 inch/hr
Bowen ratio measurements for 8 points 0.0095 inch/hr
Difference 1.3 %
Temporal variation of area-averaged playa evaporation in the Qattara Depression, Egypt

Net radiation
(inch/d)
Evaporative
Fraction (-)
Evaporation
(inch/d)
January 0.051 0.470 0.024
February 0.079 0.475 0.039
March 0.138 0.480 0.067
April 0.189 0.490 0.094
May 0.209 0.495 0.102
June 0.248 0.500 0.126
July 0.244 0.500 0.122
August 0.209 0.495 0.102
September 0.157 0.490 0.079
October 0.130 0.485 0.063
November 0.083 0.480 0.039
December 0.047 0.475 0.024
Average 0.073

source: W.G.M. Bastiaanssen and M. Menenti, 1990, Mapping groundwater losses in the Western Desert of Egypt with satellite measurements of surface reflectance and surface temperature, in (ed.) J.C. Hooghart, Water Management and Remote Sensing, TNO Committee on Hydrological Research, Proceedings and Information no.42