Satellite Mapping for Irrigation Management

…develop and demonstrate a prototype decision support system that can efficiently deliver crop coefficient and estimated crop water use information to agricultural producers and water suppliers…

NDVI from satellite

ETo from ground weather station(s)

crop coefficient Kc

ETc = ETo * Kc

canopy cover

"Reflectance-based" crop coefficients

Study area

ETo

Crop cover

Weighing lysimeter

y = 1.45x - 0.13R2 = 0.92

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9NDVI

4/54/225/75/236/86/247/107/268/279/129/2810/14

y = 1.45x - 0.13R2 = 0.92

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9NDVI

OtherBellpepperBroccoliCottonGrapeLettuceMelonsafflowerTomatoesWatermelonOnion

2008 results

<22-33-44-5>5

ET (mm)

Apr 4

Jun 7

Aug 26

Oct 13

ET maps

ETo, ETc for Study Area

0

1

2

3

4

5

6

7

8

9

0 50 100 150 200 250 300 350DOY

ETc mean, std devETo

Spatially aggregated

Daily ETc estimated by RESET2 vs NDVI (2006)

y = 1.0165x + 0.136R2 = 0.9099

y = 1.1095x - 0.1896R2 = 0.7936

0

2

4

6

8

10

12

0 1 2 3 4 5 6 7 8RESET2 ETc (mm/d)

from NDVIfrom CCLinear (from CC)

Comparison with energy-balance approach

Further research

• Modeling soil evap & stomatal regulation• Temporal interpolation• Nuts-n-bolts remote sensing issues (sun/view

angle, soil background)• Lysimeter• Validation• Optimal approach?• Decision tool development