Skip to main content
ARS Home » Pacific West Area » Maricopa, Arizona » U.S. Arid Land Agricultural Research Center » Water Management and Conservation Research » Research » Publications at this Location » Publication #154685

Title: TRACKING SPATIAL AND TEMPORAL COTTON DT PATTERNS WITH A NORMALIZED DIFFERENCE VEGETATION INDEX

Author
item Hunsaker, Douglas - Doug
item Pinter Jr, Paul
item Fitzgerald, Glenn
item Clarke, Thomas
item Kimball, Bruce
item BARNES, EDWARD - COTTON INC CARY NC

Submitted to: Irrigation Associations Exposition and Technical Conference Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 7/30/2003
Publication Date: 11/18/2003
Citation: Hunsaker, D.J., Pinter Jr, P.J., Fitzgerald, G.J., Clarke, T.R., Kimball, B.A., Barnes, E.M. 2003. Tracking spatial and temporal cotton dt patterns with a normalized difference vegetation index. Irrigation Associations Exposition and Technical Conference Proceedings. pp. 126-137.

Interpretive Summary: Irrigated agriculture faces increasing pressures to ensure that water is used as efficiently as possible. Precision irrigation is an irrigation management technique aimed at improving water use efficiency by tailoring water applications to match the actual crop water requirements for different areas within a field. However reliable estimates of the actual amount of water used in crop evapotranspiration (ETc), which is the combined crop transpiration and evaporation from the soil, are needed for the different areas to determine the appropriate irrigation amount. Although generalized crop coefficient (Kc) curves are widely used to estimate the ETc use for the typical or average crop condition within the field, these are not useful for determining the variable ETc use that occurs within large fields. This study presents a method to estimate a Kc curve for cotton from remote sensing measurements of crop reflectance that can be routinely measured either on the ground, in the air, or by satellite. Results indicate that the crop reflectance measurements can be used to predict the actual Kc exceptionally well over a wide range of cotton ETc conditions. Development of this remote sensing technique is expected to provide the precision for determining the actual crop water requirement within the field, which is required for precision irrigation management. This technique will be of interest to farmers, farm consultants, government agencies, and the irrigation industry.

Technical Abstract: Crop coefficients (Kc) are widely used to estimate crop evapotranspiration (ETc) for determining irrigation scheduling. Generalized Kc curves are limited to providing daily estimates of ETc for the 'typical' crop condition within a field. However, precision irrigation requires spatial and temporal ETc information in order to determine the proper water replacement to each management zone. An irrigation experiment conducted during 2002 in Arizona explored the use of remotely-sensed surrogate basal crop coefficients (Kcb) for quantifying spatial and temporal differences in cotton ETc. The main treatment included two irrigation scheduling approaches that were based on ETc calculation procedures of the Food and Agriculture Organization Paper No. 56 (FAO-56) but differed only by the Kcb estimation: 1) a locally-derived FAO-56 Kcb curve (FAO), and 2) Kcb values based on ground-measured normalized difference vegetation index (NDVI) using a previously defined Kcb-NDVI relationship (developed for a different cotton cultivar and row-orientation than for the experiment). Additional variables (3 plant densities, 2 N levels) were included to induce variations for crop ETc patterns within irrigation scheduling treatments. The ETc estimation and irrigation scheduling using the FAO-56 Kcb curve provided better irrigation management than the previously defined Kcb-NDVI relationship, resulting in significantly higher yields for FAO than NDVI. The Kcb-NDVI relationship employed in the experiment underestimated measured Kcb values during much of the season. The primary problem was related to factors, e.g., the different row-orientation, that effectively lowered NDVI values compared to those that occurred in the previous experiments and were used to develop the relationship. However, measured NDVI tracked the spatial and temporal variations in measured Kcb exceptionally well during the season. New Kcb-NDVI relationships based on the 2002 data were presented and are currently being tested during 2003 under a similar cotton irrigation scheduling experiment. Although additional research is needed to develop more robust NDVI-based Kcb prediction, findings to date indicate the potential for NDVI to provide near-real-time feedback for attaining Kcb that closely track actual crop ETc trends within a field, a technique that could help govern site-specific cotton irrigation scheduling.