Evaluation of Vegetation Index of Passive and Active Sensors for Estimating Crop Coefficients and Evapotranspiration.

A promising method for estimating variable crop evapotranspiration (ETc) rates in irrigated fields utilizes vegetation indices to spatially monitor crop coefficients. In previous studies in Maricopa, Arizona, the basal crop coefficient (Kcb) was derived as a unique function of the normalized difference vegetation index (NDVI) for several crops. For these functions, NDVI data were obtained with measurements from a four-band, passive radiometer. The passive radiometer relies on incident sunlight to illuminate the target and the radiance data obtained must then be calibrated to canopy reflectance for calculating NDVI. Because Kcb can change rapidly with time, using remotely-sensed Kcb for irrigation scheduling requires frequent NDVI data, generally about every few days. Thus, for gathering frequent Kcb data, a major obstacle when using passive sensors for NDVI is that data quality can be highly affected by sky conditions. Ideally, cloudless and dust-free skies are needed for high-quality passive sensor data. Active sensors, containing their own light source, also provide NDVI data and do not require calibration. Thus, they are considered much more flexible than passive sensors in terms of timeliness and variable illumination conditions. However, active sensors are generally limited to ground-based sensing because of the short range of the light source. Nevertheless, they may offer a good alternative to passive sensors for the Kcb application. Some sensor-comparison work has been done on the use of active versus passive sensors for monitoring crop development and biomass. Not much research has been conducted to date on comparing the use of active sensors versus passive sensors for assessing the basal crop coefficient. The objective of this paper is to compare the NDVI from active and passive sensors obtained during plot studies of irrigated cotton, wheat, and camelina in Maricopa, Arizona. Results will include evaluations of the derived relationships between Kcb and NDVI for the two sensors, as well as, those between fractional canopy cover and NDVI.