|Kustas, William - Bill|
|Evett, Steven - Steve|
|AGAM, N. - Agricultural Research Organization Of Israel|
Submitted to: Irrigation Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/21/2015
Publication Date: 4/12/2015
Citation: Kustas, W.P., Alfieri, J.G., Evett, S.R., Agam, N. 2015. Quantifying variability in field scale evapotranspiration measurements in an irrigated agricultural region under advection. Irrigation Science. DOI: 10.107/S00271-015-0469-1.
Interpretive Summary: While groundwater depletion has been observed in many parts of the world, it is most significant in arid and semi-arid regions where there is intensive irrigated agriculture. Since agriculture is the largest consumer of fresh water, monitoring water use, irrigation, and other agricultural practices at the field scale is prerequisite for managing water resources effectively. The objective of this study was to analyze differences in evapotranspiration (ET) measurements from current micrometeorological and water balance techniques in a pair of adjacent irrigated cotton fields. In addition to the ground-based ET measurements, high resolution aircraft imagery collected several times during the period of rapid cotton growth and development provided spatially-distributed information regarding plant cover and leaf area information, which permitted investigating the relative influence of variation in plant cover within the various ET source areas for the different ET measurement techniques. The discrepancy in the ET estimates between the methods was largely attributed to variations in vegetation cover within the source areas affecting the measurements. This suggests that ET estimates from the different methods do not necessarily reflect the field-scale water use. Moreover, this study indicates that the source area contributing to the measurements must be carefully analyzed before making any general conclusions concerning water use differences due to crop type and management practices or using the data for model validation and calibration.
Technical Abstract: This study compares the evapotranspiration (ET) measurements from eddy covariance, lysimetry, and water balance using a network of neutron probe sensors and investigates the role of within-field variability in the vegetation density in explaining the differences among the ET estimates from the various techniques. Measurements were collected in two adjacent fields of irrigated cotton during a 36-day period of rapid crop growth as part of the 2008 Bushland Evapotranspiration and Agricultural Remote Sensing Experiment (BEAREX08) conducted under strongly advective conditions in the Texas High Plains. Using neutron-probe-based water balance estimates as the reference ET, the results indicated that differences in cumulative ET estimates from the eddy covariance systems and neutron probe ranged between 2 and 14%. In contrast, differences in ET estimates from the lysimeters and neutron probe ranged between 22 and 25%. The discrepancy in the ET estimates between the three methods can be attributed in large part to variations in vegetation cover within the source areas of the sensors, which could only be reliably assessed using high resolution remote sensing imagery. Differences in the total irrigation and rainfall amounts, as well as the row orientation between the two fields, confounded the relationship between the discrepancies in ET and vegetation densities in the measurement source areas. Nevertheless, the analysis indicates that the source area contributing to the measurements must be considered, even in instances where one might consider field conditions uniform. This suggests that the ET measurements often used as “ground truth” may not necessarily reflect the field-scale ET. Any general conclusions concerning water use differences due to crop type and management practices, or use of the derived LE fluxes for model validation and calibration requires the variability in vegetation cover conditions be taken into account. The latter statement concerning model validation is reinforced by results comparing thermal-based energy balance model estimates of ET with those from lysimeter and eddy covariance measurements used for model validation.