METRIC™ Project

Overall project goal:

To accurately quantify net Consumptive Water Use (CWU) for different crops and range land vegetation by processing LANDSAT images, and to develop remote sensing image acquisition and computing expertise using the METRIC™ procedure to produce area-specific consumptive water use (CWU) maps. The 1977 maps are being developed by the University of Idaho (UI), and the 2002 and 2005 maps are being developed by the University of Nebraska-Lincoln (UNL). UNL is developing, testing and demonstrating tools that will take these CWU maps and turn them into immediately usable products for planning, managing and regulating groundwater resources. UI will provide training in applying the METRIC™ equations critical to understand the data process including assumptions, and to make consistent and dependable decisions. These CWU maps will be used to estimate net water use during the 1997, 2002 and 2005 growing seasons.

Specific objectives: To develop, test and demonstrate the tools that will take these CWU maps and turn them into immediately usable products for planning, managing and regulating groundwater resources. The tools and products that will be developed and demonstrated include:

1. Developing CWU maps for areas irrigated by groundwater sources and by surface water supplies. This will help develop water balances for these areas for assessing recharge.
2. Sampling of CWU data across the project area and develop locally calibrated Kc curves for specific crops [alfalfa, wheat, corn, and sugarbeets]. This information is critical in setting irrigation scheduling, and in ground-water modeling works.
3. Produce maps of net differences in CWU from irrigated agriculture and rain fed (dryland)
   agriculture;
4. Produce maps of net differences in CWU from irrigated agriculture and CWU from natural vegetation.

The products will be shared between UNL and natural resources districts, the Nebraska Department of Natural Resources, and USDA Natural Resources Conservation Service offices.

Expected Benefits:

1. UNL will acquire the technical knowledge to process LANDSAT images to create CWU maps. LANDSAT is much more valuable than MODIS images because of the ability to see and identify ET from individual fields and specific crop and system types. The CWU maps will be used to estimate consumptive water use of agricultural vegetation which can be used to derive water balance maps and reports. The technical tools and products will be shared with the Natural Resources Districts in the Panhandle.
2. The Natural Resources District will use the estimates of CWU, water balances, and specific studies as a basis for developing revised water conservation policies and practices.
3. New water management policies will require producers to learn to better manage and use ground water and surface water which could include adopting irrigation scheduling using crops coefficients that are specific to the watersheds.
4. This dynamic tool will be readily transferable to similar groundwater depleted regions in other parts of Nebraska. UNL will have the expertise to educate other NRD’s about using the technology to help explain the process of using water balance and crop coefficient information to help manage water resources. The transfer of technology is expected to be cost effective and adopted seamlessly across institutions as “demand” will grow. In the near future the costs of generating CWU maps and for conducting studies will be charged to the institutions.

Methods

Nebraska has a well developed system of regional weather stations that provides real-time reference ET (ETr) as part of the High Plains Regional Climate Center (HPRCC). The current procedure for estimating actual CWU includes the commonly used crop coefficient (Kc) times reference ETr method, where a ‘representative’ Kc curve is applied that describes the CWU from a specific crop relative to the ETr over specific periods of time. While HPCC data provides ETr estimates, we still do not have a practical way of applying it on a watershed scale across time to derive CWU estimates. METRIC has the ability to do large scale areas that encompass both spatial and temporal variability without time consuming and costly research. 

The Kc - ETr approach is a conservative method that has had wide application for more than 40 years. The procedure is fairly accurate with an error of approximately +/- 20% (Allen et al., 2005a). Errors can be biased high or biased low for a particular region due to specific crop characteristics and cultural behavior of the region. Biases in Kc also result from water shortage that may be undocumented or that was not reflected in the development of a specific Kc curve.  With the varying water availability during the past 5 years, this is a major concern as not all crops were ‘fully irrigated’. 

Substantial uncertainty exists in the ability of single Kc data sets typically derived for specific crop varieties to accurately estimate CWU for the wide range of conditions in the Nebraska Panhandle area. Impacts by irrigated agriculture on water resources are essentially caused by differences between CWU from irrigation and CWU from natural vegetation. These differences represent the net depletion to the ground-water source. Both of these CWU amounts are best determined by accurate satellite-based energy balance due to the potentially large spatial and temporal variations.

Since the METRIC™ model calculates ET from satellite images and weather data using the surface energy balance method, the satellite image is only for one point in time and instantaneous ET flux is only calculated for that point in time.  Satellite images throughout the growing season from the area are used to calculate ET and then ‘stitched together’ for a seasonal ET map of the area.  Requirements for accurate results using METRIC™ are:  satellite images that include thermal (surface temperature) information and that the images are cloud free; header file information for the images; high quality weather data; reference evapotranspiration; and sufficient images for the entire growing season.  A land-use map that divides the area of interest into various general classes of land-use such as agriculture, water, desert, forest, grassland, etc. is helpful.

METRIC™: METRIC™ stands for Mapping EvapoTranspiration at high Resolution using Internalized Calibration.  METRIC™ is a computer model that calculates ET via a surface energy balance that is driven by satellite (Landsat) images containing both short wave and thermal information.  The advantage of ET by energy balance is that actual ET is obtained rather than potential ET that is typically represented by crop coefficients. Actual ET can be less than potential ET due to effects of water shortage, low irrigation uniformity, salinity of soil and water, sparse vegetation, water logging and disease.

METRIC™ calculates ET through a series of computations that estimate net surface radiation, soil heat flux, and sensible heat flux to the air.  By subtracting the soil heat flux and sensible heat flux from the net radiation at the surface, a ‘residual’ energy flux can be estimated that represents the energy consumed by evapotranspiration (i.e. energy that is used to convert liquid water into water vapor).

METRIC™ is designed to produce high quality and accurate maps of ET for focused regions smaller than approximately one hundred miles in scale and at high resolution.  Internal calibration of the energy balance in METRIC™ utilizes ground-based reference ET (ETr) to ‘tie-down’ the derived energy balance.  The basic energy balance equation is shown in this graphic.

METRIC™ was developed by R. Allen, M. Tasumi, R Trezza, W. Bastiaanssen, T. Morse,
W. Kramber, J Wright, University Of Idaho, USDA/ARS, Raytheon, NASA, IDWR.

Project Team

Dr. Gary W Hergert, Professor
University of Nebraska
Panhandle Research & Extension Center
308-632-1230
ghergert1@unl.edu

Dr. Ayse Irmak, Assistant Professor
University of Nebraska
School of Natural Resources
402-472-8024
airmak2@unl.edu

Dr. Richard G Allen, Professor
University of Idaho
Kimberly R & E Center
208-423-6601
rallen@kimberly.uidaho.edu

Dr. Jeppe Kjaersgaard, Research Assistant Professor
University of Idaho
Kimberly R & E Center
208-423-6672
jeppek@kimberly.uidaho.edu

Gary Stone, Extension Educator
University of Nebraska
Panhandle Research & Extension Center
308-632-1230
gstone2@unl.edu

Peggy Penrose, GIS Technician
University of Nebraska
Panhandle Research & Extension Center
308-632-1230
ppenrose2@unl.edu

Ian C Ratcliffe, Remote Sensing & GIS Research Specialist
University of Nebraska
Center for Advanced Land Management Information Technologies
402-472-6514
Iratcliffe2@unl.edu