Scholarship Claire Williams - Flinders University
The Murray Darling Basin is Australia’s largest agricultural area and a major user of water for irrigation. The Basin’s capacity to supply water is fast reducing due to increased extraction for industrial uses, domestic supply and, most significantly, agricultural irrigation. The Basin contains approximately 72% of Australia’s irrigated crops; therefore irrigation needs to become more efficient in order to match supply with demand (MDBMC, 2007).
Recent droughts have increased investment in improving irrigation efficiency. A common measure of this efficiency is the ratio of seasonal crop water use to seasonal irrigation application. In order to assess whether increased investment is improving efficiency we need to develop accurate estimates of the rate of water use at the individual crop and district level.
The rate of water used by a crop, ET = evapotranspiration, depends on weather, growth stage and soil water availability. Crop yield is a function of water use. Crop water use can be estimated under well watered conditions as a function of reference crop evapotranspiration (ETo) and a set of crop coefficients. The coefficients (Kc) are crop specific. Crop evapotranspiration (ETc) is estimated as the product of the rate of reference crop evapotranspiration and the appropriate crop coefficient (ETc = ETo X Kc) (Allen et al., 1998). However the crop coefficient does not account for variations in canopy cover between different areas.
Reference crop evapotranspiration (ETo) is the evaporation from a grass reference crop without a shortage of water that shows certain characteristics (Allen et al., 1998). Climatic parameters are the only factors affecting ETo and so ETo can be calculated from weather data. The variations in the values of ETo with location and season reflect the temporal and spatial variation in the evaporative influence of the atmosphere. These values are not dependent on soil characteristics (Allen et al., 1998).
Evapotranspiration (ET) is the sum of evaporation and transpiration. It is the transport of water into the atmosphere from the earth’s surface. It is one of the main consumers of solar energy at the earth’s surface and is one of the most significant components of the hydrological cycle. The energy used for ET is often referred to as the latent heat flux (Burba et al., 2006).
Evaporation is the process whereby water is directly returned back into the atmosphere through evaporative loss from soil surfaces, standing water and other water surfaces. Transpiration is the process in which water is used by vegetation and consequently lost back to the atmosphere as water vapour. The water enters through the root zone of the plant and is then used for different biophysiological processes such as photosynthesis. Water then passes back to the atmosphere through the leaf stomata in the form of vapour. If the leaf becomes stressed to the wilting point, transpiration will stop (Burba et al., 2006).
Evapotranspiration is a function of soil water content (SWC), crop stage and canopy cover (CC). Matching supply with demand is a function of the rate at which water is being used by the plant and how water is being stored in the root zone. Supply will depend on SWC the sufficiency of which can be inferred from measurements of pre-dawn leaf water potentials, while water demand is dependent on crop growth stage and ET.
The eddy covariance technique (EC) is an atmospheric flux measurement technique used to measure and calculate vertical turbulent fluxes including wind speed within atmospheric boundary layers. An eddy covariance system generally measures carbon dioxide, air temperature, moisture and 3-D wind speed above a crop canopy. The net moisture flux is a
result of soil evaporation plus plant transpiration minus precipitation and condensation (Glen et al., 2008). The EC measurement represents the flux from a specific area of crop. The size and location of this area relative to the tower site depends on wind speed, wind direction, atmospheric stability and tower height above the canopy. The area is called flux footprint and is the upwind area which is the source of the atmospheric flux measured by the instruments (Glen et al., 2008).
The main source of variation in the tabulated values of Kc is the growth stage. At a given growth stage, the rate of water use by an individual vine, for example, is proportional to its canopy cover. However the tabulated values of Kc do not account for canopy cover variations between vineyards. The canopy cover of a vineyard can be estimated from remotely acquired measures of vegetative indices.
Vegetation indices are derived from measures of the way that plant canopies modify light radiation. The normalised difference vegetation index (NDVI) is a widely known example of such an index. The NDVI is a numerical indicator used to detect live green plant canopies in multi-spectral remote sensing data. It is an index used to identify the condition of vegetation in different areas. The NDVI is calculated from the visible and near-infrared light reflected by vegetation (Weier and Herring, 2010).
The NDVI can be used as a measure of canopy cover (CC). Trout and Johnson (2007) and Trout et al. (2008) have shown that there is a strong connection between NDVI and CC as NDVI was found to increase linearly with canopy cover up to approximately 0.8. Therefore NDVI can be used as a surrogate for measures of CC. Ayars et al. (2003), Williams and Ayars (2005) and Goodwin et al. (2006) have shown that the crop water use of individual peach trees (which show similar characteristics to almond trees) and vines is linearly related to projected canopy cover.
Evapotranspiration data were collected using eddy covariance towers in a vineyard and an almond orchard between February 2007 and June 2009 in South Australia’s Riverland. The project aimed to use these comprehensive data sets to explore whether the water use of an entire vineyard and an entire almond orchard can be estimated from reference crop evapotranspiration and crop coefficients adjusted with satellite measures of NDVI to account for variations in canopy cover.