# Water from 'Landsat Imagery: A Unique Resource'

## Science Center Objects

Landsat satellites provide high-quality, multi-spectral imagery of the surface of the Earth. These moderate-resolution, remotely sensed images are not just pictures, but contain many layers of data collected at different points along the visible and invisible light spectrum.

Large cumulonimbus clouds. Photo by Mindy Ritchie.

Flood Extent and Disaster Monitoring, Australia

In Australia, between 1852 and 2011, at least 951 people were killed by floods, another 1,326 were injured, and the cost of damage reached nearly $5 billion dollars. More recently, the southeast Queensland floods during the 2010–11 season left much devastation, destroying 75 percent of the banana crop and, as a result, inflating banana prices (Carbone and Hanson, 2013). Floods have devastated communities and negatively impacted the regional and national economies in Australia. Landsat imagery, in conjunction with MODIS and several other commercially acquired satellites, has been used to map flood extent to provide situational-awareness information to emergency services in order to save lives and mitigate economic impact. Landsat image of Coro, Falcon area, Venezuela. Courtesy of Exploration Signatures. Exploration Signatures: Water Exploration in Venezuela In 1995, Exploration Signatures (then the Geologic & Hydrologic Division of Earth Satellite Corporation) used Landsat ETM+ imagery to complete water exploration work in Venezuela’s Falcón State. The project was funded through a guarantee from the Export-Import Bank of the United States, administered by Harza Engineering Company, and performed for HidroFalcón C.A., the Venezuelan state water agency. National Flood Risk Information Program, Australia On July 1, 2012, Australia commenced the National Flood Risk Information Program (NFRIP), which includes mapping of flood extents from the entire historical archive of Landsat data over Australia (National Flood Information Portal, 2013). The Australian Government’s Natural Disaster Insurance Review highlighted the lack of consistency across the country in the way flood-risk information was collected and made available to users. It also recognized the need for consumers to be aware of the natural disaster risks they may face, as well as the benefits of making flood-risk information more readily accessible. The aim of the NFRIP is to improve the quality, availability and accessibility of flood-risk information across Australia, and raise community awareness of flood risks (Medhavy Thankappan, Geoscience Australia, written commun., 2013). Flood mapping in 2002 in the Barmah-Millewa forests, north-central Victoria and south-central New South Wales, Australia. Courtesy of the Victoria Department of Environment and Primary Industries. Flood Extent Monitoring in Riparian Forests, Australia The results of flood mapping are used as an input to hydrological models in some instances, and in other cases as an independent source used for the calibration of model outcomes. Hydrological models are actively used as management tools. The cost of satellite images has a direct bearing on how often flood mapping is carried out to support forest management. In the absence of Landsat data, alternative sources (for example, SPOT-5) cost about$3,853 (\$4,000 in Australian dollars) for the Barmah-Millewa forests (fig. 1), and the same amount for the Gunbower and Koondrook-Perricoota forests within the Murray Valley riparian systems. The current frequency of monitoring and flood mapping would not be possible without Landsat imagery availability (Kathryn Sheffield and Mohammad Abuzar, Department of Environment and Primary Industries , written commun., 2013).

Landsat Thematic Mapper image was taken June 2006. Courtesy of the U.S. Bureau of Reclamation.

Monitoring Consumptive Water Use for Global Crop Production

Monitoring consumptive water use is an important component of global agricultural monitoring (Curt Reynolds, USDA FAS, written commun. and oral commun., 2014), as a majority of global production relies at least in part on irrigation for crop production (fig. 1). Globally, irrigated agriculture supports production of about half of the world’s food supply (Thenkabail and others, 2010). In the United States (U.S.), agriculture accounts for around 80 percent of consumptive water, reaching upwards of 90 percent in many western states (USDA Economic Research Service, 2014).

NDVI (Normalized Difference Vegetation Index) maps derived from Landsat 5 imagery in the Wimmera Region, north-central Victoria, Australia. Courtesy of the Victoria Department of Environment and Primary Industries.

Land Cover Mapping: Australia

Australia’s agricultural industry has evolved significantly within the last decade. Change in agriculture, whether it is an increase in land used for production or the types of crops being produced, has an impact on land-cover soil properties and water availability. Landsat imagery is used in monitoring irrigation areas and dry land agricultural areas to detect changes in agricultural practices over time. Additionally, land cover is also used in benchmarking of water-use efficiency which supports irrigation water management at the catchment level.

Comparison between geologic maps of Ethiopia, the first drawn in 1972 over a topographic map from 1968 and the second consisting of a 2004 Landsat image georeferenced on SRTM. Courtesy of Radar Technologies International.

Land Cover Mapping: Africa

While working in Africa, Dr. Alain Gachet of RTI saw a need for new maps of the continent, particularly inland maps. Almost all geological maps of Africa were created during the 1960s before some African countries became independent. The coastal maps are fairly accurate, as they were required to meet the standards of the French, English, and U.S. Navies. Inland maps beyond 50 km from the shore, however, diverge significantly, with geolocation errors of several kilometers.