Data Systems for LCLUC Research
The unprecedented large volumes of data for land use research have necessitated the development of innovative data processing, delivery and analysis systems. The evolving EOS Data and Information System and a number of competed research opportunities such as REASON and ACCESS, have provided support for data systems research and development. The MODIS Advanced Data Processing System (MODAPS) at the Goddard Space Flight Center (GSFC) is generating land-cover related products from the daily MODIS instruments on board the Terra and Aqua platforms. Data products at 250m -1km are being reprocessed as the algorithms are improved to provide consistent data records. This system is currently being enhanced to provide MODIS land product distribution capabilities to augment the services provided by the NASA Distributed Active Archive Center at the Eros Data Center to meet the needs of the MODIS science community.
The Landsat Ecosystem Disturbance Adaptive processing system is developing procedures for automated atmospheric correction and mosaicing of Landsat data and the generation of high resolution disturbance time series. The Global Land Cover Facility (GLCF) at the University of Maryland has developed a low cost system for processing and distribution of large volumes of land-cover data and enhanced data sets. Similarly, the Landsat.org project developed at Michigan State University (MSU) has developed a platform independent user interface and search engine for on-line purchasing, ordering and sharing of Landsat data worldwide. The Tropical Rain Forest Information Center at MSU provides Landsat derived data sets associated with monitoring tropical deforestation.
In partnership with the private sector, NASA purchased a global data set of cloud-free Landsat imagery for 1990 and 2000. These data were orthorectified and are easily accessible and freely available. They have greatly increased the use of Landsat data for LCLUC studies worldwide. In May 2003 the Landsat 7 scan line corrector failed and although the instrument continues to receive data, the imagery are of limited use. With no Landsat instrument ready to replace Landsat 7, there is an increasing data gap, posing a critical impediment to LCLUC science. The LCLUC program, working with the USGS is developing a mid-decadal (2004-2006) high resolution global cloud-free data set to extend the previous global data sets. The data set will include data from Landsat 5, ASTER, EO1 and Landsat 7 temporal composites. This data set will include data provided by foreign ground stations and possibly foreign high resolution satellites. It is hoped that international cooperation concerning this data set could provide a prototype for future international efforts to coordinate high resolution global data acquisition from the increasing number of high resolution assets in the framework of GEOSS.
Land use and land cover change studies at regional to global scales require large numbers of field sites for algorithm development and accuracy evaluation. Rapid development in integration of digital camera, hand-held GPS device, computer and internet make it possible for both scientific communities and citizens to collect and share geo-referenced field photos. The Global Geo-Referenced Field Photo Library, developed at the Earth Observation and Modeling Facility of University of Oklahoma, offers the capacity for users to upload, query (by themes and geographically), and download geo-referenced field photos in the library. It offers interactive capacity for users to interpret and classify field photos into relevant land cover types and builds photo-based land cover database. The users can use both photos and associated databases to carry out land use and land cover analysis in a geographical information system. The users who provide field photos can decide whether individual photos are to be shared or not. This tool and the resultant photo library will enable our NASA LCLUC communities to share their field photos, and promote the NASA LCLUC effort in remote sensing.