A physically-based model, Energy: Surface Towards Atmosphere (ESTA), was used to model and map the energy and water balances of a heterogeneous land surface in a savanna environment on the southern fringe of the Okavango Delta, near Maun, Botswana. ESTA is governed by remotely sensed values of surface temperature, reflection, and vegetation density. Surface reflectance data from the MODIS sensor aboard the Terra satellite were obtained for the Okavango Delta region for September of 2001.

The subset of the MODIS MOD04_L2 aerosol product provided in this data set represents the swaths that coincide with known times of the South African Weather Bureau/Service (SAWS) Aerocommanders JRA and JRB research aircraft missions to support aerosol research and validation activities for the SAFARI 2000 region. The MODIS aerosol product monitors the ambient aerosol optical thickness over the oceans globally and over a portion of the continents. Further, the aerosol size distribution is derived over the oceans, and the aerosol type is derived over the continents. Daily Level 2 data are produced at the spatial resolution of a 10 x 10 1 km (at nadir) pixel array. The daily data files cover the period August 21, 2000, through September 26, 2000. For some data collection dates, there are two or more data files.The MOD04_L2 swath data files included in this data set are from the GSFC DAAC (V4.1.0, Collection 004). The MODIS Level 2 data files were converted from Hierarchical Data Format (HDF) to granule tables (GRANT) format. The GRANT format provides the extracted Scientific Data Set (SDS) in ASCII table form where each pixel (x,y) is represented as a row of data with georeferencing information and each SDS is provided as a separate column in the table. The ASCII tables are in comma-delimited format.

The Filled Land Surface Albedo Product for Southern Africa, which is generated from MOD43B3 Product (the official Terra/MODIS-derived Land Surface Albedo - http://geography.bu.edu/brdf/userguide/albedo.html ), is a subset of the global data set of spatially complete albedo maps computed for both white-sky and black-sky at 10 wavelengths (0.47mm, 0.55mm, 0.67mm, 0.86mm, 1.24mm, 1.64mm, 2.1mm, and broadband 0.3-0.7mm, 0.3-5.0mm, and 0.7-5.0mm). An exception is that no 2.1mm data for black-sky is being archived at this time. The data spatial extent is from approximately 5 degrees N to -30 degrees S latitude and 5 minutes E to 60 degrees E longitude and covers 7 sixteen day periods starting on July 11 through October 15, 2000.Map Products, containing spatially complete land surface albedo data, are generated at 1-minute resolution on an equal-angle grid. The maps are stored in separate HDF files for each wavelength, each 16-day period and each albedo type (white- and black-sky). Data belonging to black sky and white sky albedo have been zipped separately. This format allows the user to have flexibility to download and store only the data absolutely needed.The One-Minute Land Ecosystem Classification Product is a global (static map) data set of the International Geosphere-Biosphere Programme (IGBP) classification scheme stored on an equal-angle rectangular grid at 1-minute resolution. The dataset is generated from the official MODIS land ecosystem classification dataset, MOD12Q1 for year 2000, day 289 data (October 15, 2000). This dataset is used in generating the spatially complete albedo maps, but is also a stand-alone product designed for use by the user community. The Land Ecosystem Classification Map File product file is stored in Hierarchical Data Format (HDF).

The SAFARI 2000 project was selected as the first regional test for a prototype regional 500 m MODIS burned area product. The MODIS burned area product maps the 500 m location and approximate day of burning using a change detection algorithm based on a bi-directional reflectance model-based expectation method applied to the MODIS near-infrared and shortwave infrared bands (Roy et al., 2002). The algorithm was applied to recently reprocessed 500 m daily MODIS land surface reflectance data to produce burned area data sets for all of southern Africa for 2000 forward. This archived data set contains MODIS 500 m burned area products for two dry season months (July and September 2000).Burned area products are spatially explicit data sets that describe the approximate day of burning at 500 m resolution for all of southern Africa south of the Equator, including Madagascar. The burned area maps are compressed GeoTiff files. Several text files are included in the compressed files to aid ENVI (Research Systems, Inc.) users, including ENVI header files and ENVI density slice files. The data set also includes a projection parameters file.The MODIS burned area data set was validated using a methodology based upon the interpretation of multitemporal Landsat Enhanced Thematic Mapper plus (ETM+) data as described in Roy et al. (in press).

The MODIS (Moderate Resolution Imaging Spectroradiometer) Atmosphere Group develops remote sensing algorithms for deriving sets of atmospheric parameters from MODIS radiance data. These parameters can be integrated into conceptual and predictive global models. MODIS Atmosphere Products Subset Statistics (MAPSS) are generated over important locations around the world, as one of the ways to increase the scope of application of the MODIS atmospheric parameters. This MAPSS data set contains daily time series of the MODIS MOD05_L2 water vapor product over seventeen (17) AERONET sunphotometer measurement sites in southern Africa for the period February 24, 2000, through March 4, 2002. The process of generating the statistics involves identifying these locations on the MODIS MOD05_L2 product, extracting the values of the pixel corresponding to each coordinate point as well as surrounding pixels falling within a 50 x 50 km box centered on the coordinate point. The data product consists of column water-vapor amounts. During the daytime, a near-infrared algorithm is applied over clear land areas of the globe and above clouds over both land and ocean. Over clear ocean areas, water-vapor estimates are provided over the extended glint area. An infrared algorithm for deriving atmospheric profiles is also applied both day and night for Level 2. The data files are stored as ASCII tables in comma-separated-value (.csv) format. There is one file per site per year for each of the following two variables: total column precipitable water vapor (infrared retrieved) and total column precipitable water vapor (near-infrared retrieved).

The MODIS (Moderate Resolution Imaging Spectroradiometer) Atmosphere Group develops remote sensing algorithms for deriving sets of atmospheric parameters from MODIS radiance data. These parameters can be integrated into conceptual and predictive global models. MODIS Atmosphere Products Subset Statistics (MAPSS) are generated over important locations around the world, as one of the ways to increase the scope of application of the MODIS atmospheric parameters. This MAPSS data set contains daily time series of the MODIS MOD04_L2 aerosol product over seventeen (17) AERONET sunphotometer measurement sites in southern Africa for the period February 26, 2000, through December 31, 2001. The process of generating the statistics involves identifying these locations on the MODIS MOD04_L2 product, extracting the values of the pixel corresponding to each coordinate point as well as surrounding pixels falling within a 50 x 50 km box centered on the coordinate point. The data files are stored as ASCII tables in comma-separated-value (.csv) format. There is one file per site per year for each of the following variables: cloud fraction (land); cloud fraction (ocean); particle effective radius (ocean); optical depth (land and ocean); optical depth (land, corrected); optical depth (ocean, effective average); and optical depth ratio (small ocean).

Improved vegetation distribution and emission data for Africa south of the equator were developed for the Southern African Regional Science Initiative (SAFARI 2000) and combined with biogenic volatile organic compound (BVOC) emission measurements to estimate BVOC emissions for the southern African region. BVOC emissions were estimated for southern Africa on a monthly basis over a one-year period by combining GIS layers of vegetation, LAI, and climate with a biogenic emissions model, GLOBEIS (Guenther et al, 1993; Guenther, 1999).

The recently generated MODIS burned area product over southern Africa for the month of September 2000 was used to calculate regional biomass burning emissions from grassland and woodland fires for a number of trace gases and particulates at 1 km spatial resolution. A dynamic regional fuel load model developed for southern Africa in support of SAFARI 2000 fire emissions modeling is used to compute spatially explicit southern Africa fuel load data.

These data relate to a paper (Morisette et al., 2005) that describes the use of high spatial resolution ASTER data to determine the accuracy of the moderate resolution MODIS active fire product. Our main objective was to develop a methodology to use ASTER data for quantitative evaluation of the MODIS active fire product and to apply it to fires in Southern Africa during the 2001 burning season. We utilize 18 ASTER scenes distributed throughout Southern Africa covering the time period 5 August 2001 to 6 October 2001.

A new method is used to generate spatial estimates of monthly averaged biomass burned area and spatial and temporal estimates of trace gas and aerosol emissions from open fires in southern Africa. Global burned area data for the year 2000 (GBA2000) supplemented with the Along Track Scanning Radiometer (ATSR) fire count data are employed to quantify the area burned at 1-km resolution by using a fractional vegetation cover map derived from satellite observations.