Australian monthly fPAR derived from Advanced Very High Resolution Radiometer reflectances - version 5

These data describe the Australia-wide, monthly fraction of Photosynthetically Active Radiation absorbed by vegetation (fPAR) derived from Advance Very High Resolution Radiometer data spanning July 1981 to Oct 2011. fPAR is linearly related to fractional foliage cover. Here fPAR is split into that of persistent vegetation and of recurrent vegetation, which represent non-deciduous perennial vegetation and annual, ephemeral and deciduous vegetation, respectively. Data have been processed using the 'invariant cover triangle' method to remove the majority of errors introduced by sensor calibration and change-over effects. Data are pre-eminently suited to the analysis of long-term dynamics (trends) in vegetation cover. Data are monthly grids of 0.08 degree cell size from July 1981 onwards and of 0.01 degree from Feb 1995 onwards. Version 5 differs from the previous version in that it is current to Oct 2011 (note that values for previously existing dates may differ slightly between versions). Further details are provided in the readme file.\r\n\r\nFormat Name : ENVI byte (8-bit unsigned integer) binary file \r\nFormat Version: \r\n\r\nDataset Creation Date: 2008\nLineage: Data are derived from NOAA AVHRR Red and Near Infra-Red reflectances. Reflectances were calibrated using the invariant-cover-triangle methodology (Donohue et al., 2008). This method makes relative calibrations for both atmospheric interferences and sensor calibration errors to produce data suitable for long-term vegetation dynamics analyses. Refer to readme file for details. \n\nThese data have been corrected specifically for use in looking at long-term dynamics. The majority of sensor calibration and sensor change-over errors have been removed as have most Rayleigh (molecular) and aerosol scattering. Water vapour absorption and viewing angle (i.e., BRDF) effects have not been removed from the data. Some months in the PAL data series have had data from portions of the country masked out due to poor data quality. These months and the approximate regions covered by masks are indicated in the list at the end of the readme file (see Donohue et al., 2007 for full details). No assessment of accuracy has been performed on this data although the greatest possible precision of AVHRR-derived NDVI/fPAR data is around 0.01-0.02. Spatial accuracy of the HRPT LAC data is to within half a pixel. \n\nCredit:\nThis dataset has been developed by the Time Series Remote Sensing Team, CSIRO Land and Water in conjunction with the Environmental Biology Group at the Australian National University. The original data were supplied by CSIRO Marine and Atmospheric Research. This work has been supported by the eWater CRC.