Using AVHRR lunar observations for NDVI long-term climate change detection

Abstract

The Moon is believed to be an irrefutably stable calibration reference target for studying climate change from satellites, as demonstrated in several lunar calibration studies of satellite radiometers. However, the potential of using advanced very high resolution radiometer (AVHRR) lunar observations for climate change detection has not been recognized in the past. At the same time, it is known that despite past efforts, there are still significant uncertainties in determining the long-term trend of climate change using such indices as the Normalized Difference Vegetation Index (NDVI), partly because most AVHRR calibrations to date lack the stability required for climate change detection. This study demonstrates a novel method using the AVHRR lunar band ratio as a stability reference for long-term NDVI change detection. It shows that despite the incomplete lunar observations and complex space view mechanisms, the AVHRR lunar observations can be used to ensure the stability of the Earth observations at a better than ±1% (1 sigma) level for the lifetime of the AVHRR instrument in the long-term detection of NDVI trends. This method is compared with the Vennote and Kaufman high-altitude bright cloud method used by the NASA Long-Term Data Record (LTDR) project, and preliminary results are very encouraging. The lunar band ratio method has been tested for NOAA 11,14, and 16 AVHRR and is applicable to all AVHRRs since the early 1980s. It is conceived that the Lunar Band Ratio serves as an important stability reference for ensuring the long-term confidence for the detection of climate change. The AVHRR lunar band ratio is also useful for deriving the calibration for the problematic band 2 from that of band 1 for establishing a consistent AVHRR fundamental climate data record for a variety of climate applications. Copyright 2009 by the American Geophysical Union.