Estimating net primary productivity over the Alaskan landscape using AVHRR-derived NDVI data
Dates
Year
2001
Citation
Markon, Carl Joseph, 2001, Estimating net primary productivity over the Alaskan landscape using AVHRR-derived NDVI data: University of Alaska Anchorage.
Summary
Plant productivity is a fundamental ecological variable that provides information about the health and status of vegetation communities, but is difficult to obtain. In this study, the normalized difference vegetation index, or NDVI, derived from AVHRR data, was used to estimate NPP. Seven seasonal-based metrics were calculated using the NDVI, and utilized to model NPP over Alaska, U.S.A. They included maximum, mean and summed NDVI for the growing season, total days for the growing season, product of total days and maximum NDVI, an integral estimate of NDVI over the growing season, and a summed product of NDVI and solar radiation for the growing season. Linear relationships between NPP and each NDVI metric were analyzed at four scales: [...]
Summary
Plant productivity is a fundamental ecological variable that provides information about the health and status of vegetation communities, but is difficult to obtain. In this study, the normalized difference vegetation index, or NDVI, derived from AVHRR data, was used to estimate NPP. Seven seasonal-based metrics were calculated using the NDVI, and utilized to model NPP over Alaska, U.S.A. They included maximum, mean and summed NDVI for the growing season, total days for the growing season, product of total days and maximum NDVI, an integral estimate of NDVI over the growing season, and a summed product of NDVI and solar radiation for the growing season. Linear relationships between NPP and each NDVI metric were analyzed at four scales: plot, 1-km, 10-km, and 20-km pixels. Results show moderate to poor relationship between any of the metrics and NPP estimates for all data sets and scales. Use of NDVI for estimating NPP may be possible, but caution is required due to data seasonality, the scaling process used, and land surface heterogeneity.