Environmental controls on the spatial variability of savanna productivity in the Northern Territory, Australia

Gross Primary Productivity (GPP) is a critical measure of the health and sustainability of natural ecosystems. Understanding the magnitude, spatial patterns and processes of GPP will underpin predictions of the impact of climate change on the carbon cycle. In Australia, savannas account for one third of the terrestrial carbon stores and therefore, estimating the magnitude of savanna GPP and studying the spatial relationship between GPP and environmental determinants at the regional scale is essential in understanding ecosystem responses to increasing atmospheric CO2 concentrations and climate change. In this study we employed an integrated approach combining in situ measurements, eddy covariance based flux tower data and remote sensing techniques to examine the role of environmental drivers in controlling the spatial variation in GPP of savannas in the Northern Territory (NT), Australia. We used field based light use efficiency (LUE), regional specific meteorology and Moderate Resolution Imaging Spectro-radiometer (MODIS) based fraction of absorbed Photosynthetically Active Radiation (fPAR) data to estimate GPP. The estimated GPP agreed quite well (only a 6% error) with GPP estimated from flux tower at the Howard Springs site. The spatial pattern of GPP along the Northern Australian Tropical Transect (NATT) was calculated and showed a strong gradient in GPP from the coast (12.50°S where rainfall was 1622 mm year−1) to inland (17.73°S where mean rainfall was 643 mm year−1) with a decrease of 77%. A decreasing trend in GPP with rainfall is noticed especially at the dry end of the transect studied. However, in the wet end and middle part of the transect (e.g. dominated by different Eucalyptus species), the response of GPP to changes in rainfall is reduced. This finding suggests that the influence of rainfall on various Eucalyptus species may be dampened by biotic factors. Our results suggest that future changes in precipitation driven by climate change may affect the future distribution and dynamics of GPP in northern Australia.

• In situ measurements, eddy covariance based flux tower data and remote sensing techniques were used to model gross primary productivity (GPP) of savannas in the Northern Territory, Australia.
• The spatial pattern of GPP along the Northern Australian Tropical Transect showed a strong gradient in GPP from the coast (annual rainfall of 1622 mm) to inland (rainfall of 643 mm year−1) with a decrease of 77%.
• A decreasing trend in GPP with rainfall is noticed at the dry end of the transect studied, but in the wet end and middle part of the transect the response of GPP to changes in rainfall is reduced.
• This pattern suggests that the influence of rainfall on various Eucalyptus species may be dampened by biotic factors.
• Any future changes in precipitation may affect the future distribution and dynamics of GPP in northern Australia.