Aridity Thresholds Determine the Relationships Between Ecosystem Functioning and Remotely Sensed Indicators Across Patagonia

Año Publicación:  2021
Responsable: Y. Zhao et al.
Journal, Volumen y páginas:
IEEE Transactions on Geoscience and Remote Sensing PP(99):1-9


Y. Zhao, E. Guirado, J. J. Gaitán & F. T. Maestre


Emerging evidence suggests that ecosystem responses to increases in atmospheric aridity, a hallmark of climate change, exhibit multiple thresholds across global drylands. However, it is not clear whether aridity thresholds exist in the relationships between ecosystem functions and remotely sensed indicators (RSIs). Assessing this is important because these empirical relationships underpin the statistical models commonly used to estimate ecosystem functioning across large spatial scales, which typically uses data from RSI. We evaluated how the relationships between nutrient cycling index (NCI; a proxy of ecosystem functioning) measured in situ and RSI [albedo and normalized difference vegetation index (NDVI)] change along with a wide aridity (1 - [precipitation/potential evapotranspiration]) gradient in Patagonia (Argentina). For doing so, we used field-based NCI data from 235 ecosystems that were surveyed twice (2008-2013 and 2014-2018). Three aridity thresholds were identified when evaluating the RSI-NCI relationships. The first threshold was found around aridity values ranging from 0.44 to 0.60, while the second and third were concentrated around aridity values of 0.69 and 0.82, respectively. These results indicate that RSI-NCI relationships changed drastically along aridity gradients, and these thresholds should be considered when evaluating ecosystem functions using RSI. In addition, we also found that the relationships between NCI and albedos were not significant around aridity values of 0.82. These results were consistent across sampling dates. Our findings imply that empirical models of the RSI-NCI relationship employing only albedos/reflectance as inputs are not reliable under the most arid conditions and can be used to improve the effectiveness of the use of RSI to monitor and predict changes in ecosystem functioning across large environmental gradients in drylands.

crossmenu linkedin facebook pinterest youtube rss twitter instagram facebook-blank rss-blank linkedin-blank pinterest youtube twitter instagram