Land cover and precipitation controls over long‐term trends in carbon gains in the grassland biome of South America

Marcos Texeira,Gervasio Pineiro,Jose Maria Paruelo,Mariano Oyarzabal,Santiago Baeza

doi:10.1890/es15-00085.1

Abstract

Carbon gains are a key aspect of ecosystem functioning since they represent the energy available for upper trophic levels. Carbon gains (or primary production) are strongly correlated with other ecosystem attributes such as secondary production and they are also the support for the provision of many ecosystem services. Given the documented dependency of primary production on precipitation, we expect that altered precipitation regimes, such as those projected by climate models, will have a significant impact on carbon gains. Land use and land cover changes are also expected to have a significant impact on the dynamics of carbon gains. We generated a spectral database of the fraction of photosynthetically active radiation intercepted by vegetation (fPAR), in order to study long‐term trends (i.e., decades) in carbon gains and its spatial and temporal relationships with precipitation and land cover patterns in Uruguay, which is part of the Rio de la Plata Grasslands, one of the largest temperate grasslands biome of the world. We found that carbon gains of native forests and grassland afforestation exhibited the strongest positive spatial response to precipitation, whereas crops and rangelands the weakest. In addition, we found that the temporal response of carbon gains to precipitation was strong and positive for all land uses. Although there were not clear trends in precipitation, we found strong negative trends in carbon gains through time, particularly in rangelands of the “Northern Campos” of Uruguay, where these trends represent a decrease between 10% and 25% of the annual aboveground net primary production. On the other hand, positive trends in carbon gains through time were associated to grassland afforestation and native forests. Therefore, during the period analyzed, land cover had a stronger influence on the observed trends in carbon gains than precipitation. These patterns emerged as a consequence of the interaction among precipitation, temperature, edaphic factors and management. Present trends in the controlling factors of C gains would exacerbate the observed patterns with serious consequences for the provision of ecosystems services.

Highlights

Global climate models predict an increased incidence of extreme weather events, such as droughts and floods and to a lesser extent, in the mean annual temperature and precipitation (Raisanen 2002, IPCC 2013)
NDVIMODIS–Long Term Data Record (LTDR) We found that 97.2% of the pixels (6796) showed percent differences in the range of (À5%, þ5%) of the general mean, whereas 97.7%
For all land cover classes considered, the statistical models including spatial autocorrelation were significant (Table 1), implying that some degree of spatial dependence was present in the relation between mean integrated fPAR (I-fPAR) and mean annual precipitation (MAP)

Summary

Introduction

Global climate models predict an increased incidence of extreme weather events, such as droughts and floods and to a lesser extent, in the mean annual temperature and precipitation (Raisanen 2002, IPCC 2013). Given the documented regional relationship between precipitation and primary production (Sala et al 1988, Lauenroth and Sala 1992, Paruelo et al 1999), we expect that grasslands and shrublands, from semi-arid to sub-humid, be among the most affected biomes (Hsu et al 2012) The effects of these projected changes on carbon gains are expected to be lagged as vegetation, in these systems, shows inertial dynamics associated to its dependence on water resources and soil characteristics (Oesterheld et al 2001, Wiegand et al 2004, Fabricante et al 2009, Sala et al 2012). The availability and quality of remote sensing data has increased dramatically over the past 20 years Other related techniques, such as the empirical up-scaling of eddy covariance measurements (Baldocchi 2008, Jung et al 2009) constitute a new and powerful alternative for estimating carbon, water and energy fluxes between the biosphere and the atmosphere. The absence of eddy covariance towers in some regions of the world (southern South America, our focal region, Middle East, India and central Africa) restricts its use, in these regions (see Baldocchi 2008 and http://fluxnet. ornl.gov/maps-graphics)

Objectives

Methods

Results

Conclusion