Regional cycles of agricultural land expansion and abandonment have been common throughout history in many countries of the world. Following the cessation of agricultural practice, landscapes undergo the spontaneous process of ecological succession resulting in significant above and belowground changes over time. As agricultural lands are often severely depleted of soil carbon, they represent one of the land types with the highest potential to act as carbon sinks through the process of soil carbon sequestration. While best management practices for increasing soil carbon stocks through sustainable agriculture are understandably a key focus point in climate change research today, the lasting effect of the abandonment of agriculture on soil organic carbon has received relatively less attention. However, significant amounts of farmland have been abandoned across the globe in both developed and developing countries, especially over the last several decades. To better understand the ability of old agricultural lands to act as carbon sinks through time, this study compiles field and published data to perform a comprehensive meta-analysis on the impacts of this land use change on soil organic carbon dynamics. Using a chronosequence approach, three study sites in Catalonia, Spain, each with four fields representing different stages of ecological succession post-abandonment spanning roughly 60 years, were sampled at soil depths of 10, 20, and 30 cm. To determine soil carbon stocks at each site, bulk density samples were also collected. Samples were analyzed for organic carbon, nitrogen and pH. Additionally, published chrononsequence and paired-plot data from abandoned agricultural lands throughout the Mediterranean region were also compiled into a database to perform multiple regression analysis. Our findings are not only meant to test the hypothesis that abandoned fields can act as carbon sinks over time, but to also determine the rate of soil carbon stock increase and projected vulnerability in relation to a variety of environmental and land management variables, thereby highlighting the climate change mitigation value of an as of yet understudied global land use change.

Increases in soil organic carbon (SOC) during secondary succession in Mediterranean and semi-arid climates, global hot-spots for agricultural land abandonment, have been notoriously difficult to predict and are subject to multiple environmental and land management factors. Field studies have reported positive, negative and no change varying over extended periods of time. To better evaluate the potential carbon sink capacity of regenerating semi-natural landscapes in these climates requires an improved understanding of the rates of SOC gains and losses. We compiled Mediterranean and semi-arid chronosequences and paired plots to investigate the effects of past land use, restoration intensity, and various environmental factors on SOC stocks during post-agricultural succession. Based on a preliminary synthesis of the western Mediterranean basin, we expect significant long-term accumulation rates globally although with high variability and the potential for net losses (compared to cropland control sites) even after several decades. Losses or minimal change are likely due to high initial SOC stock at the time of abandonment (e.g. from anthropogenic organic matter inputs) and too high or too low mean annual precipitation (e.g. < 450 or > 1000 mm), among other factors. A consolidated SOC accumulation rate for both Mediterranean and semi-arid soils undergoing post-agricultural succession is provided to better inform decision-makers on the benefits and challenges of agricultural land abandonment.