Change in Soil Organic Carbon Stocks under 12 Climate Change Projections over New South Wales, Australia

Abstract

Digital soil mapping (DsM) techniques combined with space-for-time substitution (sFTs) processes were used to map and examine soil organic carbon (sOC) changes caused by projected climate change over New south Wales, Australia until ~2070. Twelve projections were derived from four global climate models downscaled with three regional climate models. A marked variation in the direction and magnitude of sOC change was demonstrated with the different projections. Mean state-wide predictions (0-30 cm depth) ranged between 2.9 Mg ha -1 gain and 8.7 Mg ha -1 sOC loss. Greater consistency among climate change projections is required before we can confidently predict sOC changes. By using averaged results from the 12 projections, broad trends were revealed for the change in sOC over two intervals (0-30 and 30-100 cm). A mean loss rate of 2.0 Mg ha -1 for the upper interval was demonstrated and a total loss of 737 Tg of CO 2 equivalent for the entire depth to 100 cm but the 95% confidence interval was wide. Although changes are primarily controlled by the balance between changing temperatures and rainfall, the extent of change also depends on the environmental regime, with differing changes demonstrated over 36 current climate-parent material-land use combinations (e.g., projected mean sOC decline is <1 Mg ha -1 for dryhighly siliceous-cropping but >15 Mg ha -1 for wet-mafic-native vegetation C limate change will impact on many aspects of the global environment and civilization generally, with changes in our soil resources being one important yet not widely understood consequence. Changes in key soil properties such as SOC content will influence agricultural productivity and our ability to feed and support the growing world population. Native ecosystems will be impacted, with modifications to species distribution and abundance at local, regional, and national scales. Changes in the potential of soils to sequester C or release it to the atmosphere are crucial for climate change modeling and mitigation strategies (Lal et al., 2007;Baldock et al., 2012).Although there has been widespread work on the relationship of climate to soils, dating back to the pioneering work of Dokuchaev in 1899 (Dokuchaev, 1967 transliteration) andJenny (1941) Core Ideas• Potential changes in soil organic C to 2070 mapped (100-m grid) and examined.• The direction and magnitude of change varied between the 12 climate projections.• Differing changes revealed for 36 current climate-parent material-land use regimes.• Digital soil mapping-space-for-time substitution is useful for climate change study

Projecting Soil Organic Carbon Distribution in Central Chile under Future Climate Scenarios

“…These results suggest a greater importance of the Andisols in subtropical climates to the C cycle, especially under climate change; however, the relationship between climate change impacts on Andisols and its SOC sequestration potential still needs to be researched. Equally remarkable is the pattern of increasing SOC from north to south, which also follows the spatial patterns of increasing rainfall and decreasing temperature (Viscarra Rossel et al, 2015; Gray and Bishop, 2016). Those results also support the hypothesis of Minasny et al (2013) at the global scale, and de Brogniez et al (2015) in Europe.…”

“…3) but also with higher amounts of losses. This pattern is similar to the work of Stockmann et al (2015), Gray and Bishop (2016), and Meersmans et al (2016), who found the projections in SOC changes to be dependent on initial SOC levels and concluded that the soils that will lose more SOC are those that initially have more SOC.…”

“…The methodology used by Gray and Bishop (2016) demonstrated that uncertainty analysis of climate change maps cannot be done due to the lack of validation points at future conditions, but to permit comparison between results, the same 50 bootstraps were used in current SOC for the simulations at each climate change condition.…”

“…Thus, it is possible to quantitatively simulate soil attributes with variations of those variables. As Gray and Bishop (2016) affirmed, for a given variable, we may assume projected changes in time based on predicted changes in space, keeping all the other variables constant. The aims of our study were (i) to predict SOC concentration under current climatic conditions using random forest modeling; (ii) to produce simulations under two scenarios of total radiative forcing, as defined in the Fifth Assessment Report of IPCC (2013) for the year 2050 and 2080; and (iii) to compare the current SOC with the predicted climate change simulations.…”

Pedology and digital soil mapping (DSM)

“…Compared with a fully dynamic simulation model, this approach has limitations, such as lack of feedback and possible extrapolation problems. However, it is a quick and relatively simple way of estimating soil changes, and it has attracted considerable application in projecting the fate of soil carbon under future climate change scenarios, with examples from Gray and Bishop (2016) and Yigini & Panagos (2016).…”