Modelling soil carbon sequestration capacity in Mediterranean soils. PhD Thesis
During the last decades, land use changes have largely affected the global warming process through emissions of CO2. However, C sequestration in terrestrial ecosystems could contribute to the decrease of atmospheric CO2 rates in the short- or medium-term. Under the Kyoto Protocol (UNFCCC, 1997), national governments are required to assess and report national atmospheric C emissions and removals reflected as changes in C pools. Accordingly, regional studies for assessing C stocks are needed. It is essential to predict soil organic C (SOC) stocks in future climate scenarios to establish adequate land use and management strategies. Models are effective tools for assessing SOC stocks and dynamics at different scales and predict C sequestration trends under projected scenarios. Soil C models are increasingly being used as decision support tools, in particular on issues related to land use or climate change. Although Mediterranean areas show a high potential for C sequestration, only a few studies have been carried out in Mediterranean systems. Several studies on soil C models in combination with climate change scenarios have been developed but new tools are needed to improve soil organic C stocks predictions.
After a general introduction (Chapter 1) Chapter 2 improves and tests methodologies to assess land cover change (LCC) dynamics between 1956 and 2007 in Andalusia (Southern Spain) and temporal and spatial variability of C stored in vegetation at a wide scale. LCCs are assessed by comparison of spatial data from 1956 and 2007 and are reclassified following land cover flows reported in major areas in Europe. Southern Spain has supported important changes during the studied period with significant consequences for vegetation C stocks, mainly due to afforestation and intensification of agriculture, resulting in a total vegetation C stock of 156.08 Tg in 2007, with an increase of 17.24 Tg since 1956. Likewise, LCCs in the 51-year period (1956-2007) have largely affected C stored in soils in Southern Spain.
In Chapter 3 a methodology is proposed to assess the impact of land use and land cover change (LULCC) dynamics on SOC contents at different depths. Soil databases and spatial datasets with soil and land use information are used to estimate SOC stocks. Additionally, SOC sequestration rates are provided for different LCCs and soil types in Andalusia. A total of 16.8 Tg of SOC has been lost in the last years (approximately 0.33 Tg year-1) and largest decreases were observed in soils types as Fluvisols and Arenosols and land use types as coastal wetlands. On the other hand, forests contributed to sequestration of 8.62 Mg C ha-1 (with a sequestration rate of 25.4%).
Chapter 4 focuses on assessing current SOC contents and identifying environmental factors which determine fluctuations and intensity of SOC dynamics. Soil and climate databases, digital elevation models and land use and soil maps were used to evaluate SOC pools and their distribution within the soil profile. The total organic C stock in 2007 in soils of Andalusia is 415 Tg for the upper 75 cm, with up to 55% stored in the top 25 cm of soil (229.7 Tg). Among all soil types, Calcisols and Vertisols show the highest values with above 65 Mg C ha-1 (0-75 cm). Significant correlations have been found between soil organic C and some environmental factors in natural areas, such as average summer and winter temperatures, annual mean precipitation and elevation.
These previous studies in Chapters 2, 3 and 4, comprise the first comprehensive analysis of the impacts of land use changes on terrestrial C stocks at a regional scale in Andalusia (S Spain). Based on this preliminary research, a soil carbon model (CarboSOIL) has been developed to predict SOC stocks at different soil depths for a range of soil management, land use and climate change scenarios (Chapter 5).
Several methodologies have been tested to design the new tool CarboSOIL and better predictions have been obtained with Multiple Linear Regression techniques and Box-Cox transformation procedures. The model has been trained in Andalusia and tested in Valencia, and is divided in four submodels (CarboSOIL25, CarboSOIL50, CarboSOIL75 and CarboSOIL TOTAL) according to different soil depths (0-25 cm, 25-50 cm, 50-75 cm and 0-75 cm). CarboSOIL model has been developed as a computer application to be implemented in the agroecological decision support system MicroLEIS, and each submodel has been built as a spatial tool in a GIS environment for spatial analysis of the inputs/outputs of the model.
In Chapter 6, CarboSOIL has been tested and validated in Andalusia in the baseline scenario and applied in different IPCC scenarios (A1B, A2 and B1) according to different Global Climate Models (BCCR-BCM2, CNRMCM3 and ECHAM5). Output data were linked to spatial datasets (soil and land use) and spatial analysis were performed to quantify SOC stocks in different soil types under a range of land uses.
Although there is an overall trend in all soil types towards decreasing of SOC stocks in the upper soil sections (0-25 cm and 25-50 cm), predicted SOC stocks tend to increase in the deeper soil section (50-75 cm). CarboSOIL model proved its ability to predict the short, medium and long-term trends (2040, 2070 and 2100) of SOC dynamics and sequestration under projected future scenarios of climate change.