Project A01, Future Carbon Storage

Project Summary

This project views future-making in rural Africa through a carbon lens, focusing on two conflicting visions: wildlife conservation and agricultural intensification. During the CRC’s Phase I, the project used space-for-time substitutions and combined biophysical and socio-economic primary and secondary data at different scales to analyze effects of conservation and intensification on (1) carbon stock dynamics in soil and vegetation, as well as carbon-related ecosystem services, and (2) the composition of farm-household income, including detailed environmental sources. We were able to show at the scale of ecological observation plots, that carbon losses in the vegetation due to increased densities of large herbivores can be offset by carbon gains in soils. Surprisingly, carbon stocks in agricultural soils were not smaller than in soils under conservation. This phenomenon may be driven by two aspects of farmers’ future-making: a future-oriented soil fertility management, and a preferential selection of fertile land for agriculture. The implications for respective social-ecological couplings will be further explored in Phase II. At the regional scale, we found that Community-Based Natural Resource Management (CBNRM), albeit positively affecting the presence of large herbivores, also led to net losses in carbon-dense woodland cover in the region. Heterogeneous impacts of CBNRM are driven by tourism opportunities. In sub-regions with relevant wildlife presence, wildlife conservation has synergistic effects on the woodland cover, while in regions without opportunities for wildlife tourism, agriculture-dominated livelihood strategies have detrimental effects on vegetation cover and corresponding carbon storage.

Research Areas: Soil science, vegetation ecology, environmental Economics

Key Questions

At which temporal scales is carbon sequestered in soils and vegetation under different land-use?

Does conservation or restoration offset carbon loss under agricultural intensification?

What are key determinants of systemic coupling between outcome dimensions?

How do bio-physical and socio-economic factors interact in determining the cost-effectiveness of transformation pathways in terms of outcomes?


Our methodological approach to answer these questions relies on primary data from soil and vegetation samples as week as on household survey data and remote sensing analyses.

Key Findings

Unfortunately, current methods to quantify woody biomass and the carbon stored therein are not well-suited for ecosystems that are shaped by frequent disturbances such as elephant browsing. In Phase I, we thus developed a novel methodology to estimate woody biomass and carbon in disturbed dryland ecosystems. The methodological toolbox also comprises a detailed damage assessment, harnessing the ecological archive of trees for past disturbances. Results indicate that in highly disturbed African savannas, previous methods may underestimate woody biomass and the C (carbon) stored therein by up to 90 %. With the aid of this novel methodology, we were able to attribute elephant browsing in Namibia’s KAZA area to reduced C storage in woody biomass by 6.4 t C ha-1. However, the soil science team found that these C losses were almost compensated by increased C storage in soils (4.7 t C ha-1). Hence, rewilding with elephants seems to only have marginal effects on total carbon storage. With increasing elephant densities, though, community composition changed considerably, hinting at a non-linear relationship between conservation and biodiversity.

Intensification efforts are preferably focused on areas, which are relatively rich in soil organic matter and clay content, contain even more C than is lost with cultivation, thus raising novel questions about how these sites were selected by local farmers. At a regional scale, community-based natural resource management (CBNRM), the dominant conservation strategy in Namibia, reduced woodland cover by 2.1% between 1994 and 2009, corresponding to an annual change of -0.14% (Meyer et al., 2021). Heterogeneous treatment effect analysis indicated that CBNRM does work for woodland conservation when communities are in and around wildlife corridors, which provide tourism income opportunities. Inside these wildlife corridors, disturbance from wildlife may still exert negative effects on woody vegetation, but not to the extent of neutralizing the gains from conservation action.

Our joint research indicated that land-use decisions at household and community levels are the main drivers of change in aboveground and belowground carbon dynamics and related ecosystem services. However, the Spatio-temporal movement patterns of large herbivores are clearly co-determined by political decisions at national or regional levels, which significantly also affect the environmental outcomes. Rural wealth varies considerably within and between villages, and CRC228 household survey data collected in 2019 suggests a significant correlation with remotely sensed vegetation biomass.


Sandhage-Hofmann, A., J. Löffler, E. Kotzé, S. Weijers, V. Wingate, D. Wundram, L. Weihermüller, R. Pape, C.C du Preez, and W. Amelung, 2020. Woody encroachment and related soil properties in different tenure-based management systems of semiarid rangelands. Geoderma 372:114399. DOI.

Luedeling, E., J. Börner, W. Amelung, K. Schiffers, K. Shepherd, and T. Rosenstock, 2019. Forest restoration: Overlooked constraints. Science 366: 315-315.Link

Börner, J., Schulz, D., Wunder, S., Pfaff, A., 2020. The effectiveness of forest conservation policies and programs. Annual Review of Resource Economics 12:45-64. Link.

Gaitán, J., Maestre, F.T., Buono, G., Bran, D., Dougill, A.J., Garcia-Martinez, G., Ferrante, D., Guuroh, R., Linstädter, A., Massara, V., Thomas, A.D. & Oliva, G., 2019. Biotic and abiotic factors controlling soil organic carbon content have similar effects in regional and global drylands. Ecosystems 4, 1141-1151. Link.

Meyer, M., Klingelhoeffer, E., Naidoo, R., Wingate, V., Börner, J., 2021. Tourism opportunities drive woodland and wildlife conservation outcomes of community-based conservation in Namibia’s Zambezi Region. Ecological Economics 180: 106863. Link.

Munjonji, L., Ayisi, K.K., Mudongo, E.I., Mafeo, T.P., Behn, K., Mokoka, M.V., Linstädter, A., 2020. Disentangling drought and grazing effects on soil carbon stocks and CO2 fluxes in a semi-arid African savanna. Frontiers in Environmental Science 8: 590665. DOI.

Duarte-Guardia, S., Peri, P., Amelung, W., Thomas, E., Borchard, N., Baldi, G., Cowie, A. & Ladd, B. (2020): Biophysical and socioeconomic factors influencing soil carbon stocks: a global assessment. Mitigation and Adaptation Strategies for Global Change: 1-20. DOI.

Outlook for phase II funding

In Phase II, we will address three hypotheses, keeping carbon as the common currency within our project. Specifically, we aim to understand how (1) historical settlement processes have co-determined current land access and land-use patterns, as well as related rural wealth dynamics and variations in soil and vegetation quality. At the farm scale, we plan to study how (2) farmers actively shape their future by spatially modulating land management for improving soil and vegetation quality in vicinity to their farms. At regional scale and beyond, we will finally analyse (3) to what extent external shocks (including the COVID-19 pandemic) and spatio-temporal variations in policy regimes affect biophysical and socio-economic outcomes of interest. .

Partner Institutions