Conventional techniques for analysing soil organic carbon involve analysing bulk density cores, a process that can be tedious, time consuming and sensitive to disturbances. Tor-Gunnar Vågen from the World Agroforestry Centre and Leigh Winowiecki from the International Centre for Tropical Agriculture, both in the Kenyan capital Nairobi, have developed a cumulative soil mass technique that overcomes these issues.

The cumulative mass methodology involves using an auger to collect soil at various depths followed by laboratory analysis to find the total oven-dried weight of soil for each depth increment for a given volume. These data are then used to model bulk density and/or soil mass at any given depth.

“This method is simple and allows for repeatable and robust measurements of soil carbon stocks in different soil types and under different land cover and land use systems,” Vågen told environmentalreseachweb. It has also enabled Vågen and Winowiecki to map soil organic carbon stocks to 30 cm depth using Landsat ETM+ reflectance across four study sites in East Africa that cover a wide range of climate and soil conditions.

The researchers collected data for 640 soil profiles from a total of four sites, each covering an area of 100 km2 (10 × 10 km) from three east African countries. “We simply would not have been able to do this using conventional methods,” said Vågen. “It would have been too expensive.”

The pair used these data to train models to map soil organic carbon stocks. They found that soil erosion affects soil organic carbon stocks strongly in all sites and across different land-cover types. They also found that inherent soil properties such as sand content are often more important than climate effects alone in determining soil organic carbon dynamics.

“Land management practices that lead to the control and reduction of soil erosion will be important mitigation strategies in these landscapes,” said Vågen. “In the dryland sites, managing woody cover is not enough to reduce erosion prevalence and mitigation strategies also need to focus on improving the condition of the herbaceous cover. This calls for the implementation of improved grazing and better livestock management in these ecosystems.”

Vågen and Winowiecki hope to make their maps and models available to other researchers early next year, through a web-based platform that will enable researchers to choose an area of interest and run predictions for soil organic carbon stocks.

The scientists published their work in Environmental Research Letters (ERL).