A preliminary assessment of the impact of landslide, earthflow, and gully erosion on soil carbon stocks in New Zealand

Abstract In geomorphically active landscapes such as New Zealand, quantitative data on the relationship between erosion and soil carbon (C) are needed to establish the effect of erosion on past soil C stocks and future stock changes. The soil C model currently used in New Zealand for soil C stock reporting does not account for erosion. This study developed an approach to characterise the effect of erosion suitable for soil C stock reporting and provides an initial assessment of the magnitude of the effect of erosion. A series of case studies were used to establish the local effect of landslide, earthflow, and gully erosion on soil C stocks and to compare field measurements of soil C stocks with model estimates. Multitemporal erosion mapping from orthophotographs was used to characterise erosion history, identify soil sampling plot locations, and allow soil C stocks to be calculated accounting for erosion. All eroded plots had lower soil C stocks than uneroded (by mass movement and gully erosion) plots sampled at the same sites. Landsliding reduces soil C stocks at plot and landscape scale, largely as a result of individual large storms. After about 70 years, soil C stocks were still well below the value measured for uneroded plots (by ~ 40% for scars and 20–30% for debris tails) indicating that the effect of erosion is very persistent. Earthflows have a small effect on estimates of baseline (1990) soil C stocks and reduce soil C stocks at landscape scale. Gullies have local influence on soil C stocks but because they cover a small proportion of the landscape have little influence at landscape scale. At many of the sites, the soil C model overestimates landscape-scale soil C stocks.