Highlights
• Long-term livestock exclusion did not improve soil carbon in dry savanna ecosystem.
• The influence of exclosure age on soil carbon was not linear.
• Soil carbon sequestration is not well understood in drylands of Africa.
• Baseline information is key to precisely evaluate exclosures.
•Target multiple benefits than carbon sequestration per se

Controlled grazing management is considered as an effective strategy to enhance soil carbon sequestration, but empirical evidences are scarce. Particularly, the role of livestock exclusion related to soil carbon sequestration is not well understood in arid and semiarid savannas of Africa. We investigated the effectiveness of long-term (14–36 years old) exclosures in enhancing soil carbon in the semiarid savanna, southern Ethiopia. We tested for differences in soil carbon content between exclosures and adjacent open-grazed rangelands, while accounting for effects of age of exclosures and soil depths. We collected soil samples at two soil depths (0–20 cm and 20–50 cm depths) from 96 plots from 12 exclosure and adjacent open grazing sites. We found no significant differences (P > 0.05) between exclosures and adjacent open-grazed rangelands in soil carbon content in both soil depths. The age chronosequence further suggested a weak non-linear trend in increasing soil carbon content with increasing duration of exclosures. These results thus challenge the opinion that controlled grazing enhances soil carbon sequestration in semiarid savannas. However, we remain cautious in regard to the conclusiveness of these findings given the paucity of information regarding other confounding factors which may disentangle the effects of the exclosure, and most importantly in the absence of soil data prior to exclosures.