Although grazing is the primary land use in the savanna lowland of southern Kenya, the effects of grazing on soil carbon dioxide flux (R_S) remain unclear. A 12-month study was conducted from January to December 2020 on the effects of six grazing intensities sites (overgrazed (OG), heavily grazed (HG), moderately grazed (MG), moderately to lightly grazed (M-LG), lightly grazed (LG) and no grazing (NG)) on R_S on. A camera trap was used to monitor the total number of animals at each site, indicating the grazing intensity. Weekly measurements of R_S were taken using static greenhouse gas chambers along with simultaneous measurements of soil temperature (T_S) and volumetric soil water content (W_S) (depth of 5 cm). Mean R_S at HG, MG, M-LG and LG sites was approximately 15–25% higher than at NG and OG sites (p < 0.001). Mean W_S increased with decrease in grazing especially in the dry season, while T_S increased with increase in grazing. We observed bimodal temporal variation in R_S and W_S due to two wet seasons in the year. Thus, variation in R_S across the study period followed the changes in W_S rather than those in T_S. Mean values of R_S in the wet seasons were significantly higher (> 45%) than those in the dry seasons, and W_S accounted for 71% of the temporal variability in R_S (p < 0.05)_. In addition, the enhanced vegetation index (EVI, interpreted as a proxy for vegetation cover) explained 60% of the variance of R_S, and W_S and EVI together explained 75%. EVI showed a negative relationship (p < 0.05) with animal intensity, indicating that more grazing reduced vegetation cover and, consequently, soil organic carbon and biomass. Soil bulk density was lower at less grazed sites. While R_S variability was unaffected by total nitrogen content, pH, and texture, correspondence analysis demonstrated that the main factors influencing R_S dynamics across the year under different grazing intensities were W_S and vegetation cover. Our results contribute to closing the existing knowledge gap regarding the effects of grazing intensity on R_S in East Africa savannas. Therefore, this information is of great importance in understanding carbon cycling in savanna grassland, as well as the identification of the potential consequences of increasing land pressure caused by rising livestock numbers, and will assist in the development of climate-smart livestock management in East Africa.

虽然放牧是在肯尼亚南部的稀树草原低地主土地使用，放牧对土壤二氧化碳通量（的效果- [R_小号）仍不清楚。2020 年 1 月至 12 月对六个放牧强度地点（过度放牧 (OG)、重度放牧 (HG)、中度放牧 (MG)、中度至轻度放牧 (M-LG)、轻度放牧）的影响进行了为期 12 个月的研究(LG) 并且没有放牧 (NG)) 在R _S上。使用相机陷阱监测每个地点的动物总数，表明放牧强度。使用静态温室气体室每周测量R _S并同时测量土壤温度 ( T _S）和体积土壤含水量（w ^_小号）的5厘米（深）。平均数ř_小号在HG，MG，M-LG和LG位点比在NG和OG位点（高约15-25％p  <0.001）。平均W _S随放牧的减少而增加，特别是在旱季，而T _S随放牧的增加而增加。由于一年中有两个雨季，我们观察到R _S和W _{S 的}双峰时间变化。因此，整个研究期间R _S的变化跟随W _S而不是T _{S 的变化}. 雨季的R _S平均值显着高于 旱季（> 45%），并且W _S占R _S时间变异的 71% （p < 0.05）_。此外，增强的植被指数（EVI，解释为植被覆盖的代理）解释了 60% 的R _S方差，而W _S和 EVI 一起解释了 75%。EVI 与动物强度呈负相关 (p < 0.05)，表明更多放牧减少了植被覆盖，从而减少了土壤有机碳和生物量。在较少放牧的地方土壤容重较低。虽然R _S变异性不受总氮含量、pH 值和质地的影响，对应分析表明，在不同放牧强度下，影响全年R _S动态的主要因素是W _S和植被覆盖。我们的结果有助于缩小关于放牧强度对东非稀树草原R _S影响的现有知识差距。因此，这些信息对于了解稀树草原的碳循环以及确定牲畜数量增加造成土地压力增加的潜在后果非常重要，并将有助于东非气候智能型牲畜管理的发展.