Abstract Climate change will likely trigger shifts in rainfall regimes that may intensify water scarcity in semi-arid regions. In the semi-arid region of Brazil, the seasonally dry ecosystem is the primary source of forage for livestock. Because the correct stocking rate of livestock (animals per area) is not well understood, overgrazing tends to advance rangeland degradation in this ecosystem. This implies that the region may become even more vulnerable under changing rainfall regimes, which in turn may exacerbate livestock and food insecurity. We developed a coupled soil water balance, vegetation, and cattle biomass model to illustrate the impacts of rainfall seasonality on the dynamics of vegetation and animal growth. The outcomes were simulated by considering different stocking rates and the timing of animal placement and removal from the rangeland. A more pronounced reduction in vegetation biomass was found in grazed vs. non-grazed paddocks. Under strongly seasonal rainfall patterns, the maximum animal weight gain decreases with average rainfall inter-arrival time and increases with total annual rainfall. Thus, a forecast of dry spells could benefit farmers in planning grazing management strategies. Our model can be used to test different management scenarios and give feedback for local herders, and to guide future experiments to reduce the time and cost of acquiring data.

中文翻译：

---

降雨波动下季节性干燥热带森林生态系统中牛放牧的优化管理

摘要 气候变化可能会引发降雨状况的变化，这可能会加剧半干旱地区的水资源短缺。在巴西半干旱地区，季节性干旱的生态系统是牲畜饲料的主要来源。由于对牲畜的正确放养率（每面积的动物）尚不清楚，过度放牧往往会加剧该生态系统中的牧场退化。这意味着该地区在不断变化的降雨情况下可能变得更加脆弱，这反过来又可能加剧牲畜和粮食不安全。我们开发了一个耦合的土壤水平衡、植被和牛生物量模型，以说明降雨季节性对植被和动物生长动态的影响。通过考虑不同的放养率以及动物放置和移出牧场的时间来模拟结果。在放牧和非放牧的围场中发现植被生物量的减少更为明显。在强季节性降雨模式下，最大动物体重增加随着平均降雨间隔时间而减少，并随着年总降雨量增加而增加。因此，干旱期的预测可以使农民在规划放牧管理策略时受益。我们的模型可用于测试不同的管理场景并为当地牧民提供反馈，并指导未来的实验以减少获取数据的时间和成本。最大动物增重随着平均降雨间隔时间的增加而降低，并随着年降雨总量的增加而增加。因此，干旱期的预测可以使农民在规划放牧管理策略时受益。我们的模型可用于测试不同的管理场景并为当地牧民提供反馈，并指导未来的实验以减少获取数据的时间和成本。最大动物增重随着平均降雨间隔时间的增加而降低，并随着年降雨总量的增加而增加。因此，干旱期的预测可以使农民在规划放牧管理策略时受益。我们的模型可用于测试不同的管理场景并为当地牧民提供反馈，并指导未来的实验以减少获取数据的时间和成本。