Abstract:

Global climate change has increasingly affected terrestrial ecosystems. The occurrence of extreme events such as high temperatures and droughts poses challenges to ecosystem services and biodiversity. Grassland ecosystems store approximately one-third of the world’s terrestrial carbon reserves and play pivotal roles in the global carbon cycle. The net primary productivity (NPP) of grasslands serves as an essential indicator of the adaptability and responsiveness of a system to climate and soil change, offering valuable insights for improving the precision of global carbon cycle models. However, the mechanisms by which grassland ecosystems respond to environmental changes across different climate zones and humidity conditions remain complex and unclear, posing challenges for effective management and conservation of grasslands. In this study, a comprehensive dataset was constructed using the observed NPP data of grasslands worldwide, covering climate factors, such as temperature, radiation, and precipitation, as well as soil factors, such as soil water content, bulk density, and soil particle composition. Through regression analysis and the random forest algorithm, the responses of grassland NPP to environmental factors in different climatic zones are systematically discussed. Regression analysis helped to identify the quantitative effects of climate and soil factors on grassland NPP, whereas random forest algorithms revealed the relative importance and complex nonlinear relationships of each factor. In addition, an asymmetry index was used to analyze the response of grassland NPP to precipitation change, revealing the adaptability of grassland systems to precipitation variability under different climatic conditions and further improving the understanding of the grassland ecosystem response model. In arid regions, grassland NPP is primarily limited by water availability, with minimal influences from other climatic factors and soil properties. These findings provide valuable guidance for grassland management in arid areas, emphasizing the importance of water resource management for maintaining and boosting the productivity and resilience of grassland ecosystems in these regions. Conversely, the response of grassland NPP to environmental factors in humid climatic zones is complex. Grassland NPP was significantly influenced by energy factors such as solar radiation and temperature. Soil factors such as soil water content and bulk density also play crucial roles. This suggests that, under relatively adequate water conditions, grassland ecosystems in humid zones exhibit a more significant range of response mechanisms to environmental fluctuations. Grasslands in these areas must adapt not only to variations in climate factors but also to changes in soil conditions to maintain ecological balance and productivity. Therefore, grassland management in humid areas should adopt an integrated approach that considers the interaction between climate and soil conditions to enhance the resilience and sustainability of grassland systems. In terms of precipitation response asymmetry, this study revealed distinct response patterns for grassland NPP across different climate zones. In arid regions, grassland NPP exhibits a pronounced positive response to precipitation under conditions of high solar radiation, low vapor pressure deficit (VPD), and high soil sand content. This pattern indicates that grassland ecosystems in arid regions develop high efficiency in utilizing sporadic precipitation events, allowing for rapid increases in productivity under limited water input conditions. This phenomenon highlights the adaptability of grassland ecosystems in arid zones to effectively utilize short-term precipitation, thereby enhancing their survival capacity in response to the dual stresses of drought and high temperatures. Additionally, in humid climate zones, grassland NPP showed a more marked positive response to precipitation under moderate levels of precipitation and temperature as well as in soils with lower clay content. This finding reveals that, in humid regions, grassland productivity peaks under moderate precipitation levels, whereas excessive rainfall may lead to decreased soil aeration, thereby inhibiting root respiration and ultimately reducing productivity. This pattern indicates the need for appropriate management of soil moisture in humid areas to avoid the negative impacts of excessive precipitation and to maintain ecological health and grassland productivity. Finally, we identified a “threshold effect” on the response of grassland NPP to environmental factors. Grassland NPP demonstrated a positive response to changes in environmental variables up to a certain threshold, beyond which the response may weaken or even turn negative. This threshold effect was particularly pronounced under extreme climatic conditions, indicating the vulnerability of grassland ecosystems to such extremes. For example, in humid regions, when precipitation levels exceed an optimal threshold, a decline in soil aeration can lead to a reduction in the grassland NPP. These results suggest that management practices should focus on maintaining environmental variables within optimal thresholds for grassland ecosystems to enhance their resilience to climate extremes and preserve their productivity and stability. Future research could expand this study by examining the specific response mechanisms of different vegetation types, such as grasslands and shrublands, to climate change and by analyzing the long-term dynamic changes in ecosystems under shifting climate conditions. Exploring the role of microbial communities in grassland ecosystems, particularly their functions in carbon and nitrogen cycling, could offer a more comprehensive understanding of grassland diversity and stability. Incorporating remote sensing technology, ecological modeling, and artificial intelligence algorithms can further enhance the accuracy of predictive models for climate adaptation, laying the scientific foundation for tackling the complexities of future climate change. In conclusion, this study systematically uncovered the intricate mechanisms by which grassland ecosystems respond to climate and soil factors across diverse climate zones, by utilizing a comprehensive global dataset of grassland NPP. These findings provide essential data to enhance global carbon cycle models and lay a solid theoretical foundation for future grassland management and climate adaptation strategies. Beyond providing actionable recommendations for grassland management and ecological restoration, this study provides scientific insights crucial for maintaining grassland ecosystem resilience in the face of intensifying climate challenges. Finally, this study has far-reaching implications for understanding the global response of ecosystems to climate change, reinforcing the urgent need for informed and sustainable ecological practices.

Key words: grassland ecosystem, net primary productivity, spatial distribution, climatic factors, asymmetric response

摘要：

草地生态系统对全球气候变化的反馈具有重要的生物指示作用，然而，不同水分条件下草地对气候和土壤因子的响应机制尚不明确。基于全球实测草地NPP数据，补充了气候和土壤因子，经过整合分析形成数据集。采用回归分析和随机森林算法探究了不同气候区草地NPP对环境因子的响应，以及利用非对称指数分析了对降水的响应模式。结果表明，干旱区草地的NPP主要受水分可利用性的限制，气候和土壤因子影响较微弱；而湿润区草地的NPP更容易受气候因子影响（太阳辐射、温度等），且土壤因子（土壤含水量、土壤容重等）也在其中起到了关键作用。此外，研究还发现NPP对降水非对称性的响应在不同气候区的草地系统中存在显著差异，在干旱区，NPP在太阳辐射高、VPD低和土壤砂含量高的环境下对降水呈现正响应，而在湿润区，NPP在降水适中、温度适中、土壤黏粒含量低的条件下对降水的正响应更为显著。NPP对环境条件具有一定的阈值效应，超过阈值范围则会导致非对称响应的正响应减弱，甚至为负响应。研究结果为未来草地管理和气候适应策略的制定提供了理论依据，同时也为提升全球碳循环模型的精确性，进一步探索复杂气候变化背景下草地生态系统的响应机制奠定了基础。

关键词: 草地生态系统, 净初级生产力, 空间分布, 气候因子, 非对称响应