Land-use generates multiple stress factors, and we need to understand their effects on plant–plant interactions to predict the consequences of land-use intensification. The stress–gradient hypothesis predicts that the relative strength of positive and negative interactions changes inversely under increasing environmental stress. However, the outcome of interactions also depends on stress factor’s complexity, the scale of analysis, and the role of functional traits in structuring the community. We evaluated plant–plant co-occurrences in a temperate forest, aiming to identify changes in pairwise and network metrics under increasing silvopastoral use intensity. Proportionally, positive co-occurrences were more frequent under high than low use, while negative co-occurrences were more frequent under low than high. Networks of negative co-occurrences showed higher centralization under low use, while networks of positive co-occurrences showed lower modularity and higher centralization under high use. We found a partial relationship between co-occurrences and key functional traits expected to mediate facilitation and competition processes. Our results shows that the stress-gradient hypothesis predicts changes in spatial co-occurrences even when two stress factors interact in a complex way. Networks of negative co-occurrences showed a hierarchical effect of dominant species under low use intensity. But positive co-occurrence network structure partially presented the characteristics expected if the facilitation was an important mechanism characterizing the community under high disturbance intensity. The partial relationship between functional traits and co-occurrences may indicate that other factors besides biotic interactions may be structuring the observed negative spatial associations in temperate Patagonian forests.

土地利用产生多种压力因素，我们需要了解它们对植物-植物相互作用的影响，以预测土地利用集约化的后果。压力梯度假说预测，在环境压力增加的情况下，正负相互作用的相对强度会发生相反的变化。然而，相互作用的结果还取决于压力因素的复杂性、分析的规模以及功能特征在构建社区中的作用。我们评估了温带森林中植物 - 植物的共生情况，旨在确定在林牧使用强度增加的情况下成对和网络指标的变化。按比例，高使用率下的积极共现比低使用率更频繁，而低使用率下的负共现比高使用率更频繁。负共现网络在低使用率下表现出较高的集中度，而正共现网络在高使用率下表现出较低的模块化和较高的集中度。我们发现共现与预期介导促进和竞争过程的关键功能特征之间存在部分关系。我们的结果表明，即使两个压力因素以复杂的方式相互作用，压力梯度假设也能预测空间共现的变化。负共现网络显示出在低使用强度下优势物种的等级效应。但如果促进是表征高干扰强度下社区的重要机制，则正共现网络结构部分呈现了预期的特征。