Abstract Crop booms in forest frontiers are a major contributor to land conversion and deforestation. In this study, we investigate the smallholder-driven northern Laos rubber boom in two case study areas (CSAs) with different speed and intensity of rubber expansion. We assess the relative importance of market, contextual, and behavioral factors in fostering or hindering the conversion of forest to rubber plantations. We develop a Bayesian network (BN) model of land-use change based on household surveys, expert interviews, market price data, and land use maps covering the period 2000–2017. We use regression analysis to inform the structure of the BN, compare model results, and analyze time-varying effects. The BN and regression models incorporate perceived price signals as a combination of market price and local price knowledge, and local self-reinforcing imitation dynamics as a combination of aggregate rubber conversion and imitation behavior elicited in the survey. Results show that deforestation was lower in strictly protected areas but not in forests with lesser protection status. Imitation had a large effect on rubber uptake in both areas. In the CSA that experienced the most intensive spread of rubber, price signals transmitted through social networks had a significant impact, especially throughout the stage of rapid expansion. Rubber expansion continued in both areas during periods of descending prices mainly because of increased cash availability and access to inputs. Our research sheds light on the underlying dynamics of crop booms and contributes to the understanding of agricultural expansion processes.

中文翻译：

---

从全球驱动因素到当地土地利用变化：了解老挝北部的橡胶繁荣

摘要 森林边境的作物繁荣是土地转化和森林砍伐的主要因素。在本研究中，我们调查了橡胶扩张速度和强度不同的两个案例研究区 (CSA) 中小农驱动的老挝北部橡胶繁荣。我们评估了市场、环境和行为因素在促进或阻碍森林向橡胶种植园转变方面的相对重要性。我们基于住户调查、专家访谈、市场价格数据和 2000 年至 2017 年期间的土地利用地图开发了土地利用变化的贝叶斯网络 (BN) 模型。我们使用回归分析来告知 BN 的结构、比较模型结果并分析时变效应。BN 和回归模型将感知价格信号作为市场价格和当地价格知识的组合，和局部自我增强的模仿动力学，作为调查中引发的骨料橡胶转化和模仿行为的组合。结果表明，严格保护区的森林砍伐率较低，但保护状况较差的森林则不然。模仿对这两个领域的橡胶吸收都有很大影响。在经历了橡胶传播最密集的CSA，通过社交网络传递的价格信号产生了重大影响，尤其是在整个快速扩张阶段。在价格下跌期间，这两个地区的橡胶扩张继续进行，这主要是因为现金供应增加和获得投入的机会增加。我们的研究揭示了作物繁荣的潜在动力，有助于理解农业扩张过程。