Local agriculture in India is highly vulnerable to a wide range of anthropogenic-induced natural disasters. Most studies regarding the vulnerability of farmers in India are performed at coarse spatial scales. These studies fail to assess the spatial distribution of vulnerability status in local regions where small and marginal farmers own lands. Consequently, the adaptation measures fail to pass on to the grassroots level. This study attempts to 1) assess the vulnerability status of local agricultural sector among the Sub-district Administrative Units (SDAUs) of Gaya district in Bihar, and 2) identify suitable adaptation models using two aggregation methods to reduce potential risks. Both aggregation methods are used to compute indices of exposure, sensitivity and adaptive capacity followed by their classification under five vulnerability categories. The degree of transition among categories is analyzed for each SDAU to find a suitable adaptation model, i.e., incremental, systemic and transformational. Our results show that, in the case of exposure, only three SDAUs shifted their categories. In sensitivity and adaptive capacity cases, 41.66% and 45.83% SDAUs are found to shift their categories respectively. Moreover, SDAUs facing higher exposure require systemic model of adaptation. SDAUs facing higher sensitivity need both systemic and transformational models, while SDAUs with lower adaptive capacity found the systemic adaptation model to be the best suited. Such a vulnerability assessment of a local area, which also facilitates the identification of suitable adaptation models, can assist agricultural agencies in reducing the risks of potential disasters by implementing efficient adaptation strategies.

印度当地农业极易受到各种人为引发的自然灾害的影响。大多数关于印度农民脆弱性的研究都是在粗略的空间尺度上进行的。这些研究未能评估当地小农和边缘农民拥有土地的脆弱性状态的空间分布。因此，适应措施未能传递到基层。本研究试图 1) 评估比哈尔邦 Gaya 区街道行政单位 (SDAU) 中当地农业部门的脆弱性状况，以及 2) 使用两种聚合方法确定合适的适应模型以降低潜在风险。两种聚合方法都用于计算暴露指数，敏感性和适应能力，然后根据五个脆弱性类别进行分类。分析每个SDAU的类别之间的过渡程度，以找到合适的适应模型，即增量的、系统的和转换的。我们的结果表明，在暴露的情况下，只有三个 SDAU 改变了他们的类别。在敏感性和适应能力情况下，发现分别有 41.66% 和 45.83% 的 SDAU 改变了它们的类别。此外，面临更高暴露的 SDAU 需要系统的适应模型。面临更高灵敏度的 SDAU 需要系统模型和转换模型，而适应能力较低的 SDAU 发现系统适应模型最适合。这种局部区域的脆弱性评估，也有助于确定合适的适应模型，