Abstract Climate change is dramatically affecting small-scale farming productivity in the global south. The phenotypic and pedoclimatic characterization of indigenous crop resources may provide an opportunity to address future challenges threatening food security. This study focuses on a vast Ethiopian ex situ teff collection, comprising 3850 accessions sampled from a range of agroecologies across the country. Eleven phenotypic traits were collected in an experimental field to describe the diversity existing in the collection, unveiling a broad variation for a number of traits of breeding relevance. The sampling information available for 2084 accessions was used to describe the teff growing range in relation to 19 bioclimatic variables and soil features. The geographic distribution of the collection showed that different teff accessions are cultivated in a broad range of environments with diverse bioclimatic features. The combination of phenotypic and pedoclimatic diversity in the collection showed the differential distribution of trait values in relation to different agroecologies. A species distribution model was developed using the bioclimatic characterization of the current teff cropping area. When applied to climate projections, the model showed that the suitability of teff cultivation will diminish by 2070, supporting the need for breeding and conservation efforts tapping the great diversity available in Ethiopian teff.

中文翻译：

---

埃塞俄比亚画眉草画眉草（Eragrostis teff）在其整个种植范围内的当前和预计的生态地理适应和表型多样性

摘要 气候变化正在显着影响全球南方的小规模农业生产力。本土作物资源的表型和气候特征可能为应对未来威胁粮食安全的挑战提供机会。这项研究的重点是埃塞俄比亚的大量异地画眉草收藏，包括从全国范围内的一系列农业生态学中采样的 3850 种材料。在一个实验田中收集了 11 个表型性状，以描述该集合中存在的多样性，揭示了许多育种相关性状的广泛变异。可用于 2084 个种质的采样信息用于描述与 19 个生物气候变量和土壤特征相关的画眉草生长范围。藏品的地理分布表明，不同生物气候特征的广泛环境中种植了不同的画眉草种质。集合中表型和土壤气候多样性的结合显示了与不同农业生态相关的性状值的差异分布。物种分布模型是使用当前画眉草种植区的生物气候特征开发的。当应用于气候预测时，该模型表明，到 2070 年，画眉草种植的适宜性将降低，从而支持利用埃塞俄比亚画眉草丰富的多样性进行育种和保护工作的必要性。集合中表型和土壤气候多样性的结合显示了与不同农业生态相关的性状值的差异分布。物种分布模型是使用当前画眉草种植区的生物气候特征开发的。当应用于气候预测时，该模型表明，到 2070 年，画眉草种植的适宜性将下降，从而支持利用埃塞俄比亚画眉草丰富的多样性进行育种和保护工作的必要性。集合中表型和土壤气候多样性的结合显示了与不同农业生态相关的性状值的差异分布。物种分布模型是使用当前画眉草种植区的生物气候特征开发的。当应用于气候预测时，该模型表明，到 2070 年，画眉草种植的适宜性将降低，从而支持利用埃塞俄比亚画眉草丰富的多样性进行育种和保护工作的必要性。