<p><i>Striga hermonthica</i> is a root-parasitic plant that attacks many important cereal crops, including sorghum, teff, and rice. It is one of the most destructive agricultural weeds, posing a major threat to food security in Ethiopia. This study aimed to map suitable habitats for <i>S. hermonthica</i> in Ethiopia and to evaluate potential changes in its range under projected climate scenarios for 2050 and 2070. One hundred thirty-two spatially filtered occurrence records were used in distribution modeling. Prior to modeling, the environmental variables were tested for multicollinearity using Pearson correlation and Variance Inflation Factor, leading to the selection of 10 uncorrelated predictors. The ensemble model, built using seven algorithms with 10 replicates each, demonstrated strong performance (mean AUC = 0.92; TSS = 0.75). The most influential environmental variables were precipitation during the warmest quarter and temperature seasonality. Under current climate conditions, approximately 234,326.35&#xa0;km² were found suitable for the weed. Projections under both intermediate and high-emission scenarios indicate a likely expansion of its range. Northern, northwestern, western, and eastern lowlands of Ethiopia are comparatively identified as risk area for this weed. Thus, all stakeholders, including researchers, policymakers, farmers and governmental and non-governmental organizations, should take early expansion-controlling management plan like create awareness for farmers.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Projecting the spatio-temporal habitat suitability of Striga hermonthica under climate change scenarios in Ethiopia using ensemble modeling

  • Mikiyas Abebe,
  • Zerihun Woldu,
  • Zemede Asfaw,
  • Bikila Warkineh

摘要

Striga hermonthica is a root-parasitic plant that attacks many important cereal crops, including sorghum, teff, and rice. It is one of the most destructive agricultural weeds, posing a major threat to food security in Ethiopia. This study aimed to map suitable habitats for S. hermonthica in Ethiopia and to evaluate potential changes in its range under projected climate scenarios for 2050 and 2070. One hundred thirty-two spatially filtered occurrence records were used in distribution modeling. Prior to modeling, the environmental variables were tested for multicollinearity using Pearson correlation and Variance Inflation Factor, leading to the selection of 10 uncorrelated predictors. The ensemble model, built using seven algorithms with 10 replicates each, demonstrated strong performance (mean AUC = 0.92; TSS = 0.75). The most influential environmental variables were precipitation during the warmest quarter and temperature seasonality. Under current climate conditions, approximately 234,326.35 km² were found suitable for the weed. Projections under both intermediate and high-emission scenarios indicate a likely expansion of its range. Northern, northwestern, western, and eastern lowlands of Ethiopia are comparatively identified as risk area for this weed. Thus, all stakeholders, including researchers, policymakers, farmers and governmental and non-governmental organizations, should take early expansion-controlling management plan like create awareness for farmers.