Tohoku Agricultural Research Center, NARO, 4 Akahira, Shimo-kuriyagawa, Morioka, Iwate, 020-0198, Japan.
Hokuriku Research Station, Central Region Agricultural Research Center, NARO, 1-2-1 Inada, Joetsu, Niigata, 943-0193, Japan.
BMC Ecol Evol. 2022 May 25;22(1):70. doi: 10.1186/s12862-022-02024-7.
Inter-regional relationships between landscape factors and biological responses in natural conditions are important but difficult to predict because of the differences in each landscape context and local environment. To examine the inter-regional variability in relation to landscape factors and the biological response of an insect pest of rice, Stenotus rubrovittatus, we extrapolated a damage prediction model (the 'original model' of our previous study) for rice using land-use data. The 'original model' comprised as fixed factors the area of source habitat (i.e. pastures and graminoid-dominated fallow fields), soybean fields, and rice paddies within 300-m radii with research years as the random intercept. We hypothesized that the original model would be applicable to new regions, but the predictive accuracy would be reduced. We predicted that fitting a new extended model, adjusting the parameter coefficients of identical fixed factors of the 'original model,' and adding regional random intercepts would improve model performance (the 'extended model'). A field experiment was conducted in two regions that had a similar landscape context with the original region, each in a different year of four years in total. The proportion of rice damage and surrounding land use within a 300-m radius was investigated, and the data were applied to the models and the applicability and accuracy of the models were examined.
When the 'original model' was assigned to the combined data from the original and extrapolated regions, the relationship between the observed and the predicted values was statistically significant, suggesting that there was an inter-regional common relationship. The relationship was not statistically significant if the model was applied only to the new regions. The extended model accuracy improved by 14% compared with the original model and was applicable for unknown data within the examined regions as demonstrated by three-fold cross validation.
These results imply that in this pest-crop system, there is likely to be a common inter-regional biological response of arthropods because of landscape factors, although we need to consider local environmental factors. We should be able to apply such relationships to identify or prevent pest hazards by offering region-wide management options.
在自然条件下，景观因素与生物响应之间的区域间关系很重要，但由于每个景观背景和局部环境的差异，很难预测。为了研究与景观因素相关的区域间变异性以及水稻害虫 Stenotus rubrovittatus 的生物反应，我们使用土地利用数据推断了一个用于预测水稻损害的模型（我们之前研究的“原始模型”）。“原始模型”包含作为固定因素的源生境（即牧场和禾本科植物占优势的休耕地）、大豆田以及 300 米半径内的稻田，以研究年份作为随机截距。我们假设原始模型适用于新的地区，但预测准确性会降低。我们预测，拟合一个新的扩展模型，调整“原始模型”中相同固定因素的参数系数，并添加区域随机截距，将提高模型性能（“扩展模型”）。在两个与原始地区具有相似景观背景的地区进行了野外实验，每个地区在四年中的四年中进行了一次。调查了 300 米半径内的水稻损害比例和周围土地利用情况，并将数据应用于模型中，检验了模型的适用性和准确性。
当将“原始模型”分配给原始和外推区域的组合数据时，观测值与预测值之间的关系具有统计学意义，表明存在区域间共同关系。如果仅将模型应用于新地区，则该关系没有统计学意义。与原始模型相比，扩展模型的准确性提高了 14%，并且在经过三折交叉验证后，该模型可适用于检查区域内的未知数据。
这些结果表明，在这种害虫-作物系统中，由于景观因素，昆虫可能存在普遍的区域间生物反应，尽管我们需要考虑局部环境因素。我们应该能够通过提供区域管理选项来应用这些关系来识别或预防害虫危害。
