| 摘 要: 文章旨在探索锡林郭勒草原的可持续放牧策略,以防止草原干旱和荒漠化。从机理分析的角度研究不同放牧政策对锡林郭勒草原土壤物理性质、植被生物量的影响,通过建立数学模型优化放牧问题,以期找到在锡林郭勒草原可持续发展情况下草场内放牧羊数量的最大阈值。此外,开发了一个基于双向长短期记忆网络的预测模型,用于准确预测不同深度土壤的湿度,以支持草原管理决策。模型在测试集上的表现与实际土壤湿度趋势一致,其中200 cm深度的土壤湿度RMSE 仅为0.2,显示出其在锡林郭勒草原土壤湿度预测中的实用价值。 |
| 关键词: 放牧策略;土壤湿度;机理分析;双向长短期记忆网络 |
|
中图分类号: TP183
文献标识码: A
|
|
| Study on the Impact of Grazing Strategies on Soil and Vegetation and Soil Moisture Prediction |
|
WANG Kai1, LI Zhong2
|
(1.School of Computer Science and Technology, Zhejiang Sci-Tech University, Hangzhou 310018, China;
2.School of Science, Zhejiang Sci-Tech University, Hangzhou 310018, China)
202130504163@mails.zstu.edu.cn; lizhong@zjhu.edu.cn
|
| Abstract: The paper aims to explore sustainable grazing strategies for the Xilingol Grassland to prevent drought and desertification. From a mechanistic analysis perspective, it studies the effects of different grazing policies on the physical properties of soil and vegetation biomass in the Xilingol Grassland. By establishing a mathematical model to optimize grazing issues, the research seeks to find out the maximum threshold for the number of sheep that can be grazed in the grassland under sustainable development conditions in Xilingol. Additionally, a prediction model based on Bidirectional Long Short-Term Memory (BiLSTM) networks is developed to accurately predict soil moisture at varying depths, supporting grassland management decisions. The model's performance on the test set is consistent with the actual soil moisture trend, with a Root Mean Square Error (RMSE) of only 0.2 at a depth of 200 cm, demonstrating its practical value for soil moisture prediction in the Xilingol Grassland. |
| Keywords: grazing strategies; soil moisture; mechanistic analysis; BiLSTM |
