Scholarlink

This study models agricultural intensity patterns in Peru using spatio-temporal Log-Gaussian Cox Processes (LGCP) applied to 74,733 events from the National Agricultural Survey (ENA) 2022–2024 covering 2.6 million km2. The LGCP framework integrated economic, climatic (CHIRPS, ERA5-Land), topographic, and water-source covariates through a hierarchical modeling approach. Spatial clustering was quantified using Ripley’s K-function, revealing significant aggregation with L(r)—r > 0 for distances below 2° (Monte Carlo test p = 0.02). The full parametric LGCP model (M3) demonstrated superior fit with ΔAIC = − 3916 and Pseudo-R2 = 0.175, showing altitude as the dominant driver (+ 81% intensity increase per standard deviation), while harvested area (− 33%), precipitation (− 30%), and temperature (− 29%) exhibited negative effects. Regional analysis revealed 71% of events concentrated in the Highlands at 2.7 × higher intensity than the Coast. Controlled coastal irrigation systems achieved 888 kg ha−1 mm−1 efficiency compared to 195 kg ha−1 mm−1 for highland natural sources. Despite water deficits increasing from 132.6 to 142.7 mm (2022–2024), productivity rose from 8439 to 9225 kg/ha, indicating climate resilience. The LGCP framework provides robust quantification of spatial heterogeneity and enables data-driven policy interventions for sustainable agricultural planning in Andean regions with extreme topographic and climatic gradients.

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

Spatio-temporal Modeling of Agricultural Patterns in Peru Using Log-Gaussian Cox Processes Based on ENA 2022–2024

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Yhack Bryan Aycaya Paco

摘要

This study models agricultural intensity patterns in Peru using spatio-temporal Log-Gaussian Cox Processes (LGCP) applied to 74,733 events from the National Agricultural Survey (ENA) 2022–2024 covering 2.6 million km2. The LGCP framework integrated economic, climatic (CHIRPS, ERA5-Land), topographic, and water-source covariates through a hierarchical modeling approach. Spatial clustering was quantified using Ripley’s K-function, revealing significant aggregation with L(r)—r > 0 for distances below 2° (Monte Carlo test p = 0.02). The full parametric LGCP model (M3) demonstrated superior fit with ΔAIC = − 3916 and Pseudo-R2 = 0.175, showing altitude as the dominant driver (+ 81% intensity increase per standard deviation), while harvested area (− 33%), precipitation (− 30%), and temperature (− 29%) exhibited negative effects. Regional analysis revealed 71% of events concentrated in the Highlands at 2.7 × higher intensity than the Coast. Controlled coastal irrigation systems achieved 888 kg ha−1 mm−1 efficiency compared to 195 kg ha−1 mm−1 for highland natural sources. Despite water deficits increasing from 132.6 to 142.7 mm (2022–2024), productivity rose from 8439 to 9225 kg/ha, indicating climate resilience. The LGCP framework provides robust quantification of spatial heterogeneity and enables data-driven policy interventions for sustainable agricultural planning in Andean regions with extreme topographic and climatic gradients.