Shifting seasons: Long-term crop dynamics across agroclimatic regions of Czechia

Abstract

This study analyzes the evolution of phenological (start-of-season, end-of-season, length-of-season, day of maximum-of-season) and productivity (small and large seasonal integrals) parameters for six major crop types in Czechia (winter cereals, spring cereals, winter rapeseed, fodder crops, sugar beetroot, and corn), using a 35-year Landsat time series (1986–2020). The leaf area index (LAI) was retrieved using an artificial neural network regression model trained on PROSAIL radiative transfer simulations and validated with extensive in situ measurements collected in 2017 and 2018 in the lowlands of Central Bohemia. The supervised classification of Landsat quarterly composites enabled the identification of crop spatial patterns for each growing season. Phenological and productivity indicators were then derived from LAI time series aggregated at the level of ten agro-climatic regions using the threshold approach. Changes in phenological and productivity parameters over the examined period were assessed through the linear least squares regression analysis and the significance of trends was tested. Results revealed significant negative trends in the end-ofseason and day of maximum-of-season for winter and spring cereals, winter rapeseed (up to –0.7 days/year), and fodder crops (up to –1.6 days/year), indicating an earlier maturation and harvest. Significant differences in trends in phenological and productivity parameters were observed between agro-climatic regions in more than 40% of cases, and the response was observed to be highly crop-specific. While the shift in harvest dates and the shortening of the season for corn and fodder crops were more pronounced in warmer regions, the shift in winter rapeseed phenology occurred more rapidly in colder regions. The findings underscore the relevance of crop type and regional climate in shaping phenological responses, offering a basis for future research and planning of agricultural adaptation strategies.

Plant phenotyping relevance

LandsatからLAIを推定し、作物のフェノロジー・生産性形質を抽出するリモートセンシング手法を、PROSAIL/ANNモデルと現地測定で検証しており、形質取得ワークフローが主要な役割を担う。

Code and data availability

All supplied blocks are the article's reference list and a table caption; no authors' public phenotype datasets, images, code, models, or supplements with availability statements are present. Cited external resources (ARCSI, TIMESAT, HR-VPP manual, CHMI yearbook, etc.) are generic third-party tools/data, not paper-phen

No evidence-backed public reproduction asset is currently recorded.