Expanding DWD phenology monitoring through crowdsourced plant reports and AI-enabled data partnership

Abstract

Phenology at the Deutscher Wetterdienst (DWD) involves the observation and analysis of recurring, seasonally driven development stages of plants, such as flowering, leaf unfolding, fruit ripening, and leaf fall. These phenological phases are closely linked to weather and climate conditions and therefore serve as biologically based indicators of climate variability and long-term change. The DWD phenological observations are used in multiple contexts, including climate monitoring and diagnostics, agrometeorological services, and scientific studies that analyse shifts in phase timing in relation to temperature trends. Phenological information also supports operational applications, for example in agricultural modelling and other environment-related services.A key element of DWD’s phenological activities is its long-running observation network, which is based on standardized and quality-controlled field observations. In recent years, DWD has started to complement this classical network with additional digital data sources in order to improve spatial coverage and temporal resolution. One important development is the crowdsourcing feature “Pflanzenmeldungen” in the DWD WarnWetter app, which allows citizens to submit geo-referenced reports of plant development stages via smartphone. These reports can help capture regional differences, extreme-year signals, and near-real-time vegetation responses, particularly in areas with fewer traditional observations. By integrating these app-based reports with established phenological datasets and expert workflows, DWD aims to extend and enrich its national phenology record while maintaining scientific usability.Further expansion of digital data sources is planned through a collaboration with Flora Incognita, a machine-learning-based plant identification platform. This cooperation aims to link phenological monitoring with scalable species recognition and user-driven reporting. The combination of supported plant identification and structured phenological input is expected to improve data consistency, encourage participation, and increase the volume and resolution of phenological information.Overall, DWD’s phenology programme is developing toward a hybrid monitoring system that integrates long-term standardized observations with crowdsourced app data and AI-supported plant identification. This approach enhances spatial and temporal coverage and strengthens the basis for monitoring climate impacts on vegetation and for providing robust phenology-based climate services.

Plant phenotyping relevance

植物の開花・葉展開・果実成熟・落葉などの表現型を、市民参加型アプリとAI植物同定を統合して取得・拡張する監視基盤が中心であり、単なる生物学的実験での routine 測定ではない。

Code and data availability

公開論文であることは確認できましたが、現在の公式API・許可済み取得経路では本文を自動取得できませんでした。

No evidence-backed public reproduction asset is currently recorded.