首页|期刊导航|Information Processing in Agriculture|Grassland ecosystem assessments:Integrating UAV-derived features for aboveground biomass estimation

Grassland ecosystem assessments:Integrating UAV-derived features for aboveground biomass estimationOA

中文摘要

Monitoring grasslands presents significant challenges due to temporal and spatial dynamics in their vegetation.This is particularly pronounced in wet grasslands,where moisture dynamics also impact vegetation patterns.Recent advancements in data acquisition and analysis via Unmanned Aerial Vehicles(UAVs)have shown potential for a more comprehensive understanding of vegetation dynamics.However,current UAV-based methods focus predominantly on structural and spectral data analysis.This often overlooks the horizontal heterogeneity within vegetation.This study addresses this gap by integrating texture analysis,alongside structural and spectral data,to enhance aboveground biomass(AGB)estimation.The research was conducted in a heterogeneous wet grassland in eastern Germany under three different cutting frequencies.Regular UAV flights were carried out to obtain RGB(Red,Green,and Blue)and multispectral images,analyzed alongside ground-reference data from 108 plots,to evaluate canopy height and biomass.We tested the performance of Random Forest and Partial Least Squares Regression models for AGB estimation considering different combinations of features including canopy height model vegetation indices and texture analysis.The results demonstrate that texture analysis when combined with traditional spectral and structural data,enhances predictive accuracy,yielding the best R2 values of up to 0.84 for AGB and reducing the relative root mean square errors to 26.6%.The results underline the potential of combining UAV-based features in AGB estimation of heterogeneous grassland ecosystems offering a path forward for more effective ecological monitoring and sustainable grassland management.

Clara Oliva Gonçalves Bazzo;Bahareh Kamali;Murilo dos Santos Vianna;Dominik Behrend;Hubert Hueging;Farshid Jahanbakhshi;Inga Schleip;Paul Mosebach;Almut Haub;Axel Behrendt;Thomas Gaiser

Institute of Crop Science and Resource Conservation(INRES),University of Bonn,Katzenburgweg 5,53115 Bonn,GermanyInstitute of Crop Science and Resource Conservation(INRES),University of Bonn,Katzenburgweg 5,53115 Bonn,GermanyInstitute of Crop Science and Resource Conservation(INRES),University of Bonn,Katzenburgweg 5,53115 Bonn,GermanyInstitute of Crop Science and Resource Conservation(INRES),University of Bonn,Katzenburgweg 5,53115 Bonn,GermanyInstitute of Crop Science and Resource Conservation(INRES),University of Bonn,Katzenburgweg 5,53115 Bonn,GermanyInstitute of Crop Science and Resource Conservation(INRES),University of Bonn,Katzenburgweg 5,53115 Bonn,GermanyFaculty of Landscape Management and Nature Conservation,Eberswalde University for Sustainable Development,16225 Eberswalde,GermanyFaculty of Landscape Management and Nature Conservation,Eberswalde University for Sustainable Development,16225 Eberswalde,GermanyFaculty of Landscape Management and Nature Conservation,Eberswalde University for Sustainable Development,16225 Eberswalde,GermanyLeibniz Centre for Agricultural Landscape Research(ZALF),Gutshof 7,14641 Paulinenaue,GermanyInstitute of Crop Science and Resource Conservation(INRES),University of Bonn,Katzenburgweg 5,53115 Bonn,Germany

农业科技

Remote sensingVegetationEcological monitoringCanopy heightTextureAgriculture digitalization

《Information Processing in Agriculture》 2026 (3)

P.396-414,19

funded by the German Federal Ministry of Education and Research(BMBF)through the Digital Agriculture Knowledge and Information System(DAKIS)Project[Grant number 031B0729E]funded by the Deutsche Forschungsgemeinschaft(DFG,German Research Foundation)-SFB 1502/1-2022􀀀Projektnummer:450058266.

10.1016/j.inpa.2026.01.003

评论