Skip to main content
ARS Home » Pacific West Area » Pullman, Washington » Grain Legume Genetics Physiology Research » Research » Publications at this Location » Publication #328717
Title: Low altitude remote sensing technologies for crop stress monitoring: a case study on spatial and temporal monitoring of irrigated pinto bean

Author
item ZHOU, JIANFENG - Washington State University
item KHOT, LAV - Washington State University
item Boydston, Rick
item Miklas, Phillip - Phil
item Porter, Lyndon

Submitted to: Precision Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/29/2017
Publication Date: 9/18/2017
Citation: Zhou, J., Khot, L.R., Boydston, R.A., Miklas, P.N., Porter, L. 2017. Low altitude remote sensing technologies for crop stress monitoring: a case study on spatial and temporal monitoring of irrigated pinto bean. Precision Agriculture. DOI: 10.1007/s11119-017-9539-0.

Interpretive Summary: Site-specific crop management is a promising approach to maximize crop yield with optimal use of inputs of water and nutrients. Availability of meaningful high resolution crop data at specific growth stages is critical for real-time data driven decisions during the production season. The goal of this study was to evaluate the possibility of using small unmanned aerial system (sUAS) based remote sensing technologies to monitor the crop stress of pinto beans grown under full and deficit irrigation treatments. Multispectral and infrared thermal imaging sensors mounted on a small unmanned aerial system were used to collect data from dry bean field plots at bloom stage, pod fill, and mature pod stages. Indicators such as green normalized vegetation index (GNDVI), canopy cover (CC) and canopy temperature (CT) of dry beans were extracted from imaging data and compared with actual crop yield. Indicators were also correlated with leaf area index (LAI) estimated with a handheld device. The GNDVI, CC and CT were able to distinguish dry beans grown under full and deficit irrigation treatments at each of the three growth stages. The developed indicators (GNDVI and CC) collected in the bloom stage were strongly correlated with dry bean yield. Canopy temperature also showed high correlation with yield when collected at the pod fill stage and mature pod stage. Measurements of GNDVI, CC and CT collected from the aerial unmanned system were equal to or better in monitoring crop stress and estimating crop yield than ground-based leaf area estimates, indicating the potential of such emerging remote sensing tools for monitoring crop stress, use in crop management, and predicting yield.

Technical Abstract: Site-specific crop management is a promising approach to maximize crop yield with optimal use of rapidly depleting natural resources. Availability of high resolution crop data at critical growth stages is a key for real-time data-driven decisions during the production season. The goal of this study was to evaluate the possibility of using small unmanned aerial system (UAS)-based remote sensing technologies to monitor the crop stress of irrigated pinto beans (Phaseolus vulgaris L.) with varied irrigation and tillage treatments. A small UAS with onboard multispectral and infrared thermal imaging sensors was used to collect data from bean field plots on three growth stages in 2015 and 2016. Indicators including green normalized vegetation index (GNDVI), canopy cover (CC, ratio of ground covered by crop canopy to the total plot area) and canopy temperature (CT, °C) of crops were extracted from imaging data and correlated with crop yield data and leaf area index (LAI) estimated with a handheld ceptometer. Results show that GNDVI, CC and CT were able to differentiate crops with full and deficit irrigation treatments at each of the three growth stages in both years. Developed indicators were strongly correlated with crop yield with Pearson correlation coefficients (r) of approximate 0.7 for GNDVI and CC, respectively, in the early growth stage (54 day after planting) in both years. Canopy temperature also showed even stronger correlation with yield with r > 0.8 at early growth stage. Performance of small UAS-based indicators in crop stress monitoring and crop yield estimation was better than or comparable to that of the ground-based LAI estimates, indicating the potential of such a remote sensing tool in rapid crop stress monitoring and management.