Search

The development of approaches to determine the spatial variability of nitrogen (N) into coffee leaves is essential to increase productivity and reduce production costs and environmental impacts associated with excessive N applications. Thus, this study aimed to assess the potential of the Random Forest (RF) machine learning method applied to vegetation indices (VI) obtained from Remotely Piloted Aircraft (RPA) images to measure the N content in coffee plants. A total of 10 VI were obtained from multispectral images by a camera attached to a rotary-wing RPA. The RGB orthomosaic was used to determine sampling points at the crop area, which were ranked by N levels in the plants as deficient, critical, or sufficient. The chemical analysis of N content in the coffee leaves, as well as the VI values in sample points, were used as input parameters for the image training and its classification by the RF. The suggested model has shown global accuracy and a kappa coefficient of up to 0.91 and 0.86, respectively. The best results were achieved using the Green Normalized Difference Vegetation (GNDVI) and Green Optimized Soil Adjusted Vegetation Index (GOSAVI). In addition, the model enabled the evaluation of the spatial distribution of N in the coffee trees, as well as quantification of N deficiency in the crop for the whole area. The GNDVI and GOSAVI allowed the verification that 22% of the entire crop area had plants with N deficiency symptoms, which would result in a reduction of 78% in the amount of N applied by the producer. [ABSTRACT FROM AUTHOR]

The challenges of securing future food security will require deployment of innovative technologies to accelerate crop production. Plant phenotyping methods have advanced significantly, spanning low-cost hand-held devices to large-scale satellite imaging. Field-based phenotyping aims to capture plant response to the environment, generating data that can be used to inform breeding and selection requirements. This in turn requires access to multiple representative locations and capacities for collecting useful information. In this paper we identify the current challenges in access to field phenotyping in multiple locations in Europe based on stakeholder feedback. We present a map of current infrastructure and propose opportunities for greater integration of existing facilities for meeting different user requirements. We also review the currently available technology and data requirements for effective multi-location field phenotyping and provide recommendations for ensuring future access and co-ordination. Taken together we provide an overview of the current status of European field phenotyping capabilities and provides a roadmap for their future use to support crop improvement. This provides a wider framework for the analysis and planning of field phenotyping activities for crop improvement worldwide. • Field phenotyping is a common challenge within the area of crop phenotyping. • Capturing infrastructure availability and accessibility supports co-ordination. • Multi-site crop examples provide a framework for future recommendations. • New technologies, common access and data policies will ensure long-term value. • Co-ordinated field phenotyping will support future crop research and breeding. [ABSTRACT FROM AUTHOR]

• Global sensitivity analysis highlighted most influential parameters. • A guiding framework for data collection for process-based crop models was developed. • Calibration data for AquaCrop should include multiple years and soil types. • Soil moisture sensors for continuous monitoring are preferred. • Crop observations every two weeks are advised. The collection of high-quality calibration data is essential for the estimation of parameter values and reliability of crop models. However, few tools are available to quantify the minimum number of observations needed for parameter estimation. We therefore here applied practical identifiability analysis, based on global sensitivity analysis, to design measurement campaigns on farmers' fields. We applied the method for parameterization of the AquaCrop model for mid-early potatoes in Belgium. We generated several virtual observational datasets, considering multiple weather and soil conditions, and measurement frequencies and variables. This analysis resulted in experimental designs where measurement campaigns should be conducted over at least two growing seasons and in different soil types, using soil moisture sensors combined with field observations every two weeks. This method showed to be a useful planning tool for the collection of sufficient data for the calibration of process-based crop models. [ABSTRACT FROM AUTHOR]