Skip to main content
ARS Home » Research » Publications at this Location » Publication #338463

Title: Workflow to establish time-specific zones in precision agriculture by spatiotemporal integration of plant and soil sensing data

item SCUDIERO, ELIA - University Of California
item TEATINI, PIETRO - University Of Padua
item MANOLI, GABRIELE - Eth Zurich
item BRAGA, FEDERICA - National Research Council - Italy
item Skaggs, Todd
item MORARI, FRANCESCO - University Of Padua

Submitted to: Agronomy
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/1/2018
Publication Date: 11/7/2018
Citation: Scudiero, E., Teatini, P., Manoli, G., Braga, F., Skaggs, T.H., Morari, F. 2018. Workflow to establish time-specific zones in precision agriculture by spatiotemporal integration of plant and soil sensing data. Agronomy. 8(11):253.

Interpretive Summary: Because agricultural soils are by nature non-homogeneous, different sections of agricultural fields require different levels of inputs (water, nutrients, etc.) for crop production. Developing agricultural operations that optimize resource allocation according to space- and time-varying crop requirements will lead to a more efficient and sustainable agriculture. In this work, we proposed an approach that can be used to identify plant needs and management zones throughout the growing season, using spatial information on soil properties and in-season measurements of crop status. A case study for a 21-ha corn field in northeastern Italy was developed using soil maps and satellite imagery. The research demonstrates the potential benefits of incorporating information on in-season, time- and space-varying soil and crop conditions into site-specific management plans and decisions. Farmers, extension specialists, and research scientists can benefit from the proposed workflow, which should lead to more efficient use of resources, including water and nutrients.

Technical Abstract: Management zones (MZs) are used in precision agriculture to diversify agronomic management across a field. According to current common practices, MZs are often spatially static: they are developed once and used thereafter. However, the soil–plant relationship often varies over time and space, decreasing the efficiency of static MZ designs. Therefore, we propose a novel workflow for time-specific MZ delineation based on integration of plant and soil sensing data. The workflow includes four steps: (1) geospatial sensor measurements are used to describe soil spatial variability and in-season plant growth status; (2) moving-window regression modelling is used to characterize the sub-field changes of the soil–plant relationship; (3) soil information and sub-field indicator(s) of the soil–plant relationship (i.e., the local regression slope coefficient[s]) are used to delineate time-specific MZs using fuzzy cluster analysis; and (4) MZ delineation is evaluated and interpreted. We illustrate the workflow with an idealized, yet realistic, example using synthetic data and with an experimental example from a 21-ha maize field in Italy using two years of maize growth, soil apparent electrical conductivity and normalized difference vegetation index (NDVI) data. In both examples, the MZs were characterized by unique combinations of soil properties and soil–plant relationships. The proposed approach provides an opportunity to address the spatiotemporal nature of changes in crop genetics × environment × management interactions.