This proposal aims to create a field-specific, grower-friendly DRSI (Disease Risk Sensitivity Index) to measure potential disease risk based on a set of available remotely-sensed geospatial layers that can be integrated to reveal potential threats from common Michigan diseases of soybeans. These available geospatial layers when combined provide a base map that is used to denote potential bean yield reduction risk. Aerial imagery can confirm or deny these problem areas by using the optical wavelength to reveal differences in the crop’s canopy. The thermal wavelength reveals differences in plants affected by water deficits, showing signs of “fever”, or affected by diseases, as they close their stomata as protection mechanism, and become water stressed.

The main objective of this proposal is to create a disease risk sensitivity index (DRSI) by integrating relevant geospatial layers of information to discern indications of potential disease in soybeans in Michigan. To validate the large number of acres covered by aerial imagery, ground truthing data of disease visible signatures will be collected from fields where known diseases occur in cooperation with traditional field scouting. Specific objectives include:

1) Over 800 acres of soybeans will be enrolled into a remote sensing program that captures 7-10 images a year in multiple VIs (vegetation indices)

2) High resolution UAV imagery will be collected in conjunction with scouted fields where diseases are known to be present or of high risk based on grower experiences.

3) Compare and analyze trends by relating remotely sensed imagery from plot- to field-scale to determine relationships and enhancing understanding of disease early detection and identification

4) Use 3D digital elevation data, topographic wetness index maps, and imagery (optical and thermal) to identify potential high risk sites within each field as indicated in the following three types of field specific maps: yield stability, topography, soil moisture and thermal stability.

Precision technologies provide ways to visualize data as it pertains to different fields from various implements or platforms. All these data are related through different geospatial concepts that help describe these trends of variability. Historical yield analysis reveals spatial trends in yield variability (Maestrini and Basso, 2018) where field productivity is categorized and mapped. These trends from yield monitor data provide an important understanding into the field’s yield response to biotic and abiotic stressors. Digital elevation models provide insight into topographic features throughout each field, showing areas where water frequently moves and pools.

Combined with the topographic wetness index (TWI), which uses slope and contour (elevation change) to quantify wetness, these maps confirm prominent areas that are more prone to potential disease occurrence. In Martinez-Feria and Basso (2020), areas of fields that fluctuated substantially were categorized as unstable and are further identified as hilltops or depressions. These hilltops where drought is more likely to occur can be useful spots for scouting for potential diseases like charcoal rot, as it favors hot and dry conditions. Depressions accumulate more water due to runoff and water routing during precipitation events and are candidates for white mold.

Remote sensing in the optical wavelength visualizes how the plant canopy size differs throughout the field which can indicate potential threats from disease. Thermal imagery captures heat as it’s reflected from plants in response to the ability to transpire water. Plants that reflect more heat are potentially water stressed or affected by disease, while plants that can transpire with sufficient plant available water and absence of diseases have cooler canopies. These images will clearly inform the scouting process during the growing season to visualize parts of the field that are possibly under threats of disease or insect pressure. Output from this proposal will directly impact farmers and agribusiness by providing a novel method to scout more efficiently and effectively in both small and large soybean fields

Field scouting is a crucial element of a proper integrated management scheme. Remote sensing alone cannot replace the informative and knowledgeable scouts that walk fields to observe potential threats firsthand. Yet, remote sensing as a component of an integrated vulnerability index containing relevant geospatial layers like yield stability, DEMs, and TWI could improve the ability of scouts to pinpoint locales of high disease probability. Michigan soybean farmers collect an inordinate amount of data, many of which gets packaged and shipped to companies with minimal benefit to the farmer. This proposal applies these already available data and processes them to create a map that reveals where potential threats might originate. With this knowledge in hand, scouts could cover more acres efficiently by using remotely sensed imagery to confirm disease presence.