Remote Sensing in Hemp

Printer-friendly .pdf version of this page | Glossary

Maylin Murdock, Savanna Shelnutt, Tian Qiu, Schuyler Seyram, Sam Reisinger, Alex Wares, Kathleen Kanaley, Haoyu Niu, Katie Gold, Yu Jiang, Mike Irey, David Suchoff, Larry Smart

Remote sensing is collecting and learning information about the earth’s surface and atmosphere without being in contact with it using electromagnetic radiation (EM). Depending on the source of radiation, sensors of EM are characterized as either active (Figure 1b) or passive (Figure 1a) as well as imaging or non-imaging. Current phenotyping methods for hemp rely primarily on extensive, destructive data collection and processing, which isn’t sustainable given the urgency for genetic characterization and improvement. Our remote sensing research team works to address this issue and current gaps in precision agriculture practices in hemp breeding and production systems with a handful of sensors (Table 1). We aim to offer high-throughput plant phenotyping (HTPP) approaches to advance breeding efforts in hemp programs.

Sensors & Applications

Sensors	Characteristics	Common Applications	Example VIs
RGB	High Spatial Detail Visual Observations	Crop Monitoring Weed Detection	NGRDI – Biomass RGBVI – Growth Stage
Multispectral	l More Spectral Bands Near-Infrared	Biomass & Yield Mapping Plant Health	NDVI – Plant Stress EVI – Plant Health
Hyperspectral	tral High Spectral Detail Continuous Spectral Bands	Disease Detection Nutrient Detection	NDII – Water Content NDNI – Nitrogen Content
Thermal	Temperature Differences	s Irrigation Management Disease Mapping	CWSI – Crop Water Stress
Lidar	3D Information	Crop Yields Modeling & Mapping	g LAI – Leaf Area

Table 1. Sensors and common applications used in precision agriculture. Table adapted from Drones for Agriculture, edX Course.

Wide-Spaced Trial

Our goal for this research trial (Figure 2) is to characterize plant architecture, growth stages, and yield estimation in hemp.

Research Objectives

• Extensive phenotyping
• Reconstructing individual plants with an aerial-based lidar sensor
• Evaluating vegetation indices using an aerial-based hyperspectral sensor
• Identifying HTPP traits optimal for yield estimation

Research Prospects

This trial enables future research and development toward remote sensing methods in phenotyping and quantifying flowering time, sex determination, and
more (Table 2).

Applications	Sensors
Chemotype & Cannabinoid Content	Hyperspectral
Downey Mildew & Powdery Mildew	Thermal & Multi or Hyperspectral
Grain Yield & Fiber Quality	Lidar & Multi or Hyperspectral

Table 2. Remote sensing research prospects.

Glossary

• EM – Electromagnetic
• EMR – Electromagnetic Radiation
• HTPP – High-throughput Plant Phenotyping
• EOH – Essential Oil Hemp
• NRG – Near-infrared Red Green
• NDVI – Normalized Difference Vegetation Index
• TLS – Terrestrial Laser Scanner
• DEM – Digital Elevation Model
• VI – Vegetation Index
• RGB – Red Green Blue
• NGRDI – Normalized Green Red Difference Index
• RGBVI – Red Green Blue Vegetation Index
• EVI – Enhanced Vegetation Index
• NDII – Normalized Difference Infrared Index
• NDNI – Normalized Difference Nitrogen Index
• CWSI – Crop Water Stress Index
• LAI – Leaf Area Index

Acknowledgments: This project was funded by a grant from the Foundation for Food and Agriculture Research (FFAR) with matching support provided by U.S. Sugar Corporation.