Coincident detection of crop water stress, nitrogen status and canopy density using ground-based multispectral data.

TLDR

Remotely sensed data is a key tool for precision crop management, used for zoning, nutrient mapping, and pest detection, yet most studies rely on single‑factor experiments, limiting the link between multispectral signatures and crop condition variability. This study aimed to evaluate whether remotely sensed data could differentiate water stress from nitrogen stress in cotton during the 1999 season near Phoenix, Arizona. The authors collected visible, near‑infrared, and thermal multispectral data with a prototype sensor mounted on a linear‑move irrigation system, quantified crop water status with a neutron probe, and assessed nitrogen status using petiole samples.

Abstract

Remotely sensed data has been identified as an important tool for precision crop management (PCM). The data has been used to assist in the identification of management zones, map crop nutrient status, and detect pest infestations. However, in many of the examples cited, the correlation between a multispectral signature and the variation of interest was limited to single factor experiments (i.e., only one factor was primarily responsible for the variability in crop condition). A water by nitrogen experiment was conducted during the 1999 cotton season near Phoenix, Arizona, where one objective was to test the ability of remotely sensed data to distinguish between water and nitrogen stress. Multispectral (visible, near infrared and thermal) data were collected using a prototype sensor mounted on a linear move irrigation system. Neutron probe data were used to quantify crop water status, and petiole samples were used to