Frost Damage Index: The Antipode of Growing Degree Days

Abstract

Abiotic stresses such as heat and frost limit plant growth and productivity. Image-based field phenotyping methods allow quantifying not only plant growth but also plant senescence. Winter crops show senescence caused by cold spells, visible as declines in leaf area. We accurately quantified such declines by monitoring changes in canopy cover based on time-resolved high-resolution imagery in the field. Thirty-six winter wheat genotypes were measured in multiple years. A concept termed "frost damage index" (FDI) was developed that, in analogy to growing degree days, summarizes frost events in a cumulative way. The measured sensitivity of genotypes to the FDI correlated with visual scorings commonly used in breeding to assess winter hardiness. The FDI concept could be adapted to other factors such as drought or heat stress. While commonly not considered in plant growth modeling, integrating such degradation processes may be key to improving the prediction of plant performance for future climate scenarios.

Fig. 1.

HTFP-based RGB image (color), plant-soil segmentation (black/white), and extracted CC per area, and plot (%) for 2 wheat varieties, Runal (susceptible to freezing stress) and CH Combin (more tolerant to freezing stress), before (20 February 2018) and after (5 March 2018) a cold spell.

Fig. 2.

Overview of the data collection and processing workflow for CC, measured with the FIP using the example of 2018 and the variety Ludwig. The FIP enabled to take RGB images repeatedly throughout the growing season (A). RGB images were then segmented in plant and nonplant pixels (B). From these images, the CC per specific area was extracted (C). In addition to RGB images, temperature data were collected from a nearby weather station (D). The RGB and segmented images (A and B) and red line in (C) depict the variety Ludwig, and gray lines in (C) depict all other examined varieties.

Fig. 3.

The concept of the FDI (B) based on air-temperature measurements (A) and its relation to CC decline (C). The FDI is defined as the sum of temperatures below a certain threshold (A, red line), therefore summarizing the duration and severity of an event (D), with no frost stress leading to zero values and stress leading to negative values (B). Measured CC changes (ΔCC) are positive when no stress occurs and negative when stress occurs (C). Relating the FDI to declining CC requires 2 additional parameters, the time lag between frost events and visible damage (B and C, red arrow) and the sensitivity to the FDI (B and C, black arrows). The time lag allows to account for a delayed appearance of CC decline after a frost event, e.g., due to delayed biotic and/or abiotic processes. Temperature courses may be smoothed before deriving the FDI [light blue versus dark blue lines in (A)], and smoothing reduces the leverage of extreme frost events.

Fig. 4.

Rotated and scaled original image (bottom) and segmented plant pixels (top) with identified sowing rows (black rectangles) and plot shape (outer bound).

Fig. 5.

Three steps of the evaluation and parameter estimation process: the first 2 on row-based levels using data from all genotypes and the third step on a genotype-specific level.

Fig. 6.

Predicted frost damage based on the FDI plotted against CC decline (ΔCC measured per sowing row). A time lag of 3 d and temperature course smoothing with an 18-h moving average were applied. The solid line shows the 1:1 relation, red dots represent the marked variety CH Combin, and blue dots indicate the variety Runal (see Fig. 1). The different years are depicted by the shape. The optimization was once performed with fixed T_base to −9 °C and a global parameter s (A), and once with fixed T_base to −9 °C and a genotype-specific parameter s (B). The distribution of genotype-specific estimations for s for single years (2018 and 2021) and the multiyear combination are shown in (C). MAE, mean absolute error.