Skip to main content
Dryad

Temperature-dependent reproduction in Douglas-fir (Pseudotsuga menziesii): Machine learning-based phenotyping reveals a distinct reproductive thermal niche

Hsu, Hsin-Wu1; Zerrade, Sabrina1; Kim, Soo-Hyung1

Published Sep 25, 2026 on Dryad. https://doi.org/10.5061/dryad.djh9w0wgg

Data files

Sep 25, 2026 version files 1.06 MB

Click names to download individual files

Abstract

Pollen germination and elongation are temperature-sensitive processes that can influence plant reproduction, seed production, and forest regeneration, but their quantification is labor-intensive. This study developed a deep learning-based approach for pollen phenotyping and characterized the temperature dependence of pollen germination and elongation in Douglas-fir (Pseudotsuga menziesii). A convolutional neural network (CNN) was trained to segment pollen grains and classify germination status from microscopic images. We quantified germination percentage and pollen length at incubation temperatures from 5 to 40 °C and fitted gamma functions to estimate optimal temperatures across populations collected along an elevational gradient. See the Materials & Methods section in the corresponding manuscript for a more detailed description of data collection and analysis.