Assessing onion salt tolerance using soil apparent electrical conductivity directed soil sampling, planet scope derived yield maps, and boundary analysis

Abstract

Plant-salt tolerance parameters support decision making directed at conserving water in salt-affected farmland. Traditional methods for determining crop salt tolerance are labor-intensive and often do not accurately represent on-farm growing conditions. This research on dehydrated processing onion evaluates a novel approach utilizing boundary line analysis (BLA) for plant salt tolerance determination. We used soil and plant near-ground and remote sensing to generate extensive datasets (over 3900 yield-salinity data pairs) with limited ground-truth data (20 or fewer samples per field over four onion fields). First, root-zone soil salinity was mapped (R²=0.82) using apparent electrical conductivity through ensemble modeling with field-specific and multi-field regressions. Second, onion yield was mapped (R²=0.82) using a novel time-series analysis of daily PlanetScope surface reflectance. Third, a novel resampling procedure was applied to fit a BLA-based Maas and Hoffman salt tolerance curve to the upper boundary of the salinity × yield data point cloud. The examined onion variety was classified as moderately salt-tolerant, with a soil salinity threshold of 3.05 dS m⁻¹ and a relative yield decrement slope of 2.27 % at higher salinity levels. This proof-of-concept research demonstrated the feasibility of BLA-based crop salt tolerance determination using geospatial soil and plant sensing with limited ground sampling. Crop breeding programs may use the proposed methodology to determine salt tolerance in new crops in realistic environmental and management conditions and potentially reduce the labor required for traditional salt tolerance curve determination.