Phosphorus Rate Optimization for Snap Bean on Florida’s Sandy Soils: A Multi-Year Linear–Plateau Analysis

Phosphorus availability is extremely limited in Florida’s sandy soils due to intense sorption by aluminum (Al), iron (Fe) oxides, and fertilizer retention. Current fertilization recommendations do not account for P-fixation, a defining characteristic of Florida’s soils. Site-specific and multi-year yield-based thresholds for snap bean under these conditions have not been established. This study is among the first to derive yield-based thresholds from a multi-year linear–plateau model using nonlinear mixed-effects modeling that accounts for stochastic variability across sites and years, thereby defining a threshold range for this crop in this soil system. This work assessed snap bean (Phaseolus vulgaris L.) pod yield responses to phosphorus fertilization from 2022 to 2025. Field experiments employing increasing P2O5 rates and fertilizer sources were conducted. Hastings and Citra were selected to represent sandy soil conditions across northeast and north-central Florida’s commercial snap bean production areas, where soil tests consistently indicated elevated extractable Al and Fe in the rhizosphere, key drivers of P fixation and fertilizer demand. At low-to-moderate P2O5 rates, yield increased linearly over site-years before plateauing. A breakpoint of 215.6 kg ha−1 P2O5 was found in Hastings by the multi-year model. A single-year fit at Citra in 2025 revealed a breakpoint of 265.7 kg ha−1 P2O5. Confidence intervals were wide due to year and plot variability, with values of 148.2–283 kg ha−1 P2O5. When all site-years were pooled, the population-level breakpoint was estimated at 223.5 kg ha−1 P2O5, with 90% and 95% model estimates of maximum yield obtained at about 164 and 194 kg ha−1 P2O5, respectively. These findings provide a fertilizer range for snap bean production in Florida’s sandy soils under similar conditions, with implications for regional fertilizer guidelines.