Development and improvement of the CROPGRO-Strawberry model

Abstract Strawberry is a high-value horticultural crop with a global market and it has a strong regional importance in production areas such as Florida. Strawberry growers face many challenges related to weather, cultivation, and markets. Decision support tools can help optimize strawberry production but require sound models or other predictive tools as a foundation. The goal of this study was to improve a new CROPGRO-Strawberry model in the Decision Support System for Agrotechnology Transfer (DSSAT) using experimental data with observations from different seasons and multiple cultivars. Model improvements were made in three primary areas: 1) cardinal temperatures for different development processes, 2) a module for the dynamic assimilate partitioning based on photothermal age, and 3) improvement of the cultivar and ecotype coefficients. The model predicts the growth, development, and fruit production of strawberry over time using weather, soil, management and physiological parameters as inputs. Overall, the results show a good simulation of development and growth (d = 0.77 to 0.99, RRMSE = 0.26 to 0.68), but an overestimation of vegetative growth during the early season. Periodic and cumulative fruit harvests were well-simulated (d = 0.91, RRMSE 0.47), capturing the seasonal dynamics and representing differences among cultivars and harvest intervals. A strategic analysis showed the applicability of the crop model to understand and manage the impact of seasonal climate variability on total strawberry yield as well as the distribution of fruit production across different harvest months. Future work should include improvements in the simulation of vegetative growth and evaluation against new datasets representing different production environments and cultivars. The improved CROPGRO-Strawberry model will enable future work on applications to other specialty crops where modeling of continuous fruiting and multiple harvests of individual fruit instead of one end-of-season harvest is desired.