Skip to main content
ARS Home » Southeast Area » Fort Pierce, Florida » U.S. Horticultural Research Laboratory » Citrus and Other Subtropical Products Research » Research » Publications at this Location » Publication #400106

Research Project: Determination of Flavor and Healthful Benefits of Florida-Grown Fruits and Vegetables and Development of Postharvest Treatments to Optimize Shelf Life an Quality for Their Fresh and Processed Products

Location: Citrus and Other Subtropical Products Research

Title: Dynamic prediction of fruit quality traits as a function of environmental and genetic factors

item HOPF, ALVIN - University Of Florida
item Plotto, Anne
item RIZWAN, RAFIQ - University Of Florida
item ZHANG, CONGMU - University Of Florida
item BOOTE, KENNETH - University Of Florida
item SHELIA, VAKHTANG - University Of Florida
item HOOGENBOOM, GERRIT - University Of Florida

Submitted to: Acta Horticulturae
Publication Type: Proceedings
Publication Acceptance Date: 11/16/2022
Publication Date: 12/31/2022
Citation: Hopf, A., Plotto, A., Rizwan, R., Zhang, C., Boote, K., Shelia, V., Hoogenboom, G. 2022. Dynamic prediction of fruit quality traits as a function of environmental and genetic factors. Acta Horticulturae. 1353:145-152.

Interpretive Summary: Fruit quality is influenced by many factors, and starts in the field soon after bloom and during fruit growth. Environmental conditions (geographic location and weather), horticultural practices and postharvest management strongly effect fruit quality, in addition to the inherent characteristics due to genotype. This work presents an overview of a proposed set of methodologies to assess and model the variability of fruit quality using process-based crop statistical models. Four case studies show the application on fruit quality datasets for strawberry, blueberry, grapevine and grapefruit.

Technical Abstract: Fruit quality is a complex trait affected by interactions among horticultural management, genotype, and environment. Growers and the horticulture value-chain are challenged by the increasing importance of quality aspects driven by consumer demand. Process-based crop growth models and other data-driven decision support tools are actively used to increase and improve fruit production, but further work is required to include fruit quality predictions in this process. This work presents an overview of proposed methodologies to assess and model the variability of fruit quality and the possible integration with process-based crop models. Four case studies provided here show the application on fruit quality datasets for strawberry (Fragaria ×ananassa), blueberry (Vaccinium spp.), grapevine (Vitis vinifera) and grapefruit (Citrus paradisii MacFad.), production in the USA and Pakistan. Statistical relations between measured fruit quality traits and both observed and simulated pre-harvest conditions were revealed by regression analysis. A quality prediction model was built by identifying and integrating the key correlations into a quality prediction module. For strawberry, the module can predict soluble solid content and titratable acidity based on average temperature during key phenological phases, e.g., individual fruit growth from end of flowering to harvest maturity. The quality prediction model was integrated with the CROPGRO-Strawberry model of the Decision Support System for Agrotechnology Transfer (DSSAT) as a module to predict both quality and quantity dynamics of strawberry production. A strategic analysis with historic weather data was conducted to reveal the impact of seasonal climate variability on strawberry yield and quality. For grapefruit, a similar correlation between temperature and the ratio of soluble solids to titratable acidity was modeled and extended to include post-harvest quality development for combined harvest and storage recommendations. Overall, the proposed quality prediction methodology is an important advancement towards the objective assessment of genotype by environment by management effects on fruit quality, particularly when analyzing data from multiple sites or years. Future work will include more refined, e.g., multivariate, statistical methods or inclusion of process-based approaches based on the availability of suitable data.