Strawberry (Fragaria × ananassa) production in the United States is a $2.5 billion industry, traditionally dominated by field cultivation. Controlled Environment Agriculture (CEA) is emerging as a promising alternative, offering year-round production and greater control over growing conditions. Despite its potential, strawberry cultivation in CEA systems remains cost-intensive, primarily due to high labor requirements. Additionally, strawberry production follows a cyclical pattern, with fruit developing in discrete peaks known as flushes. These fluctuations present challenges for consistent resource management, labor planning, and market supply, highlighting the need for predictive tools to optimize production efficiency. Our primary research objective is to develop a strawberry growers’ decision support tool for crop management through yield prediction based on flower mapping, a method of describing floral developmental stages through meristem dissection. Using a soilless hanging gutter system designed to mimic a commercial greenhouse production system, we grew a widely used cultivar Albion. The greenhouse maintained average daytime and nighttime air temperatures of 22.5 ± 3.1°C and 18.2 ± 2.8°C, respectively. Daily light integral (DLI) averaged 20.0 ± 3.0 mol·m⁻²·d⁻¹, daytime CO₂ concentration averaged 580 ± 207 ppm, and vapor pressure deficit (VPD) averaged 1.0 ± 0.6 kPa. Supplemental lighting provided a 16-hour photoperiod with a photosynthetic photon flux density (PPFD) of ~250 μmol·m⁻²·s⁻¹. Plants were fertigated through a drip irrigation system and grown in a commercial strawberry substrate composed of 100% coconut coir fiber. We performed weekly flower mapping on randomly sampled plants and yield measurements for the rest of plants for 19 weeks. We hypothesized that yield of a future week can be predicted based on counts of floral buds at each of 11 developmental stages. We found that floral meristem stages 4 and 5 (when calyx and trichomes differentiate on the floral bud) exhibit significant positive correlations with yield occurring nine weeks later. In addition, stage 11 meristems (anthesis) showed a significant positive correlation with yield occurring three weeks later. The remaining developmental stages exhibited weaker correlations and were less reliable predictors of upcoming yield. By using these key developmental stages, we will develop a methodology for forecasting near-future yield. This will help U.S. greenhouse strawberry growers to make informed decisions about resource allocation, labor scheduling, and market planning, ultimately optimizing yield and production efficiency in CEA systems. Our research outcomes lay the groundwork for more comprehensive yield predictions in the future.
