Genotype × environment interactions influence early fruit traits in young Mediterranean sweet chestnut orchards

Technical Summary: Genotype × Environment Interactions in Young Mediterranean Sweet Chestnut Orchards

Problem Statement
Mediterranean chestnut (Castanea sativa Mill.) production faces increasing challenges from climate variability, including rising temperatures, irregular rainfall, and heightened biotic pressure. While previous research has established that genotype × environment (G×E) interactions regulate physiological performance and water relations in chestnut, there is a scarcity of integrated data linking these interactions to early fruit-level traits. Specifically, limited information exists regarding how G×E dynamics influence phenology, early commercial yield, sanitary quality, and kernel nutritional composition during the initial productive years of orchard establishment. Identifying cultivars capable of maintaining stable productivity and quality across contrasting environments is critical for the long-term resilience of chestnut systems, yet current selection criteria often rely on mature orchard data rather than early indicators of adaptation.

Methodology
This study, conducted within the ClimCast experimental network, evaluated four chestnut genotypes across four contrasting Portuguese environments during the 2024 harvest season. The plant material included three traditional cultivars (Judia, Longal, and Martaínha) grafted onto the ColUTAD rootstock, alongside ColUTAD grown as a non-grafted genotype. The four sites—Carrazedo de Montenegro, Parada, Penela da Beira, and Marvão—represented a gradient of altitude and temperature, ranging from 583 to 885 meters above sea level, with mean annual temperatures between 12.5°C and 15.5°C.

The experimental design utilized three biological replicates (trees) per cultivar per site. Data collection followed a sequential workflow:

1. Phenology: Weekly monitoring from June to July 2024 (and October for maturation) using the BBCH scale to track reproductive stages, including catkin emergence, flowering, and bur development.
2. Fruit Production and Quality: Harvests occurred at physiological maturity. Nuts were classified into marketable (well-formed and wrinkled) and non-marketable categories. Defects were categorized as hollow, rotten (fungal decay), or insect-damaged (primarily Cydia spp.). Commercial performance was assessed via marketable yield percentage, sanitary quality (proportion of healthy marketable nuts), and morphological calibre (nuts kg⁻¹).
3. Kernel Composition: Lyophilized samples were analyzed for proximate composition (dry matter, organic matter, crude protein, total fat, total dietary fibre) and carbohydrate profiles (soluble carbohydrates and starch) using standard AOAC and colorimetric methods.
4. Statistical Analysis: Differences among cultivars within sites were evaluated using ANOVA (or non-parametric equivalents where assumptions were violated), with post-hoc tests (Tukey, Games-Howell, or Dunn's) applied at p < 0.05.

Key Results
The study revealed strong G×E interactions affecting nearly all assessed traits, indicating that cultivar performance is highly context-dependent even in young orchards.

• Phenology and Yield: Reproductive development varied significantly by site. While some cultivar-site combinations initiated reproductive development (bur formation), they failed to produce harvestable nuts (specifically Judia at Parada and Marvão, and Martaínha at Penela da Beira). ColUTAD consistently exhibited earlier and more frequent bur formation across sites, likely due to its non-grafted status and more advanced establishment, whereas grafted cultivars showed variable establishment success.
• Commercial and Sanitary Quality: Marketable yield and sanitary quality were not consistently correlated. ColUTAD demonstrated high early productivity and marketable yield (60.7–77.9%) across sites but suffered from significantly poorer sanitary quality, particularly in the warmest environment (Marvão), where rot incidence reached 38.5%. Conversely, traditional cultivars like Martaínha and Judia showed more favorable sanitary profiles at productive sites, with Martaínha achieving high marketability and excellent sanitary quality in Marvão.
• Fruit Size: Morphological calibre and mean nut weight varied inversely and were strongly influenced by the environment. Penela da Beira generally produced smaller nuts, while Carrazedo de Montenegro and Marvão supported larger fruit sizes, particularly for Martaínha and Longal.
• Kernel Composition: Nutritional profiles were highly sensitive to G×E interactions. Protein content varied considerably, with the highest concentrations found in ColUTAD at Penela da Beira (9.38% DM). The greatest variability was observed in carbohydrates; soluble carbohydrate concentrations ranged from 31.6 to 94.8 mg g⁻¹ DM, and starch from 47.0 to 71.0 mg g⁻¹ DM. Notably, ColUTAD at Carrazedo de Montenegro and Marvão exhibited higher fat content (approx. 3.2–3.5% DM) compared to other combinations (1–2% DM). The data suggested that environmental conditions influenced carbon partitioning between starch and soluble sugars rather than merely altering total carbohydrate accumulation.

Significance and Claims
The authors claim that this study provides the first integrated assessment of fruit traits within the ClimCast network, demonstrating that G×E interactions are already expressed during the early productive stage of orchard establishment. The paper asserts that:

1. Early Indicators: Differences in reproductive development, commercial quality, sanitary status, and kernel composition serve as early indicators of cultivar adaptation, appearing well before orchards reach full maturity.
2. No Universal Superiority: There is no universally superior cultivar; instead, distinct adaptive profiles exist. High productivity does not guarantee high sanitary quality or stable kernel composition.
3. Integrated Selection: Cultivar selection for Mediterranean environments must integrate productive, commercial, sanitary, and nutritional traits. Relying solely on yield or commercial classification may overestimate the effective value of a harvest, as seen in cases where high marketable yields were accompanied by significant sanitary losses.
4. Resilience: Understanding these interactions is essential for site-specific cultivar selection to support resilient chestnut production systems under increasing climatic variability.

The paper concludes that while continued monitoring is necessary to confirm the stability of these responses as orchards mature, the current findings validate the utility of fruit-level traits for early-stage cultivar evaluation.