Rapid ripeness assessment of fruit after harvest is important to reduce post-harvest losses by sorting fruit according to the duration until they become ready to eat. However, there has been little research on non-destructive estimation of the ripeness and ripening speed of avocado fruit. Unlike previous methods, which classify the ripeness of fruit into a few categories (e.g., unripe and ripe) or indirectly estimate ripeness from its firmness, we developed a method using hyperspectral imaging coupled with deep learning regression to directly estimate the duration until ripeness of Hass avocado fruit. A set of 44,096 sub-images of 551 Hass avocado fruit images was used to train, validate, and test a convolutional neural network (CNN) to predict the number of days until ripeness. Training, validation, and test samples were generated as sub-images of Hass fruit images and were used to train a spectral–spatial residual network to estimate the duration to ripeness. We achieved predictions of duration to ripeness with an average error of 1.17 days per fruit on the test set. A series of experiments demonstrated that our deep learning regression approach outperformed classification approaches that rely on dimensionality reduction techniques such as principal component analysis. Our results show the potential for combining hyperspectral imaging with deep learning to estimate the ripeness stage of fruit, which could help to fine-tune avocado fruit sorting and processing.

Fatty acid composition and mineral nutrient concentrations can affect the nutritional and postharvest properties of fruit and so assessing the chemistry of fresh produce is important for guaranteeing consistent quality throughout the value chain. Current laboratory methods for assessing fruit quality are time-consuming and often destructive. Non-destructive technologies are emerging that predict fruit quality and can minimise postharvest losses, but it may be difficult to develop such technologies for fruit with thick skin. This study aimed to develop laboratory-based hyperspectral imaging methods (400–1000 nm) for predicting proportions of six fatty acids, ratios of saturated and unsaturated fatty acids, and the concentrations of 14 mineral nutrients in Hass avocado fruit from 219 flesh and 194 skin images. Partial least squares regression (PLSR) models predicted the ratio of unsaturated to saturated fatty acids in avocado fruit from both flesh images (R² = 0.79, ratio of prediction to deviation (RPD) = 2.06) and skin images (R² = 0.62, RPD = 1.48). The best-fit models predicted parameters that affect postharvest processing such as the ratio of oleic:linoleic acid from flesh images (R² = 0.67, RPD = 1.63) and the concentrations of boron (B) and calcium (Ca) from flesh images (B: R² = 0.61, RPD = 1.51; Ca: R² = 0.53, RPD = 1.71) and skin images (B: R² = 0.60, RPD = 1.55; Ca: R² = 0.68, RPD = 1.57). Many quality parameters predicted from flesh images could also be predicted from skin images. Hyperspectral imaging represents a promising tool to reduce postharvest losses of avocado fruit by determining internal fruit quality of individual fruit quickly from flesh or skin images.

Tree nuts play an important role in healthy diets, but their economic value and nutritional quality may be affected by their size and paternity. We assessed relationships between nut size and kernel recovery, the incidence of whole kernels, fatty acid composition and mineral nutrient concentrations in three macadamia cultivars, “Daddow”, “816” and “A4”. We determined to what extent differences in nut size and quality were the result of different levels of cross- or self-paternity. Small nuts of all cultivars had lower kernel recovery than large nuts, and small nuts provided lower incidence of whole kernels in “Daddow” and “A4”. Small kernels had a lower relative abundance of the saturated fatty acid, palmitic acid, in all cultivars and higher relative abundance of the unsaturated fatty acid, oleic acid, in “Daddow” and “A4”. Small kernels had higher concentrations of many essential nutrients such as nitrogen and calcium, although potassium concentrations were lower in small kernels. Most nuts arose from cross-pollination. Therefore, nut size and kernel quality were not related to different levels of cross- and self-paternity. Identified cross-paternity was 88%, 78% and 90%, and identified self-paternity was 3%, 2% and 0%, for “Daddow”, “816” and “A4”, respectively. Small macadamia kernels are at least as nutritious as large macadamia kernels. High levels of cross-paternity confirmed that many macadamia cultivars are predominantly outcrossing. Macadamia growers may need to closely inter-plant cultivars and manage beehives to maximise cross-pollination.

Many tree crops experience sub-optimal yields and low fruit quality due to inadequate pollination, low fruit set, and poor crop nutrition. Boron (B) is a critical crop nutrient for fruit set because B levels affect pollen germination and pollen tube growth. However, the relationship between floral B concentration and fruit set is not well understood. The aim of this study was to determine the effect of B applications on the initial fruit set, yield, quality, and paternity of macadamia (Macadamia integrifolia). Cultivar ‘816’ trees received one of three treatments: (a) 0 g, (b) 15 g, or (c) 30 g B per tree prior to flowering. Boron application increased the B concentration of macadamia flowers. Application of 15 g B increased fruit set at 3 weeks after peak anthesis, but this higher initial fruit set was not translated into higher fruit set at 6 or 10 weeks after peak anthesis or higher yield. Boron application increased B concentrations in kernels but did not affect nut-in-shell (NIS) mass, kernel mass, kernel recovery, kernel oil concentration or incidence of whole kernels. Cultivar ‘816’ was highly outcrossing, with 97–98% cross-paternity among kernels from all treatments. Our results indicate that higher B concentration in macadamia flowers can be associated with an increased initial fruit set. However, high B levels did not affect yield, nut quality, or the proportion of self-pollinated fruit at maturity. The heavy dependence on outcrossing highlights the importance of inter-planting different cultivars and managing bee hives to sustain the productivity of macadamia orchards.

Pollination services are critical in entomophilous crops. Relationships between pollinator visits and fruit set and mass are complex and optimal strategies to improve pollination are poorly defined. We often do not understand what proportions of the crop result from self‐ or cross‐pollination, how cross‐pollination affects crop quality and how far cross‐pollen is transported. We quantified proportions of self‐ and cross‐paternity in nuts of the partially self‐incompatible macadamia cultivar 816 at increasing distances from cross‐pollen sources. We estimated the distance of effective pollen movement, evaluated how pollination by different parents affected nut size and nutritional quality, and assessed whether distance from a cross‐pollen source determined proportions of cross‐pollinated nuts, nut size and number of nuts harvested. Almost all nuts resulted from cross‐pollination, even at 184 m (23 rows) from a cross‐pollen source. The cross‐pollen cultivars in closest proximity were usually the pollen parents. Cross‐pollination increased nut size and altered kernel nutritional quality. The number of harvested nuts decreased with increasing distance from a cross‐pollen source in one orchard. Nut size was generally not affected by distance from a cross‐pollen source. Synthesis and applications. Macadamia nut production resulted almost entirely from cross‐pollination. Crop production was dependent on across‐orchard transport of cross pollen, sometimes over at least 184 m. This shows the importance of pollen genotype, not just pollen quantity, in ensuring an effective pollination service. The number of harvested nuts decreased at 24 m from a cross‐pollen source, suggesting that different cultivars should be inter‐planted more closely in macadamia orchards. We need to better understand pollinator behaviour, including what genotype of pollen the different flower visitors carry, and how far and in what direction they transport pollen across orchards, to optimise crop yield and quality.