Rhododendron species are typical calcifuges that do not grow well in calcareous soils characterized by alkaline pH and high concentrations of Ca²⁺. In this study, we investigated the effects of three pH levels and a Ca²⁺ treatment on the in vitro germination and seedling growth of three Rhododendron species: R. chihsinianum, R. fortunei, and R. vernicosum. Alkaline pH had no significant effect on germination parameters (g_MAX, mean germination time and germination uniformity) but significantly increased abnormal leaf development (AL) and mortality in all three species. Adding extra Ca²⁺ reduced the mean germination time for R. vernicosum. The negative influence of alkaline pH on seedlings was already visible on the second day of treatment as measured by chlorophyll fluorescence imaging parameters (F_v/F_m and Φ_PSII) on R. fortunei. Extra Ca²⁺ alleviated the negative effect of alkaline pH and increased F_v/F_m 41 days after seed germination in R. fortunei and R. chihsinianum and reduced mortality for all three species. In conclusion, alkaline pH mainly influenced seedling development and growth but not the germination process itself. Chlorophyll fluorescence imaging can be an efficient way to perform high-throughput in vitro screening of Rhododendron seedlings for alkalinity tolerance.

The genus Rhododendron is known for its preference for acidic soils, although some genotypes can tolerate a more neutral or alkaline pH. In this study, a greenhouse experiment was set up for 140 days to examine different parameters to assess pH stress in the progeny of R. fortunei and the cross combination R. ‘Pink Purple Dream’ x ‘Belami’. Additional cultivars ‘Gomer Waterer’ and ‘Cunningham’s White’ were included in the greenhouse test. The plants were divided into two groups. One group was planted in a substrate with a neutral pH (treatment, pH 6.3) and the other group of plants was planted in an acidic pH substrate (control, pH 4.5). Tolerance to pH stress was evaluated for the individual genotypes on both substrates 140 days after the start of the experiment. The following parameters were analyzed: shoot length, root development, chlorophyll fluorescence (Fv/Fm), leaf color and weight (fresh and dry). In intolerant genotypes, all parameters except for number of shoots were negatively affected by pH stress; especially, the development of roots was negatively impacted by the neutral pH, resulting in above-ground symptoms of pH stress, including decreased height and lower fresh and dry weight. The results show variation in pH tolerance within the genotypes tested and point to the potential for the selection of Rhododendron genotypes with improved tolerance to neutral pH.

To investigate interspecific variation in heat tolerance and underlying adaptation mechanisms in Rhododendron, three-year-old potted seedlings of eight taxa, representing four subgenera within the genus Rhododendron, were subjected to 40 °C high-temperature stress. Heat tolerance was comprehensively assessed using phenotypic observation, chlorophyll fluorescence imaging, microscopic examination, and physiological measurements. Results revealed that leaf damage in Rhododendron oldhamii and Rhododendron × pulchrum reached grade III, whereas Rhododendron latoucheae exhibited only grade II injury with rapid recovery. Chlorophyll fluorescence analysis showed a significant decrease in the maximum quantum efficiency of PSII (F_v/F_m) in R. liliiflorum and R. × pulchrum, followed by rapid recovery, while R. latoucheae maintained stable F_v/F_m values. Stomatal closure occurred in all taxa post-stress; stomatal characteristics of R. liliiflorum and R. simiarum remained stable, and leaf tissue structure was least affected in R. kiangsiense. R. × pulchrum demonstrated the most pronounced structural recovery. Physiologically, R. oldhamii exhibited the greatest increases in electrolyte leakage (EL) and malondialdehyde (MDA) content. R. simiarum accumulated the highest proline content under stress, while R. latoucheae showed the most significant proline reduction during recovery. By integrating multiple indicators through principal component analysis (PCA) and a membership function, and assigning weights based on variance contribution, the heat tolerance was comprehensively evaluated and ranked as follows: R. latoucheae > R. simiarum > R. oldhamii > R. ovatum > R. fortunei > R. liliiflorum > R. kiangsiense > R. × pulchrum. These findings demonstrate significant differences in heat tolerance among Rhododendron taxa at the subgenus level, with the subgenus Azaleastrum generally possessing stronger short-term heat tolerance compared to the subgenus Tsutsusi. This study provides a theoretical basis for heat-tolerant cultivar breeding and landscape application of Rhododendron.

To address the low estimation accuracy of deep learning-based crop yield image recognition methods under untrained shooting distances, this study proposes a shooting distance adaptive crop yield estimation method by fusing RGB and depth image information through multi-modal data fusion. Taking strawberry fruit fresh weight as an example, RGB and depth image data of 348 strawberries were collected at nine heights ranging from 70 to 115 cm. First, based on RGB images and shooting height information, a single-modal crop yield estimation model was developed by training a convolutional neural network (CNN) after cropping strawberry fruit images using the relative area conversion method. Second, the height information was expanded into a data matrix matching the RGB image dimensions, and multi-modal fusion models were investigated through input-layer and output-layer fusion strategies. Finally, two additional approaches were explored: direct fusion of RGB and depth images, and extraction of average shooting height from depth images for estimation. The models were tested at two untrained heights (80 cm and 100 cm). Results showed that when using only RGB images and height information, the relative area conversion method achieved the highest accuracy, with R² values of 0.9212 and 0.9304, normalized root mean square error (NRMSE) of 0.0866 and 0.0814, and mean absolute percentage error (MAPE) of 0.0696 and 0.0660 at the two untrained heights. By further incorporating depth data, the highest accuracy was achieved through input-layer fusion of RGB images with extracted average height from depth images, improving R² to 0.9475 and 0.9384, reducing NRMSE to 0.0707 and 0.0766, and lowering MAPE to 0.0591 and 0.0610. Validation using a developed shooting distance adaptive crop yield estimation platform at two random heights yielded MAPE values of 0.0813 and 0.0593. This model enables adaptive crop yield estimation across varying shooting distances, significantly enhancing accuracy under untrained conditions.