关键词: 无患子, 微域环境因子, 果实性状, 回归模型

Abstract:

【Objective】Fruit yield and quality characteristics are closely related to the canopy micro-environment, but the specific relationships have been unclear. Analyzing the relationship between micro-environmental factors and fruit characteristics can reveal the effect of the canopy micro-environmental factors on Sapindus mukorossi fruit characteristics, and provide a theoretical basis and technical information for optimization of canopy structure and the construction of favorable micro-environments. 【Method】In this study, in Jianning County of Fujian Province, 10-year-old S. mukorossi trees with a stable fruiting stage were chosen as test samples. The whole of the natural round-head canopy was divided into 144 small measuring areas. During the whole growing period from May to October in 2017, the canopy micro-environmental factors were measured monthly in these areas. The fruit yield and quality characteristics were determined immediately after harvest. Based on the measured values of micro-environmental factors and fruit yield in different parts of the canopy, 3D distribution maps were drawn. According to the distance from the trunk and the vertical stratification of the canopy, the small measuring areas were grouped into nine different canopy areas, and the significant differences in micro-environmental factors and fruit characteristics were analyzed. The correlations between micro-environmental factors and fruit characteristics in different vertical layers were also analyzed, and the regression models of fruit characteristics were established. Finally, the micro-environmental conditions for the high fruit yield and quality were assessed. 【Result】① Within the canopy from inside to outside and from bottom to top, the average temperature in the growing season increased by 0.3 and 1.5 ℃, respectively; the average photosynthetically active radiation of the growing season increased by 330.76 and 463.53 μmol/(m ²·s), respectively; the average relative humidity of growing season decreased by 1.1% and 3.0%, respectively. The ratio of fruit yield of the lower, middle and upper layers was 8:17:75, and that of the inner, middle and peripheral layers was 24:32:44. The differences in the micro-environment and fruit yield in the horizontal direction were obviously smaller than that in the vertical direction. Fruits were mainly distributed in the periphery and upper layer. ② There were significant differences in the fruit number, fresh fruit yield, dry fruit yield, the water content of the fruit, single dried fruit weight and oil yield in nine canopy areas, but there were no significant differences in the dried aril content ratio, the dried seed content ratio, the kernel content of the dried seed, and the oil content of the kernel. ③ Fruit yields, the kernel content of the dried seed, and the oil yield all had significant positive correlations with temperature and photosynthetically active radiation, but a significant negative correlation with relative humidity. Photosynthetically active radiation had the greatest impact on fruit yield and quality, followed by temperature and relative humidity. ④ From the single factor regression models, it was deduced that suitable micro-environmental conditions for high fruit yield and quality are: temperature between 28-30 ℃, relative humidity no more than 74%, and photosynthetically active radiation between 830-1 203 μmol/(m ²·s). ⑤Multivariate regression models were suitable for the S. mukorossi fruit and oil yield prediction with a natural round-head canopy. 【Conclusion】The synergistic effect of micro-environmental factors greatly influences fruit yield and quality characteristics of S. mukorossi. The fruit characteristics in the central and lower layers are strongly affected by the micro-environment, so the improvement of the micro-environment of the middle and lower layers through horticultural measures based on the canopy structures and fruit-bearing habits can improve fruit characteristics.

Key words: Sapindus mukorossi, micro-environmental factors, fruit characteristics, regression models