Shipin Kexue (Feb 2024)

Evaluation of Suitability for Green Tea Processing of Different Tea Cultivars Based on Multivariate Statistical Analysis

  • WANG Huanhuan, YUAN Liping, LEI Zhendong, YANG Shuya, LI Yuchuan, ZHOU Jingtao, QIN Xinxue, YU Zhi, NI Dejiang, CHEN Yuqiong

DOI
https://doi.org/10.7506/spkx1002-6630-20230516-153
Journal volume & issue
Vol. 45, no. 3
pp. 9 – 15

Abstract

Read online

Principal component analysis (PCA) and multiple linear regression analysis (MLRA) were used to analyze and investigate the suitability of green tea for tea tree varieties in different tea cultivars. The fresh leaves of 63 tea cultivars planted in the same garden were processed into roasted green tea by the same method. The sensory and major physicochemical qualities (tea polyphenols, free amino acids, water extracts and chlorophyll) and the color (L*, a* and b*) of green tea were analyzed. To evaluate the suitability of the tested cultivars for green tea processing and its major influential factors, the data obtained were analyzed by PCA and MLRA. The results showed that the coefficient of variation (CV) of chlorophyll content in different cultivars was the highest (26.7%), followed by tea a* value (22.2%), sensory score for tea infusion color (17.1%) and sensory score for tea color (16.2%). A significant correlation was found between the polyphenol content of green tea and its color, tea infusion color and the color of infused tea leaves; between the chlorophyll content, |a*| and b* values of green tea and its color; between the chlorophyll content and |a*| value of green tea and the color of infused tea leaves; and between the |a*| and b* values of green tea and tea infusion color (P < 0.01). The PCA results showed that the contribution rate of the first five principal components (PCs) was 76.895%, and that of the first principal component was 31.918%, mainly pointing to the color quality. According to the evaluation model constructed based on the first five PCs, the top 10 cultivars were Zhongcha 108, Wuniuzao, Pingyangtezao, Mengshan 9, Soubeizhong, Fudingdabai, Jiukeng 16, Fudingdahao, Maolv, and Longjing 43. Using MLRA, the regression function between overall sensory score (Y) and |a*| value (x) was obtained as Y = 68.668 + 5.174x (R2 = 0.313) (P < 0.001). The top 10 varieties determined from this equation were highly consistent with the results of PCA, indicating that a* value is an important indicator for the evaluation of the suitability of tea cultivars for green tea processing.

Keywords