Caconical correlation analysis for the determination of relationship between weather variables and yield components of selected tall coconut (Cocos nucifera L.) varieties in Zamboanga City [Philippines]

The study aimed to determine the relationship between weather variables [X set - amount of rainfall (RF12), number of dry days (DRY12), maximum temperature (TMAX12), minimum temperature (TMIN12), relative humidity (RH12), solar radiation (SR12) 12 months prior, amount of rainfall (RF24), number of dry days (DRY24), maximum temperature (TMAX24), minimum temperature (TMIN24), relative humidity (RH24), and solar radiation (SR24) 24 months prior] and coconut yield components [Y set - number of leaves (NOL), number of bunches (NOB), number of buttons (BUT), number of tennis-ball sized nut (NTB) and total potential yield (TOTNUT)] through canonical correlation analysis (CCA). The study was conducted in Philippine Coconut Authority - Zamboanga Research Center comprising of selected tall coconut varieties namely Laguna Tall, Baybay Tall, Tagnanan Tall, and Rennel Tall. Thirty palms from each identified block were randomly selected, and variable were observed regularly from 2015 to 2016. On the other hand, daily weather information was obtained from the Philippine Atmospheric, Geophysical and Astronomical Services Administration (PAGASA), Canonical correlation coefficients between coconut yield components and weather information during inflorescence opening (12 months prior) and male and female flower development (24 months prior) was done using SAS version 9.4. Results showed that the first canonical correlation coefficient between the first pair of canonical variables was found to be significant at a=10% (r=0.9618, P 0.0001). Findings indicate that NOB, TOTNUT, RF24, DRY24, TMIN24 and RH24 had the largest influence for the explanatory capacity of the first canonical variable estimated from yield components and weather variables, respectively (0.9306, 0.7251, -0.8582, 0.9495, -0.6778 and -0.6496, respectively). Hence, high number of bunches and high total potential yield is associated with low amount of rainfall, more dry days, lower minimum temperature and lower relative humidity during male and female flower development.