ENSO and PDO strongly influence Taiwan spruce height growth

Abstract

To assess the influences of the El Niño-Southern Oscillation (ENSO) and the Pacific Decadal Oscillation (PDO) on the mean annual height growth of the Taiwan spruce (Picea morrisonicola), we constructed two height growth indices (HGI1 and HGI2) using an ensemble empirical mode decomposition approach based on stem analysis of 19 trees. The most significant periodicities for HGI1 and HGI2 were 4.5 and 15.5years, respectively. Both indices positively correlated with the mean annual height growth and reflected the influences of local thermal environment variability at various lag times, up to 3years prior to the shoot extension season. The lagged relationships were likely caused by a combination of the species’ two-year height growth process, the delayed responses of the local climate to sea surface temperature (SST) variation, and persistence due to needle lifespan. The correlations between HGI1 and a detrended SST field showed a spatial pattern similar to that of the ENSO. At lags of both 3 and 4years, the correlations between HGI2 and the SST field revealed a spatial pattern resembling that of PDO. When leading by 3 and 4years, respectively, the NINO4 and PDO indices positively associated with the mean annual height growth and together explained 37% of the variance; the marginal effect of the NINO4 index was twice as large as that of the PDO index. The results show that the height growth of Taiwan spruce responded to both ENSO and PDO in a positive manner. This study demonstrates that quasi-periodic climate variation on an interannual (ENSO) to interdecadal (PDO) time scale can significantly influence tree growth and should be taken into account when assessing the impact of climate changes on forest productivity, and the results of this study provide a basis for incorporating the influences of such quasi-periodic climate variations into future model-based assessments.