Global nonlinear and nonstationary climate change effects on regional precipitation and forest phenology in Panama, Central America

The inherent effects of global sea surface temperature (SST) anomalies on hydrological cycle and vegetation cover complicate the structure of tropical climate at the regional scale. Assessing hydrological processes related to climate forcing is important in Central America because it is surrounded by both the Pacific and Atlantic oceans and two continental landmasses. In this study, the use of high-resolution remote sensing imagery in wavelet analysis helps identify nonstationary characteristics of hydrological and ecological responses. The wavelet-based empirical orthogonal function (WEOF) further reflects the nonlinear relationship between the Atlantic and Pacific SST and the greenness of a pristine forested site in Panama, La Amistad International Park. Integrated WEOF and descriptive statistics for data analysis reveal a higher temporal variability in terrestrial precipitation relative to in situ land surface temperature and its probable effects on the presence of dry periods. Such teleconnection signals of SST were identified as a driving force of decline in tropical forest greenness during dry periods. The results of our remote sensing-based wavelet analysis showed intra-annual high-frequency and biennial to triennial low-frequency signals between enhanced vegetation index/precipitation datasets and SST indices in both Atlantic and Pacific oceans. A spatiotemporal priority search further confirmed the importance of the effects of the El Niño-Southern Oscillation (ENSO) over terrestrial responses in the selected study site. Coincidence of the effect of ENSO teleconnection patterns on precipitation and vegetation suggests possible impacts of El Niño-associated droughts in Central America, accompanied by reduced rainfall, especially during the first months of rainy season (June, July, and August), and decline in vegetation cover during the dry season (March and April).