气候变化领域集成服务门户

The seasonal and interannual variability of photosynthetically available radiation (PAR) over the tropical oceans is examined using satellite imagery acquired from 1997 to 2017. Spatial and temporal biases between monthly PAR estimates from different instruments are determined and corrected, resulting in a consistent time series over the 20-year record. Uncertainty is evaluated in comparisons with in situ measurements at various sites. Empirical orthogonal function (EOF) analysis is performed with both seasonal and nonseasonal PAR signals, and linear trends are quantified. Seasonal cycles dominate PAR variability, with the first three seasonal EOF modes explaining 84.7% of the total variance. The seasonal patterns are related to solar position and monsoon. Canonical El Nino-Southern Oscillation (ENSO) and Modoki ENSO are related to the two leading nonseasonal EOF modes, with a correlation coefficient of 0.84 between the first mode and the multivariate ENSO index and of 0.48 between the second mode and the El Nino Modoki index. Trend analysis reveals that PAR tends to decrease by 0.2%/year in the central Pacific north of the equator and to increase by 0.2%/year in the central Pacific around 5 degrees S. The tendency is also for PAR to increase west of Central and South America. These changes are consistent with patterns of cloud change evidenced in the satellite cloud record and predicted by global climate models. The long-term satellite PAR data set, together with information of nutrient availability and temperature, enables further studies to elucidate the causes of phytoplankton variability in the tropical oceans.

Plain Language Summary Knowing the spatial and temporal distribution of photosynthetically available radiation (PAR) over the global oceans is critical to understanding biogeochemical cycles of carbon, nutrients, and oxygen and to address important climate issues such as the fate of anthropogenic atmospheric carbon dioxide. This paper examines the seasonal and interannual variability of PAR at the surface of the tropical oceans using 20-year satellite imagery. It is found that the seasonal patterns are related to solar position and monsoon and nonseasonal patterns are affected by El Nino events. The PAR time series exhibits significant trends in some regions due to cloudiness change. Such long-term satellite PAR data set, together with information of nutrient availability and temperature, enables further studies to elucidate the causes of phytoplankton variability (spatial, seasonal, and interannual) in the tropical oceans.