Precipitation controls isoprene emissions from tropical ecosystems

Isoprene emissions from tropical regions account for a majority of isoprene produced globally. Current estimates of global isoprene emissions use meteorological inputs (temperature and light), ecosystem leaf area, and a time invariant, ecosystem specific emissions factor. This approach has been verified to work well for deciduous mid-latitude forests, but the approach has not been tested for tropical ecosystems where seasonality is induced by precipitation. Recent flux studies at two field stations in the tropics found strong effects of precipitation regime (dry vs. wet season) on isoprene emissions. A flux study conducted during the wet season (October 1999) at the La Selva Biological Station (10° 26' N, 83° 59' W, precipitation 4000 mm yr{^-1}) found whole system isoprene emissions rates between 2--10 mg C m^-2 h^-1, while a second campaign during the dry season (April 2003) found values ranging 8--16 mg C m^-2 h^-1. This difference could not be explained by changes in ambient temperature or light using established emissions algorithms. The second field site near Santarém, Brazil in the Floresta Nacional do Tapajós (2° 51' S, 54° 58' W, precipitation 2000 mm yr{^-1}), part of the Large scale Biosphere-atmosphere experiment in Amazônia (LBA), showed a similar pattern. Additionally, a 13 month isoprene concentration record at this station found a 4 fold increase during the dry season. Application of a one dimensional chemistry model predicts a similar change in isoprene source strength. A standard emission model using temperature and light could not account for these seasonal changes, but adding an empirical term that accounted for previous precipitation greatly enhanced the fit.