Change of tropical cyclone and seasonal climate state in a global warming experiment with a global cloud-system-resolving model

Abstract

Recent increase in computation power allows a use of high-resolution global model to investigate possible future change of tropical cyclones. In this chapter, we propose a new approach here to pursue the issue: the use of global cloud-system resolving model (GCRM).1 It is the model designed with the clear aim of resolving cloud cluster, an essential component of tropical cyclogenesis, and therefore expected to provide a new result for the projection of future change of tropical cyclone. This chapter highlights our first attempt of time-slice prediction of future tropical cyclone (TC) under a global warming condition and associated climate change of season-long period with 14-km mesh version of Nonhydrostatic ICosahedral Atmospheric Model (NICAM), a prototype GCRM. Notable changes in the seasonal-mean state include decrease in outgoing longwave radiation (OLR) in the western to central Pacific, in particular east of the maritime continent, and increase in OLR and decrease in outgoing shortwave radiation in the mid and high latitudes. The former (the latter) is associated with an enhanced activity of precipitation (decrease in cloud amount) over the region. Global frequency of TCs in the seasonal period is projected to decrease, in agreement with the general statement in the Intergovernmental Panel on Climate Change Fourth Assessment Report. On a regional basis, the frequency decreases over the North Atlantic, and remains almost unchanged in the western Pacific. The tendency of global frequency is found to be insensitive to detection threshold of the surface wind speed in the tropical cyclone tracking algorithm. The control experiment is designed for the particular year of 2004, which had a more El Nino flavor than normal year; therefore some caution is necessary in interpreting the results for this particular choice of experimental design.