An ensemble of regional climate simulations from the Coordinated Regional Down-scaling Experiment in East Asia (CORDEX-East Asia) was analysed to evaluate the ability of 5 regional climate models (RCMs) and their ensemble mean in reproducing the key features of present- day precipitation (1989-2008). We emphasised (1) an extreme rainfall event, (2) seasonal climatology, (3) annual cycles and inter-annual variability and (4) the monsoon characteristics. We highlighted 4 sub-monsoon regions, viz. South Asian Summer Monsoon (SAS), the East Asian Summer Monsoon (EAS), the Western North Pacific Tropical Monsoon (WNP) and the Australian-Maritime Continent Monsoon (AUSMC). We found that the RCMs showed a reasonable performance to capture the extreme rainfall event in 1998. The RCMs simulated the seasonal mean, annual cycle and inter-annual variability acceptably. However, individual models exhibited significant biases in some sub-regions and seasons. Moreover, most of the RCMs significantly improved their performance in capturing precipitation climatology and monsoon characteristics over the Korean Peninsula, the Korea Strait and southern Japan. Based upon this performance study, we conclude that the present set of RCMs from CORDEX can be used to provide useful information on climate projections over East Asia.

The Indian Monsoon Variability during the past Millennium has been simulated with the ECHAM5 model in two different time slices: Medieval Climate Anomaly and the Little Ice Age. The simulations are compared with new centennial-resolving paleo-reconstructions inferred from various well-dated multi-proxies from two core regions, the Himalaya and Central India. A qualitative moisture index is derived from the proxies and compared with simulated moisture anomalies.

Abstract. The general circulation model ECHAM5 has been used to simulate the Indian monsoon and its variability during the Medieval Warm Period (MWP; 900–1100 AD), the Little Ice Age (LIA; 1515–1715 AD) and for recent climate (REC; 1800–2000 AD). The focus is on the analysis of external drivers and internal feedbacks leading to extreme rainfall events over India from interannual to multidecadal time scale. An evaluation of spatiotemporal monsoon patterns with present-day observation data is in agreement with other state-of-the-art monsoon modeling studies. The simulated monsoon intensity on multidecadal time scale is weakened (enhanced) in summer (winter) due to colder (warmer) SSTs in the Indian Ocean. Variations in solar insolation are the main drivers for these SST anomalies, verified by very strong temporal anticorrelations between Total Solar Irradiance and All-India-Monsoon-Rainfall in summer monsoon months. The external solar forcing is coupled and overlain by internal climate modes of the ocean (ENSO and IOD) with asynchronous intensities and lengths of periods. In addition, the model simulations have been compared with a relative moisture index derived from paleoclimatic reconstructions based on various proxies and archives in India. In this context, the Lonar record in Central India has been highlighted and evaluated the first time. The simulated relative annual rainfall anomalies in comparison to present-day climate are in agreement (disagreement) with the reconstructed moisture index for MWP (LIA) climate. In order to investigate the interannual monsoon variability with respect to monsoon failures, dry summer monsoon composites for 30-yr-long periods of MWP, LIA and REC have been further analysed. Within dry years of LIA, the summer rainfall over India and surrounding oceans is less than in MWP indicating stronger drying conditions due to a stronger summer solar insolation forcing coupled with variations in ENSO. To quantify the ECHAM5 simulated long-term drought conditions within Monsoon Asia, the Palmer Drought Severity Index has been additionally estimated for recent climate showing strong pattern correlation between global SST anomalies and EOF variability signal of the drought index, whereas the temporal relationship is weak.