Search

Extreme precipitation events (EPEs) have garnered considerable social concerns due to their hazardous and destructive nature. To identify the possible causes of EPEs in China, this paper presents an in-depth investigation of how EPEs coincide with atmospheric conditions, i.e., atmospheric instability, moisture availability and moisture convergence, as well as driving variables, i.e., vertical velocity, relative humidity and air temperature. Specifically, the classic coincidence probability is devised to explicitly relate 72-h EPEs to convective available potential energy (CAPE), precipitable water (PW), vertical velocity at 700 hPa (verV), relative humidity at 700 hPa (RH), average air temperature between 850 and 500 hPa (Tavg) as well as air temperature difference between 850 and 500 hPa (Tdiff). The results show that at the annual timescale, EPEs in Southeast and Southwest China are dominantly controlled by verV, in North and Central China by PW and in Northwest China by CAPE roughly. At the seasonal timescale, the spatial distributions of coincidence probability values and "competition" among atmospheric conditions and driving variables exhibit similar patterns as observed throughout the entire year except for December–January–February. Moreover, the diagnostic plots generated for three case study regions in China provide valuable insights into the temporal evolution of precipitation events, cumulative distribution function curves of influential factors and dominant controlling factors of EPEs. This paper contributes to understandings of the dominant controlling factors of EPEs for the whole of China. The spatial patterns of EPEs and their related atmospheric conditions and driving variables yield useful information for rainstorm forecasting and disaster risk management. Plain Language Summary: Atmospheric conditions, i.e., atmospheric instability, moisture availability and wind convergence, and driving variables, i.e., vertical velocity, relative humidity and air temperature, play important parts in the occurrence of extreme precipitation events (EPEs). In this paper, the 72-h EPEs in China are explicitly related to atmospheric conditions and driving variables by using the classic coincidence probability. Specifically, the relationships are measured by the ratio of the time when precipitation and the given factor both reaching at extreme to the whole time in EPEs. Assuming that the factor exhibiting the highest coincidence probability influences EPEs the most, it is observed that at the annual timescale, vertical velocity at 700 hPa has the greatest impact on EPEs in Southeast and Southwest China as well as part of Northeast China, precipitable water in North and Central China and convective available potential energy in Northwest China. Also, it is found that the results at the seasonal timescale are consistent with that throughout the entire year except for winter. Overall, the results of this paper can help us to understand the possible causes of EPEs and serve in rainstorm forecasting and risk management at grid cell level for the whole of China. [ABSTRACT FROM AUTHOR]

Precipitation forecasts generated by global climate models (GCMs) are valuable for environmental modelling and management. At the seasonal timescale, GCMs' forecast skills are generally attributed to low-frequency climate signals, particularly to the internal climate variability mode of El Niño-Southern Oscillation (ENSO). This study proposes a diagnostic method targeting the correlation coefficients for both Niño3.4 index and GCM forecasts in relating them to global precipitation. The similarities versus differences are identified. For the Climate Forecast System version 2 (CFSv2) forecasts, regions with significantly positive anomaly correlation corresponding to significant ENSO teleconnection are highlighted, and regions with significantly positive anomaly correlation in the absence of significant ENSO teleconnection are revealed. As December–January–February (DJF) is the peak season of ENSO, it is estimated that for every 100 grid points with significant ENSO teleconnection, ENSO teleconnection determines significantly positive anomaly correlation for 37 of them; the ratio is respectively 34/100 in March–April–May (MAM), 30/100 in June–July–August (JJA), and 38/100 in September–October–November (SON). It is found that anomaly correlation benefits substantially from ENSO teleconnection over Southern Africa, Central America/Mexico, Amazon and Central North America in DJF, Southeast Asia, North-East Brazil and Central Asia in MAM, Southeast Asia and West Coast South America in JJA, and Southeast Asia, Australia and New Zealand in SON. Limited contributions of ENSO teleconnection to anomaly correlation are observed for North Europe in DJF, West Cost South America and East North America in MAM, North-East Brazil and Central America/Mexico in JJA, and Amazon in SON. Overall, the proposed method contributes to disentangling the sources of GCM forecast skill of seasonal precipitation and fostering better seasonal forecast benchmarking and diagnosis. [ABSTRACT FROM AUTHOR]