The Role of Sea Surface Temperature in Shaping the Characteristics of Future Convective Afternoon Rainfall in Taiwan

Observational data reveals a remote influence of eastern Pacific SST changes on Taiwan's summer diurnal rainfall. The question arises: how will this influence evolve in the future? To address this, we employed dynamical downscaling projections with varied sea surface temperature patterns.
The Role of Sea Surface Temperature in Shaping the Characteristics of Future Convective Afternoon Rainfall in Taiwan

Diurnal Rainfall Characteristics in Taiwan: Past

During Taiwan's summer, diurnal rainfall typically occurs in the late afternoon, with the peak observed between 14:00 and 17:00 local standard time (Fig. 1). These convective afternoon rainfall (CAR) events contribute significantly, making up over 40~50% of the total summer rainfall in Taiwan. Previous research has identified local and large-scale atmospheric thermodynamic changes, including decreased daytime wind convergence, increased thermal stability, and enhanced moisture supply, as key factors influencing CAR activity in Taiwan.

Huang et al. (2019), utilizing observational data, demonstrated that variations in sea surface temperature (SST) over the eastern tropical Pacific correlate with the frequency of CAR in Taiwan. Cooler eastern tropical Pacific SST leads to more frequent CAR, while warmer SST results in less frequent occurrences, highlighting the influence of the El Niño Southern Oscillation (ENSO) on local weather. In our recent study published in npj Climate and Atmospheric Science, we explored the projected uncertainties in summer CAR in Taiwan related to SST changes. This investigation utilized a dynamical downscaling approach, with projections driven by four different SST categories derived from 28 CMIP5 model simulations under the RCP8.5 high emission scenario.

Fig. 1 Location and the diurnal rainfall variability of Taiwan

Figure 1 .  Location and Diurnal Rainfall Variability in Taiwan. (a) The positioning of Taiwan within the regional domain employed in the HiRAM-WRF model simulations, overlaid with topography (shaded) and 925 hPa wind circulation (vectors). (b) Diurnal Rainfall Variability (DRV) assessed using gridded rain-gauge observations (OBS). (c) The peak timing phase of the diurnal evolution of rainfall, estimated from OBS, presented in local standard time (LST).

Diurnal Rainfall Characteristics in Taiwan: Future

All future projections, dynamically downscaled using the Weather Research and Forecasting model (WRF) driven by the HiRAM global model, indicate that CAR events in Taiwan will become less frequent but more intense in the future. Nevertheless, substantial differences in the magnitude of CAR changes were observed among simulations with varying degrees of eastern tropical Pacific SST warming (Fig. 2). The La Niña-like SST pattern (FC1) projects a higher CAR frequency than other SST warming patterns. In contrast, the El Niño-like SST pattern (FC2) shows the least intense CAR among the future runs.

In response to these changes in CAR frequency and intensity, FC1 exhibits the highest, while FC2 exhibits the smallest CAR amount in Taiwan. These findings underscore the potential impact of the ENSO in modulating changes in CAR activities over Taiwan under global warming conditions, where La Niña-like circulation patterns induce more CAR activities over Taiwan than El Niño-like circulation patterns. These projected results further emphasize the importance of understanding the underlying physical mechanisms that influence CAR event characteristics, a crucial aspect for enhancing the accuracy of CAR event projections in Taiwan.

Projected mean sea surface temperature (SST) changes and CAR characteristic

Figure 2.  Projected changes in global SST patterns and CAR characteristics in Taiwan. (a) Projected SST changes, estimated by calculating the differences between the future projection using the FC0 run (ensemble mean of the 28 CMIP5 models) and the present-day simulation (Pday). (b) to (d) illustrate the differences between FC0 and the other three future runs: (b) FC1-FC0, (c) FC2-FC0, and (d) FC3-FC0. (e) Demonstrates the respective percentage differences in the area-averaged CAR activities over Taiwan for CAR frequency (denoted as FC1 - FC0 (purple bar), FC2 - FC0 (green bar), and FC3 - FC0 (pink bar)), estimated from HiRAM-WRF simulations. (f) and (g) follow a similar format to (e), but represent CAR intensity and CAR amount, respectively.

Possible Mechanisms for Projected Differences in CAR Activities

Upon investigating the potential causes and associated mechanisms for the observed changes, it became evident that the variations in the projected differences in CAR activities can be ascribed to differences in daytime thermal instability, local inland wind convergence, and moisture flux convergence over Taiwan. The study establishes a link between these local thermodynamic changes and regional circulation changes.

We also identified that vertically integrated moisture flux convergence plays a more crucial role than changes in local moisture in explaining the projected changes in CAR intensity. Additionally, we discussed the connection between regional circulation changes and global circulation changes driven by the four SST patterns, highlighting the potential role of the ENSO in shaping the characteristics of future CAR activities in Taiwan.

Our paper provides a detailed discussion of these findings, which hold significant implications for informed decision-making in hydrology and climate-related policies, contributing to better planning for the future.


  • Huang, W.-R., Chien, Y.-T., Cheng, C.-T. et al. The role of sea surface temperature in shaping the characteristics of future convective afternoon rainfall in Taiwan. npj Clim. Atmos. Sci. 6, 198 (2023).
  • Huang, W.-R., Chang, Y.-H. & Huang, P.-H. Relationship between the Interannual Variations of Summer Convective Afternoon Rainfall Activity in Taiwan and SSTA(Nino3.4) during 1961-2012: Characteristics and Mechanisms. Sci. Rep. 9, 9378, (2019).

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in

Subscribe to the Topic

Earth and Environmental Sciences
Physical Sciences > Earth and Environmental Sciences

Related Collections

With collections, you can get published faster and increase your visibility.

Digital Paleoclimate: Integration, Simulation, and Assimilation

Paleoclimate studies towards a digital paradigm by using paleoclimate records integration, model simulation, and data assimilation for promoting our understanding of climate dynamics and future prediction.

Publishing Model: Open Access

Deadline: Sep 30, 2024