Executive Summary

In the second year of the 6th phase of the Asia-Pacific project (2022-2024), non-linear composite analyses were performed to identify the detailed physical-dynamic processes in which major predictors affect the climate variability of the Korean Peninsula, and to use the results to develop objective and story-based forecasting methods. And for the optimal integration of various observation-based prediction information, a method for selecting the importance of predictors based on the reproducibility of detailed dynamic processes of the predictors was proposed. In addition, considering the impact of climate change, future renewable energy and drought characteristics changes on the Korean Peninsula were projected and their causes were analyzed. Lastly, in order to save resources and improve efficiency in analysis work, the monitoring processes for 3-month climate prediction were automated, and an advanced system for composite analysis and impact analysis on climate variability in Korean Peninsula was established.

In the analysis study of detailed dynamic process of summer predictors, a framework that can reveal the detailed physical-dynamic process of predictors was developed and applied to the temperature predictors in July. In the case of the positive phase of March European Z500 predictor, the overlapping of the subtropical/mid-latitude wave propagation resembling CGT pattern related to the South Asian monsoon activity in June and the east-west wave propagation from the Atlantic Ocean to East Asia seems to strengthen the barotropic high pressure in the Korean Peninsula and causes high temperatures. It was found that in the case of the positive phase, monitoring the strong South Asian monsoon index in June increases the reliability of the prediction and that in the negative phase, the large amount of snow in Central Asia in April and the low SST in the Gulf of Mexico in June increase the reliability of the prediction. In the case of the positive phase of April tropical SST tripole, after the high temperature of SST in the Philippine Sea is maintained until May in the situation where the Central Pacific type La Niña developed, low pressure and convection develops in the northern Philippine Sea in June-July, resulting in a positive P-J pattern. It was analyzed that the resulted high pressure in northeastern East Asia tends to cause high temperatures in the Korean Peninsula. It was found that prediction reliability could be increased by monitoring the strong South Asian monsoon index in June in the case of a positive phase and the low tropical SST tripole in May in the case of a negative phase. As such, the possibility of discriminating the importance of various predictive information was confirmed. As a result, in the case of the March European Z500 and the April tropical SST tripole, it was confirmed that the prediction reliability improved when the weight was assigned to the prediction of the factor according to the tendency of the monitoring factors. In the case of operational prediction, a method is suggested to determine the importance of the predictor by estimating the reproducibility of the monitoring factor using the model prediction result.

A significant positive relationship is found between the ENSO and Korean temperature and precipitation anomalies in December. However, the relationship is distinctively broken in January, so it is somewhat difficult to predict the Korean climate in January. The North Pacific Oscillation (NPO) in December is responsible for significant warm anomalies over Korean Peninsula that peak in January, exhibiting a 1-month leading role. The NPO-related anticylconic anomalies in December may affect the precipitation decrease in the western North Pacific, then the East Asian anticylconic circulation remains until January through the Rossby wave propagation. Additionally, as a result of NPO, the East Asian anticyclonic anomalies accompany the easterlies on the southern side in December modulate the Jet weakening. The SST anomalies near East Asia tend to increase in January due to the wind-evaporation-SST feedback, and the resultant warm SST is favorable for warm Korean Peninsula in January. The December NPO index as a new precursor may help us to better understand and predict the temperature variability over Korea in January with a 1-month lag, and shows an overall improved predictability than that of the Nino3.4 index.

For investigating the impact of the climate change on the wind and solar power generation potentials over South Korea considering ensemble projections from downscaled high-resolution bias-corrected future climate change scenario data taken from the National Institute of Meteorological Science in Korea. Under future global warming, solar power potentials over South Korea were projected to decrease in spring (March-May) and winter (December-February) seasons relative to present climate in the late 21st century (2081-2100), particularly showing the relatively large decrease in the northern part of South Korea. The decrease tendency was more significant and larger in the high-CO2 emission scenario (SSP5-8.5) than low-CO2 emission scenario (SSP2-4.5). The projected decrease in solar power potential in spring was mainly due to increased air temperature by future global warming and the decrease in winter was attributable to the projected increase in the air temperature and the decrease in solar radiation at the surface. Wind power potentials which were estimated with the wind energy density was generally projected to be decreased with future global warming in all seasons except for summer. This decrease tendency was also larger in the late 21st century of the SSP5-8.5 scenario, especially over the southern part of South Korea in winter and spring and over the northern part in fall. These results may help optimize the regional renewable energy generation system development and plans.

For analyzing the future changes in drought indices for regions in Korean Peninsula, we used the Standardized Precipitation Index (SPI), widely used as a meteorological drought index, and the Standardized Evapotranspiration Deficit Index (SEDI), used as an agricultural drought index. SPI calculates the drought index based only on precipitation data and does not reflect the impact of continued temperature rises. Therefore, using SEDI, which incorporates surface temperature, wind, net radiation, humidity, and precipitation, we analyzed the future changes and causes of drought indices in Korean Peninsula. Comparing historical periods(1985~2014), the results suggest that SEDI forecasts a more severe drought situation in Korean Peninsula compared to SPI, especially during autumn and in high carbon emission scenario(SSP585). We investigated the intensification factors behind future drought indices in Korean Peninsula. During spring, although precipitation increases more than the present, the steep rise in temperature leads to a higher demand for evapotranspiration, surpassing atmospheric evaporation, resulting in more severe drought than the present. In autumn, the weakening of the southern winds entering Korean Peninsula leads to reduced precipitation. Furthermore, higher net radiation and temperature compared to the present are anticipated, indicating a more severe drought than during spring.

Due to frequent extreme climate events around the world, the need to quickly establish a climate monitoring system is increasing. It has become necessary to collect the latest observation data and establish a periodic extreme climate monitoring system in an effort to minimize damage to property and human life through continuous monitoring of increasingly severe extreme weather phenomena around the world. To build the system, a climate monitoring service and climate analysis service were developed based on observation data provided by NCEP and the Korea Meteorological Administration. The climate monitoring service provides up-to-date information on various climate variables, and the climate analysis service not only synthesizes data or provides results of time series analysis, but also standardizes the data collection system for quick response to abnormal climate. In addition, in order to provide rapid monitoring results, an automatic prediction factor production and information provision system and a time series-based synthetic information provision function were additionally built for user convenience. By automatically calculating and providing monthly related climate factors for monthly forecast discussions, work efficiency can be further improved.