• Journal of Information Science Theory and Practice
• /
• v.10 no.spc
• /
• pp.135-142
• /
• 2022

Recently, with the development of data processing technology and the increase of computational power, methods to solving social problems using Artificial Intelligence (AI) are in the spotlight, and AI technologies are replacing and supplementing existing traditional methods in various fields. Meanwhile in Korea, heavy rain is one of the representative factors of natural disasters that cause enormous economic damage and casualties every year. Accurate prediction of heavy rainfall over the Korean peninsula is very difficult due to its geographical features, located between the Eurasian continent and the Pacific Ocean at mid-latitude, and the influence of the summer monsoon. In order to deal with such problems, the Korea Meteorological Administration operates various state-of-the-art observation equipment and a newly developed global atmospheric model system. Nevertheless, for precipitation nowcasting, the use of a separate system based on the extrapolation method is required due to the intrinsic characteristics associated with the operation of numerical weather prediction models. The predictability of existing precipitation nowcasting is reliable in the early stage of forecasting but decreases sharply as forecast lead time increases. At this point, AI technologies to deal with spatio-temporal features of data are expected to greatly contribute to overcoming the limitations of existing precipitation nowcasting systems. Thus, in this project the dataset required to develop, train, and verify deep learning-based precipitation nowcasting models has been constructed in a regularized form. The dataset not only provides various variables obtained from multiple sources, but also coincides with each other in spatio-temporal specifications.

• Journal of the Korean earth science society
• /
• v.43 no.6
• /
• pp.712-736
• /
• 2022

To understand the characteristics of low-level clouds (CLs), environmental variables are composited on each CL using individual surface observations and six-hourly upper-air meteorologies around the globe. Individual CLs has its own distinct environmental conditions. Over the eastern subtropical and western North Pacific Ocean in JJA, stratocumulus (CL5) has a colder sea surface temperature (SST), stronger and lower inversion, and more low-level cloud amount (LCA) than the climatology whereas cumulus (CL12) has the opposite characteristics. Over the eastern subtropical Pacific, CL5 and CL12 are influenced by cold and warm advection within the PBL, respectively but have similar cold advection over the western North Pacific. This indicates that the fundamental physical process distinguishing CL5 and CL12 is not the horizontal temperature advection but the interaction with the underlying sea surface, i.e., the deepening-decoupling of PBL and the positive feedback between shortwave radiation and SST. Over the western North Pacific during JJA, sky-obscuring fog (CL11), no low-level cloud (CL0), and fair weather stratus (CL6) are associated with anomalous warm advection, surface-based inversion, mean upward flow, and moist mid-troposphere with the strongest anomalies for CL11 followed by CL0. Over the western North Pacific during DJF, bad weather stratus (CL7) occurs in the warm front of the extratropical cyclone with anomalous upward flow while cumulonimbus (CL39) occurs on the rear side of the cold front with anomalous downward flow. Over the tropical oceans, CL7 has strong positive (negative) anomalies of temperature in the upper troposphere (PBL), relative humidity, and surface wind speed in association with the mesoscale convective system while CL12 has the opposite anomalies and CL39 is in between.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2023.05a
• /
• pp.158-159
• /
• 2023

With the development and state-of-the-art of ocean satellites, wide-area management of the waters around Korea has become possible. In particular, in the field of ship operation, as autonomous navigation technology based on artificial intelligence and big data is being developed, there is a need for additional analysis and observation through ocean satellite data.. Researches that can combine ship operation with ocean satellite data include ship detection based on ocean satellites and ship navigation assistance using marine weather forecasting.

• Journal of Software Assessment and Valuation
• /
• v.15 no.2
• /
• pp.59-66
• /
• 2019

Since Jeju is the absolute weight of agriculture and tourism, the analysis of precipitation is more important than other regions. Currently, some numerical models are used for analysis of precipitation of Jeju Island using observation data from meteorological satellites. However, since precipitation changes are more diverse than other regions, it is difficult to obtain satisfactory results using the existing numerical models. In this paper, we propose a Jeju precipitation pattern analysis method using the texture analysis method based on Convolution Neural Network (CNN). The proposed method converts the water vapor image and the temperature information of the area of ​​Jeju Island from the weather satellite into texture images. Then converted images are fed into the CNN to analyse the precipitation patterns of Jeju Island. We implement the proposed method and show the effectiveness of the proposed method through experiments.

• Atmosphere
• /
• v.23 no.2
• /
• pp.187-204
• /
• 2013

This study analyzed the synoptic distribution and vertical structure about four cases of precipitation occurrences using NCEP/NCAR reanalysis data and upper level data of winter intensive observation to be performed by National Institute of Meteorological Research at Bukgangneung, Incheon, Boseong during 63days from 4 JAN to 6 MAR in 2012, and Observing System Experiment (OSE) using 3DVAR-WRF system was conducted to examine the precipitation predictability of upper level data at western and southern coastal regions. The synoptic characteristics of selected precipitation occurrences were investigated as causes for 1) rainfall events with effect of moisture convergence owing to low pressure passing through south sea on 19 JAN, 2) snowfall events due to moisture inflowing from yellow sea with propagation of Siberian high pressure after low pressure passage over middle northern region on 31 JAN, 3) rainfall event with effect of weak pressure trough in west low and east high pressure system on 25 FEB, 4) rainfall event due to moisture inflow according to low pressures over Bohai bay and south eastern sea on 5 MAR. However, it is identified that vertical structure of atmosphere had different characteristics with heavy rainfall system in summer. Firstly, depth of convection was narrow due to absence of moisture convergence and strong ascending air current in middle layer. Secondly, warm air advection by veering wind with height only existed in low layer. Thirdly, unstable layer was limited in the narrow depth due to low surface temperature although it formed, and also values of instability indices were not high. Fourthly, total water vapor amounts containing into atmosphere was small due to low temperature distribution so that precipitable water vapor could be little amounts. As result of OSE conducting with upper level data of Incheon and Boseong station, 12 hours accumulated precipitation distributions of control experiment and experiments with additional upper level data were similar with ones of observation data at 610 stations. Although Equitable Threat Scores (ETS) were different according to cases and thresholds, it was verified positive influence of upper level data for precipitation predictability as resulting with high improvement rates of 33.3% in experiment with upper level data of Incheon (INC_EXP), 85.7% in experiment with upper level data of Boseong (BOS_EXP), and 142.9% in experiment with upper level data of both Incheon and Boseong (INC_BOS_EXP) about accumulated precipitation more than 5 mm / 12 hours on 31 January 2012.

• Journal of the Korean earth science society
• /
• v.32 no.7
• /
• pp.750-760
• /
• 2011

In this study, the reproducibility of the simulated current climate by using two regional climate models, such as Seoul National University Regional Climate Model (SNURCM) and Weather Resuearch and Forecasting (WRF), is evaluated in advance to produce the standard regional climate scenario of future climate. Within the evaluation framework of a COordinated Regional climate Downscaling EXperiment (CORDEX), 28-year-long (1978-2005) regional climate simulation was conducted by using the Hadley Centre Global Environmental Model (HadGEM2-AO) global simulation data of the National Institute of Meteorological Research (NIMR) as a lateral boundary forcing. The simulated annual surface temperatures were in good agreement with the observation; the spatial correlation coefficients between each model and observation were over 0.98. The cold bias, however, were shown over the northern boundary in the both simulated results. In evaluation of the simulated precipitation, the skill was reasonable and good. The spatial correlation coefficients for the precipitation over the land area were 0.85 and 0.79 in SNURCM and WRF, respectively. It is noted that two regional climate models (RCMs) have different characteristics for the distribution of precipitation over equatorial and midlatitude areas. SNURCM shows better distribution of the simulated precipitation associated with the East Asia summer monsoon in the mid-latitude areas, but WRF shows better in the equatorial areas in comparison to each other. The simulated precipitation is overestimated in summer season (JJA) rather than in spring season (MAM), whereas the spatial distribution of the precipitation in spring season corresponds to the observation better than in summer season. Also the RCMs were capable of reproducing the annual variability of the maximum amount and its timing in July, in which the skills over the inland area were in better agreement with the observation than over the maritime area. The simulated regional climates, however, have the limitation to represent the number of days for extremely hot temperature and heavy rainfall over South Korea.

• Journal of the Korean earth science society
• /
• v.36 no.4
• /
• pp.330-340
• /
• 2015

For this study, we developed an algorithm to estimate the total amount of clouds using sky image data from the Skyviewer equipped with CCD camera. Total cloud amount is estimated by removing mask areas of RGB (Red Green Blue) images, classifying images according to frequency distribution of GBR (Green Blue Ratio), and extracting cloud pixels from them by deciding RBR (Red Blue Ratio) threshold. Total cloud amount is also estimated by validity checks after removing sunlight area from those classified cloud pixels. In order to verify the accuracy of the algorithm that estimates total cloud amount, the research analyzed Bias, RMSE, and correlation coefficient compared to records of total cloud amount earned by human observation from the Gangwon Regional Meteorological Administration, which is in the closest vicinity of the observation site. The cases are selected four daily data from 0800 LST to 1700 LST for each season. The results of analysis showed that the Bias in total cloud amount estimated by the Skyviewer was an average of -0.8 tenth, and the RMSE was 1.6 tenths, indicating the difference in total cloud amount within 2 tenths. Also, correlation coefficient was very high, marking an average of over 0.91 in all cases, despite the distance between the two observation sites (about 4 km).

• Journal of the Korean earth science society
• /
• v.30 no.6
• /
• pp.709-720
• /
• 2009

Field observation and numerical experiments were conducted to understand the impact of water retention pavement on the surface heat budget and on the regional circulation. The numerical model applied in this study is the atmospheric dynamic model Local Circulation Model (LCM) with two dimensional grid system, and a field observation was carried out under the clear sky and calm conditions of the weather on 19 July 2007. In the field observation, the maximum value of surface temperature on pavement covered with water retention material reached the $41.2^{\circ}C$ at 1430 LST and the values was lower for $16.1^{\circ}C$ than that of asphalt without the material. The Case BET03 assumed to be 0.3 for the surface evaporation efficiency was in good agreement with the observation and its sensible and latent heat fluxes were numerically estimated to be 229 and 227 $W/m^2$, respectively. Results of the numerical experiments demonstrated that the water retention pavement tends to induce the increase of latent heat flux associated with the lower surface temperature and mixing height during the daytime. Discontinuity of latent heat caused by the water retention pavement also tends to promote the development of mesoscale circulation called as land-land breeze or country breeze.

• Journal of the Korean earth science society
• /
• v.30 no.2
• /
• pp.185-195
• /
• 2009

We investigated the change of several meteorological variables due to deforestation. We established two sets of automatic weather observation system: one on a hill where forest was destructed by lumbering (Point 1) and the other in a neighboring district (Point 2) of fairly preserved forest. The observations were continued for one year (2006. 12-2007. 12). In this study, we analysed the data observed for one week from the nea day after summertime rainfall. The results showed that the air temperatures of Point 1 were about $1.5^{\circ}C$ higher than those of Point 2 during the daytime. But there were small gaps between the two poults during the nighttime. The relative humidities also differed greatly between the two during the daytime. It was as high as about 10% at Point 2. The surface and underground (15 cm in depth) soil temperatures were also fealty different between the two points during the daytime. They were $3-10^{\circ}C$ higher at Point 2 than those of Point 1. And the gaps reduced drastically during the nighttime. The averaged soil moistures were 7.1% at Point 1 and 19.5% at Point 2 during the observation period, respectively. The differences of wind direction were small, but the wind speeds differed between the two points. The observed wind speeds during the observation period were roughly estimated to be about 0.5m/s at Point 1 and 0.3m/s at Point 2. The heat budget analysis was also performed based on the observation data.

• Journal of the Korean earth science society
• /
• v.42 no.5
• /
• pp.504-513
• /
• 2021

Different physical processes, according to the atmospheric boundary layer types, were used in the Local Data Assimilation and Prediction System (LDAPS) of the Unified Model (UM) used by the Korea Meteorological Administration (KMA). Therefore, it is important to verify the atmospheric boundary layer types in the numerical model to improve the accuracy of the models performance. In this study, the atmospheric boundary layer types were verified using observational data. To classify the atmospheric boundary layer types, summer intensive observation data from radiosonde, flux observation instruments, Doppler wind Light Detection and Ranging(LIDAR) and ceilometer were used. A total number of 201 observation data points were analyzed over the course 61 days from June 18 to August 17, 2019. The most frequent types of differences between LDAPS and observed data were type 1 in LDAPS and type 2 in observed(each 53 times). And type 3 difference was observed in LDAPS and type 5 and 6 were observed 24 and 15 times, respectively. It was because of the simulation performance of the Cloud Physics such as that associated with the simulation of decoupled stratocumulus and cumulus cloud. Therefore, to improve the numerical model, cloud physics aspects should be considered in the atmospheric boundary layer type classification.