In this study, we focus on the improvement of data quality transmitted from a weather buoy that guides a route of ships. The buoy has an Internet-of-Thing (IoT) including sensors to collect meteorological data and the buoy’s status, and it also has a wireless communication device to send them to the central database in a ground control center and ships nearby. The time interval of data collected by the sensor is irregular, and fault data is often detected. Therefore, this study provides a framework to improve data quality using machine learning models. The normal data pattern is trained by machine learning models, and the trained models detect the fault data from the collected data set of the sensor and adjust them. For determining fault data, interquartile range (IQR) removes the value outside the outlier, and an NGBoost algorithm removes the data above the upper bound and below the lower bound. The removed data is interpolated using NGBoost or long-short term memory (LSTM) algorithm. The performance of the suggested process is evaluated by actual weather buoy data from Korea to improve the quality of ‘AIR_TEMPERATURE’ data by using other data from the same buoy. The performance of our proposed framework has been validated through computational experiments based on real-world data, confirming its suitability for practical applications in real- world scenarios.
Recently, the importance of impact-based forecasting has increased along with the socio-economic impact of severe weather have emerged. As news articles contain unconstructed information closely related to the people’s life, this study developed and evaluated a binary classification algorithm about snowfall damage information by using media articles text mining. We collected news articles during 2009 to 2021 which containing ‘heavy snow’ in its body context and labelled whether each article correspond to specific damage fields such as car accident. To develop a classifier, we proposed a probability-based classifier based on the ratio of the two conditional probabilities, which is defined as I/O Ratio in this study. During the construction process, we also adopted the n-gram approach to consider contextual meaning of each keyword. The accuracy of the classifier was 75%, supporting the possibility of application of news big data to the impact-based forecasting. We expect the performance of the classifier will be improve in the further research as the various training data is accumulated. The result of this study can be readily expanded by applying the same methodology to other disasters in the future. Furthermore, the result of this study can reduce social and economic damage of high impact weather by supporting the establishment of an integrated meteorological decision support system.
We present a novel method that can enhance the detection success rate of interstellar objects. Interstellarobjects are objects that are not gravitationally bound to our solar system and thus are believed to haveoriginated from other planetary systems. Since the nding of two interstellar objects, 1l/`Oumuamua in2017 and 2l/Borisov in 2019, much attention has been paid to nding new interstellar objects. In thispaper, we propose the use of Heliospheric Imagers (HIs) for the survey of interstellar objects. In particular,we show HI data taken from Solar TErrestrial RElation Observatory/Sun Earth Connection Coronal andHeliospheric Investigation and demonstrate their ability to detect `Oumuamua-like interstellar objects. HIs are designed to monitor and study space weather by observing the solar wind traveling throughinterplanetary space. HIs provide the day-side observations and thus it can dramatically enlarge theobservable sky range when combined with the traditional night-side observations. In this paper, we rstreview previous methods for detecting interstellar objects and demonstrate that HIs can be used for thesurvey of interstellar objects.
PURPOSES : Due to the frequent occurrence of accidents on icy roads during nighttime, it would be advantageous to notify road managers and drivers about the most perilous areas. This would allow road managers to treat the icy roads with de-icing chemicals and enable drivers to be better prepared for potential hazards. Essential information about pavement temperature is required to identify icy spots on the road. METHODS : With the goal of estimating nighttime pavement temperature on the National Highways in Korea using atmospheric data, the current study investigated a widely recognized forecasting method known as deep neural network (DNN). To achieve this objective, the input data for the models were gathered from the weather agency's website. The dataset comprised of relative humidity, air temperature, dew point temperature, as well as the differences in air temperature and humidity between two consecutive days. RESULTS : In order to assess the effectiveness of the built DNN model, a comparison was made using baseline pavement temperature data gathered through an infrared-based pavement temperature sensor installed in a highway patrol car. The results indicated that the DNN model achieved a mean absolute error (MAE) of 0.42 and a root mean square error (RMSE) of 0.62. In comparison, a conventional regression model yielded an MAE of 2.07 and an RMSE of 2.64. Thus, the DNN model demonstrated superior performance in comparison to the conventional regression model. CONCLUSIONS : Considering the increasing focus on preventive maintenance, these newly developed prediction models can be implemented proactively as a preventive measure against icing. This proactive approach has the potential to significantly improve traffic safety on winter roads.
PURPOSES : The purpose of this study was to develop techniques for forecasting black ice using historical pavement temperature data collected by patrol cars and concurrent atmospheric data provided by the Korea Meteorological Administration.
METHODS : To generate baseline data, the physical principle that ice forms when the pavement temperature is negative and lower than the dew-point temperature was exploited. To forecast frost-induced black ice, deep-learning algorithms were created using air, pavement, and dew point temperatures, as well as humidity, wind speed, and the z-value of the historical pavement temperature of the target segment.
RESULTS : The suggested forecasting models were evaluated against baseline data generated by the above-mentioned physical principle using pavement temperature and atmospheric data gathered on a national highway in the vicinity of Young-dong in the Chungcheongbukdo province. The accuracies of the forecasting models for the bridge and roadway segments were 94% and 90%, respectively, indicating satisfactory results.
CONCLUSIONS : Preventive anti-icing maintenance activities, such as applying anti-icing chemicals or activating road heating systems before roadways are covered with ice (frost), could be possible with the suggested methodologies. As a result, traffic safety on winter roads, especially at night, could be enhanced.
본 연구에는 가뭄의 유발 요인으로 강수량, 기온, 상대습도 등의 기상현상을 활용하고 가뭄 피해로 인한 대응 요소로서 대체수원, 제한급수, 운반급수 등의 비상급수를 적용하여 AI기반 가뭄 대응 정보 구축 방안을 구성하였다. AI 머신 러닝 기법 중 널리 사용되는 의사결정나무 모형을 통하여 예측 기법을 수립하였다. 연구 대상 지역은 비상급수 활용 빈도가 높고 종관기상관측소가 존재하는 충주시, 안동시, 의성군을 선정하였다. 자료 기간은 2014년부터 2019년까지의 자료를 이용하였으며, 가뭄 유발 기상요인으로 ASOS의 강수량 및 기온, 습도를 이용하고 가뭄 피해 요소로 국가 가뭄정보 포털의 비상급수 현황 자료를 활용하였다. 모형 학습 결과 정확도가 약 0.97, F1-Score가 약 0.5로 나왔으며, 이는 비상급수가 필요한 상황과 그렇지 않은 상황을 97%의 확률로 예측할 수 있음을 의미하며, 비상급수가 필요했던 표본만을 대상으로 했을 경우 약 50%의 확률로 예측이 가능한 것을 의미한다. 따라서 의사결정나무 모형을 적용하여 예측 정확도를 분석한 결과 가뭄 대응 비상급수 준비지역 예측을 위한 적용성이 높은 것으로 평가되었다. 그러나 본 연구에서는 기상 조건만을 가뭄 유발 요인으로 반영하였기 때문에, 공급수량 부족 등의 요인을 추가적으로 검토할 필요가 있으므로 가뭄과 연관된 요소인 저수지 용량 등을 추가하고 비상급수 이외의 피해 요소 역시 확장하여 연구를 개선하고자 한다.
This study was aimed to find yield prediction model of Italian ryegrass using climate big data and geographic information. After that, mapping the predicted yield results using Geographic Information System (GIS) as follows; First, forage data were collected; second, the climate information, which was matched with forage data according to year and location, was gathered from the Korean Metrology Administration (KMA) as big data; third, the climate layers used for GIS were constructed; fourth, the yield prediction equation was estimated for the climate layers. Finally, the prediction model was evaluated in aspect of fitness and accuracy. As a result, the fitness of the model (R2) was between 27% to 95% in relation to cultivated locations. In Suwon (n=321), the model was; DMY = 158.63AGD –8.82AAT +169.09SGD - 8.03SAT +184.59SRD -13,352.24 (DMY: Dry Matter Yield, AGD: Autumnal Growing Days, SGD: Spring Growing Days, SAT: Spring Accumulated Temperature, SRD: Spring Rainfall Days). Furthermore, DMY was predicted as 9,790±120 (kg/ha) for the mean DMY(9,790 kg/ha). During mapping, the yield of inland areas were relatively greater than that of coastal areas except of Jeju Island, furthermore, northeastern areas, which was mountainous, had lain no cultivations due to weak cold tolerance. In this study, even though the yield prediction modeling and mapping were only performed in several particular locations limited to the data situation as a startup research in the Republic of Korea.
Agrometeorological information used in agriculture field depends heavily on productivity in accordance with analysis and forecasting accuracy. In this study, we used necessary weather information for potato crops during the growing season and developed a decision support system to help farmers' management activities. The core of the solution is to utilize the real-time weather data observed in my field. By using application weather information in agricultural field rather than simple weather information transmission, we understood control effect through various pest information services. These ICT (Information & Communication Technologies) solutions in the field of weather-based agriculture provide appropriate intention information for growing crops, thus it is expected that effective control will reduce the risk of insect pests and reduce direct costs of farmers. In addition, it is expected that it will contribute not only to reduce soil pollution but also to safe food production by controlling indiscriminate use of pesticides properly.
본 연구에서는 최근 20년간(1992-2011년)의 기상 데이터를 이용하여 태풍의 전향에 대하여 분석하였다. 연구 결과, 연별 전향하는 태풍의 수는 연별 태풍 발생 수의 감소와 함께 서서히 줄어드는 경향을 보였다. 전향하는 태풍은 특히 8-10월에 많고, 7-10월에 전향하는 태풍의 수는 전체 전향 태풍의 71 %를 차지하였다. 남중국해를 제외한 북태평양에서 전향한 태풍의 수명은 5일과 7일이 가장 많았고, 전향한 태풍의 평균 수명은 6.8일이었다. 이는 전향, 비전향의 모든 태풍의 평균 수명보다 다소 긴 값이다. 전향 태풍의 대부분은 태풍으로 발달한 후 5일 이내에 북위 20-34도, 동경 120-139도에서 북동쪽으로 전향하였다. 평균 전향 위도는 북위 25도, 평균 전향 경도는 동경 135도이었다.
본 연구는 국립기상연구소에서 생산한 기상자원지도 및 기상정보의 원시자료로부터 Web2.0기반의 기상자원정보시스템 설계 및 구축을 목적으로 한다. 세부 목표로는 (1)기상자원 DB의 공간정보화, 통계적 정보 제공에 적합한 DB 서비스 개발 (2)사용자 경험 중심의 RIA 서비스 개발, (3)기상자원 KML 배포 서비스 웹 제공을 통해 사용자가 참여와 기상정보 콘텐츠 창작환경 조성에 있다. 이를 위해 풍력과 태양광 관련 기상자원지도와 원시자료를 이용해 GeoDB 및 웹서비스를 설계 및 구현하였다. 결과로서 구축된 기상자원 GeoDB 및 웹서비스가 Web2.0 패러다임을 지향하는 지에 대해 그 활용성을 평가하였다. 본 연구를 통해 생산생산된 기상자원 공간/통계 정보 및 RIA서비스는 입지분석, 의사결정지원, 창작활동 등 여러 응용분야에 유용하게 활용될 것이다.
This study had two main objectives. We first investigated which weather phenomena people were most concerned about in the context of climate change or global warming. Then, we conducted content analysis to find which words were more commonly used with climate change or global warming. For this, we collected web data from Twitter, Naver, and Daum from April to October 2019 in the Republic of Korea. The results suggested that people were more concerned about air quality, followed by typhoons and heat waves. Because this study only considered one warm period in the year of 2019, winter-related weather phenomena such as cold wave and snowfall were not well captured. From Twitter, we were able to find wider range of terminologies and thoughts/opinions than Naver and Daum. Also, more life-relevant weather events such as typhoons and heat waves in Twitter were commonly mentioned compared to Naver and Daum. On the other hand, the comments from Naver and Daum showed relatively narrower and limited terms and thoughts/ opinions. Especially, most of the comments were influenced by headlines of articles. We found many comments about air quality and energy/economic policy. We hope this paper could provide background information about how to promote the climate change education and public awareness and how to efficiently interact with general audiences.
본 연구는 건물에너지 효율 향상을 위한 목적으로 기상데이터 변화에 따른 건물 냉 난방부하량을 예측하고 결과를 비교 분석한 것으로, 연구 성과는 다음과 같다. 1)기상청에서 입수데이터를 평가툴인 ESP-r에 활용할 수 있도록 항목별 기상데이터를 개발하였다. 표준기상 데이터의 외기온도, 습도, 풍속은 대부분의 경우 기상청데이터 보다 크거나 높았다. 수평면전일사량은 기상청데이터가 높았고, 직달일사량은 겨울철에는 표준기상데이터가, 여름철에는 기상청데이터가 많은 것으로 나타났다. 2)대학교 캠퍼스 내에 신축된 후생복지관을 대상으로 한 시뮬레이션 결과, 최대난방부하의 경우 표준년도, 2006년, 2009년이 비슷한 반면 2007년은 표준년도 대비 81%, 2008년은 96% 수준이었고, 연간난방부하는 2006년, 2008년의 순으로 난방수요가 많았다. 한편, 냉방부하의 경우에는, 상대적으로 최대냉방부하가 큰 2007년, 2009년의 연간 냉방부하보다 최대냉방부하가 가장 적은 2008년의 연간냉방부하가 더 큰 결과를 보였다. 3)냉 난방기기의 상당시간가동률을 평가한 결과, 표준년도의 최대부하대비 상당시간가동률은 2006~2009년이 표준년도에 비해 대부분 가동률이 낮았다.