• Korean Journal of Agricultural and Forest Meteorology
• /
• v.5 no.3
• /
• pp.185-190
• /
• 2003

This study was conducted to examine rice growth duration by analyzing agricultural climatic conditions at different latitudes in unified Korea. The climatic conditions of nine sites from Wunggi (latitude 42N) to Jeju (latitude 31N) were examined in this study. The rice growth duration of various cropping patterns was determined by analyzing consecutive days when effective daily mean air temperature was suitable for rice growth from the first seeding date to the last maturing date. The rice growth duration in Wunggi located in North Korea was available 138 days for machine transplanting, 115 days for direct seeding on dry paddy cultivation, and 97 days for direct seeding on a flooded surface with cultivation after seeding. On the other hand, the rice growth duration in Kwangju (latitude 35N) located in South Korea was 195 days for machine transplanting, 180 days for direct seeding on dry paddy cultivation, and 170 days for direct seeding on a flooded surface cultivation after seeding.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.25 no.2
• /
• pp.80-98
• /
• 2023

This study explores the effectiveness of various data preprocessing algorithms for improving subseasonal to seasonal (S2S) climate predictions from six climate forecast models and their Multi-Model Ensemble (MME) using a deep learning-based postprocessing model. A pipeline of data transformation algorithms was constructed to convert raw S2S prediction data into the training data processed with several statistical distribution. A dimensionality reduction algorithm for selecting features through rankings of correlation coefficients between the observed and the input data. The training model in the study was designed with TimeDistributed wrapper applied to all convolutional layers of U-Net: The TimeDistributed wrapper allows a U-Net convolutional layer to be directly applied to 5-dimensional time series data while maintaining the time axis of data, but every input should be at least 3D in U-Net. We found that Robust and Standard transformation algorithms are most suitable for improving S2S predictions. The dimensionality reduction based on feature selections did not significantly improve predictions of daily precipitation for six climate models and even worsened predictions of daily maximum and minimum temperatures. While deep learning-based postprocessing was also improved MME S2S precipitation predictions, it did not have a significant effect on temperature predictions, particularly for the lead time of weeks 1 and 2. Further research is needed to develop an optimal deep learning model for improving S2S temperature predictions by testing various models and parameters.

• Proceedings of the Korean Environmental Sciences Society Conference
• /
• 2004.05a
• /
• pp.32-36
• /
• 2004

본 연구는 최근 인구와 산업체가 급증하고 있는 김해시를 대상으로 김해시.군을 통합한 1996년 이후부터 2002년까지 김해시 지상오존농도의 변화경향과 고농도 오존일에 대한 사례 연구결과 다음과 같다. 김해시 대기오염농도의 시계열변화는 증가추세가 뚜렷하였으며, 최근 대기환경기준이 강화되면서 환경지정기준을 초과하기도 하였다. 계절별 오존농도의 일변화는 사계절 가운데 봄철이 가장 농도가 높게 나타났으며 매년 증가하는 농도폭도 가장 크게 나타났다. 이는 여름철의 경우 태양고도가 높고 단위시간당 일사량은 많으나, 장마와 같은 운량 증가로 인한 일사량의 감소되거나, 기온이 상승할 경우 해안으로부터 해풍이 증가하여 여름보다는 봄철에 더 높은 농도를 나타낸 것으로 생각되며, 인구와 자동차 등의 증가로 인한 배출량이 증가한 것도 크게 영향을 미친 것으로 생각된다. 연구기간동안 60 ppb 이상의 고농도 오존일의 총 발생빈도는 237일로 나타났으며, 매년 발생빈도가 증가하고 있으며, 대기환경기준 100 ppb/hr를 초과한 날도 8일이나 되었는데, 1999년이 4일 발생하여 오존에 의한 대기오염규제지역으로 선정되었는데, 고농도 오존일은 지난 1997년 이후 매년 증가하고 있으며 겨울을 제외한 전 월에 발생하는 특징을 나타내어 오존의 저감을 위한 실천 계획 수립뿐 아니라 고농도 오존일에 대한 집중적인 연구와 빠른 시간 내에 오존의 예.경보제를 도입 운영하는 것이 김해시민의 건강과 복지에 도움이 될 것으로 생각된다. 고농도 오존일에 대한 사례일 첫 번째인 5월 1일은 일 최고 기온은 그리 높은 상태는 아니었지만 광범위한 이동성 고기압에 의해 대기가 정체하고 바람이 약한 시점에 광화학반응에 의한 오존생성이 용이하였고 해안가의 높은 농도의 오존이 수송되어 고농도 오존이 발생하였으며, 사례 2의 경우 대륙에서 이동해 오는 이동성 고기압의 영향으로 대기는 안정하고 바람이 약하여 기온이 급상승하였으며, 광화학반응에 의한 오존 생성이 용이하였다. 사례 3의 경우는 남북으로 놓여 있는 대규모 기압계 사이에 안상부 형태의 대상고기압이 놓여 대기는 매우 안정하고 바람이 약하며 때때로 기압계에 의한 바람이 불 경우 다소 강한 바람이 불어 일사량이 많은 기압계에서 광화학반응과 수소에 의한 오존생성이 용이하여 고농도 오존을 발생하였다.

• Journal of the Korean Geographical Society
• /
• v.39 no.5 s.104
• /
• pp.711-721
• /
• 2004

The aim of this study is to clarify whether frequency and/or severity of extreme climate events have changed significantly in Korea during recent years. Using the best available daily data, spatial and temporal aspects of ten climate change indicators are investigated on an annual and seasonal basis for the periods of 1954-1999. A systematic increase in the $90^{th}$ percentile of daily minimum temperatures at most of the analyzed areas has been observed. This increase is accompanied by a similar reduction in the number of frost days and a significant lengthening of the thermal growing season. Although the intra-annual extreme temperature range is based on only two observations, it provides a very robust and significant measure of declining extreme temperature variability. The five precipitation-related indicators show no distinct changing patterns for spatial and temporal distribution except for the regional series of maximum consecutive dry days. Interestingly, the regional series of consecutive dry days have increased significantly while the daily rainfall intensity index and the fraction of annual total precipitation due to events exceeding the $95^{th}$ percentile for 1901-1990 normals have insignificantly increased.

• Journal of the Korean Geographical Society
• /
• v.48 no.1
• /
• pp.37-55
• /
• 2013

Previous studies have shown that Foehn was mainly observed in Young-seo area in Korea. However, they have failed to indicate the area where Foehn can be observed most frequently in Young-seo area and how Foehn is distributed in that area. This study targets HongCheon area in Young-seo province and examines the frequency and extent of Foehn in local scale through documenting a daily maximum air temperature map of Foehn. The period examined in this study is the months between March and June from 2003 to 2012. CoKriging method, which uses temperature and the altitude above sea, generates a higher level of accuracy in making daily maximum air temperature map of Foehn occurring days. We have found that Foehn is observed in certain areas, not all areas of HongCheon region, by compiling the daily maximum air temperature map. In particular, Foehn was found to be frequent and strong in the downstream of HongCheon river. In addition, we surveyed the residents of HongCheon about their perception of Foehn. They did not know whether high temperature and dryness in spring are caused by Foehn. The methods and techniques used to examine Foehn in local climate scale by this study will enhance the understanding of regional climate and contribute towards the research in this area. In particular, they can be applied to high temperature that recently occurred between spring and summer, excessive hotness in summers, agricultural plant growth in springs and etc.

• 한국지구과학회:학술대회논문집
• /
• 2002.02a
• /
• pp.50-50
• /
• 2002

• Horticultural Science & Technology
• /
• v.28 no.6
• /
• pp.937-941
• /
• 2010

This study investigated the effect of global warming on full bloom date (FBD) of 'Niitaka' pear ($Pyrus$ $pyrifolia$ Nakai) tree by calculating the development stage index by hourly temperatures recorded at Pear Research Station, estimating the distribution of average FBD and the change of FBD according to temperature rising by integrating development rate at 67 locations in Korea Meteorological Administration site. Development stage index of 'Niitaka' pear tree was 0.9593 at Naju location. Differences between full bloom dates observed at Cheonan region and predictions by development stage index were 0-7 days, and matched year was 35.3%. FBDs of 67 locations were distributed from April 4 to May 28. When yearly temperature was raised 1, 2, 3, 4, and $5^{\circ}C$ at 67 locations, predicted FBD was accelerated at most of the locations. However, FBD decelerated at south coast locations from $3^{\circ}C$ rise and did not bloom at 'Gosan', 'Seogwipo', and 'Jeju' locations from $4^{\circ}C$ rise. When monthly temperature was raised 1, 3, and $5^{\circ}C$ at 67 locations, predicted FBD was the most accelerated at March temperature rise, and followed by April, February, January and December. Therefore, global warming will cause acceleration of the full bloom date at pear production areas in Korea.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.40 no.4
• /
• pp.444-450
• /
• 1995

It was very hard to gather the seeds of Eleutherococcus senticosus Max. known as a medicinal plant for they tend to drop under the high temperature condition during the summer period in Korea. Therefore, this study was conducted to select seed production site for Eleutherococcus senticosus in Korea, comparing the climate of Hokkaido of Japan, in which the seeds have been produced, with that of various place in this country. It was low that the average maximum temperature during the hottest summer two months (July and August) as a 24.4$^{\circ}C$ in Hokkaido and 21.2$^{\circ}C$ in Daegwanryeong compared with 27.4$^{\circ}C$ in Changsu. Especially in Daegwanryeong, average maximum temperature from June to September remained as low as 21$^{\circ}C$. Effective accumulated temperature(>5$^{\circ}C$) was 807$^{\circ}C$ in Hokkaido and 964$^{\circ}C$ in Daegwanryeong during the ripening period. Monthly sunshined hours from July to August were 121.7~128 hours in Daegwanryeong and 83.5~85.4 hours in Hokkaido. The Eleutherococcus senticosus sprouts at 8.5$^{\circ}C$, comes to flowering season in mid-August, and ripens during late-August and October in Hokkaido, the climate of which is similar to that of Daegwanryeong.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.12 no.4
• /
• pp.217-224
• /
• 2010

This study was carried out to quantify potential effects of the surrounding ocean on the observed air temperature at coastal weather stations in the Korean Peninsula. Daily maximum and minimum temperature data for 2001-2009 were collected from 66 Korea Meteorological Administration (KMA) stations and the monthly averages were calculated for further analyses. Monthly data from 27 inland sites were used to generate a gridded temperature surface for the whole Peninsula based on an inverse distance weighting and the local temperature at the remaining 39 sites were estimated by recent techniques in geospatial climatology which are widely used in correction of small - scale climate controls like cold air drainage, urban heat island, topography as well as elevation. Deviations from the observed temperature were regarded as the 'apparent' sea effect and showed a quasi-logarithmic relationship with the distance of each site from the nearest coastline. Potential effects of the sea on daily temperature might exceed $6.0^{\circ}C$ cooling in summer and $6.5^{\circ}C$ warming in winter according to this relationship. We classified 25 sites within the 10 km distance from the nearest coastline into 'coastal sites' and the remaining 15 'fringe sites'. When the average deviations of the fringe sites ($0.5^{\circ}C$ for daily maximum and $1.0^{\circ}C$ for daily minimum temperature) were used as the 'noise' and subtracted from the 'apparent' sea effects of the coastal sites, maximum cooling effects of the sea were identified as $1.5^{\circ}C$ on the west coast and $3.0^{\circ}C$ on the east and the south coast in summer months. The warming effects of the sea in winter ranged from $1.0^{\circ}C$ on the west and $3.5^{\circ}C$ on the south and east coasts.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.2 no.4
• /
• pp.175-182
• /
• 2000

Spatial interpolation of daily temperature forecasts and observations issued by public weather services is frequently required to make them applicable to agricultural activities and modeling tasks. In contrast to the long term averages like monthly normals, terrain effects are not considered in most spatial interpolations for short term temperatures. This may cause erroneous results in mountainous regions where the observation network hardly covers full features of the complicated terrain. We developed a spatial interpolation model for daily minimum temperature which combines inverse distance squared weighting and elevation difference correction. This model uses a time dependent function for 'mountain slope lapse rate', which can be derived from regression analyses of the station observations with respect to the geographical and topographical features of the surroundings including the station elevation. We applied this model to interpolation of daily minimum temperature over the mountainous Korean Peninsula using 63 standard weather station data. For the first step, a primitive temperature surface was interpolated by inverse distance squared weighting of the 63 point data. Next, a virtual elevation surface was reconstructed by spatially interpolating the 63 station elevation data and subtracted from the elevation surface of a digital elevation model with 1 km grid spacing to obtain the elevation difference at each grid cell. Final estimates of daily minimum temperature at all the grid cells were obtained by applying the calculated daily lapse rate to the elevation difference and adjusting the inverse distance weighted estimates. Independent, measured data sets from 267 automated weather station locations were used to calculate the estimation errors on 12 dates, randomly selected one for each month in 1999. Analysis of 3 terms of estimation errors (mean error, mean absolute error, and root mean squared error) indicates a substantial improvement over the inverse distance squared weighting.