• Journal of the Korean Geographical Society
• /
• v.41 no.3 s.114
• /
• pp.361-372
• /
• 2006

The purpose of this paper is to analyze temporal trends and spatial distribution of frost occurrence days using 54 weather stations under the Korea Meteorological Administration during the recent 30years ($1971{\sim}2000$). The frost occurrence days in the middle inland area was more than in the coastal area, but the period of free-frost days in the coastal area was longer than that in the middle inland. In the southern and eastern coastal areas, the frost day was more than in the western coastal area. The distribution of the first frost day was similar to that of the last frost day. The first frost day in the coastal area appeared a month later than that in the middle inland. The last frost day in the coastal area appeared $20{\sim}25$ days later than that in the middle inland. At most of stations which the first frost day was getting delayed and the last frost day earlier, the frost occurrence days showed a decreasing trend. However, in some stations, the frost occurrence day was increasing. As the frost days in October, March and April at some stations showed an increasing trend, the damage due to the frost might be increased in some areas in South Korea.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.24 no.2
• /
• pp.83-94
• /
• 2022

In order to actively prepare to frost damage that occurs in the process of growing crops, the spatial and temporal distribution of frost occurrence in South Korea was derived using frost observation data from 20 regions over the past 21 years (2000~2020). The main products are the number of frost days, first frost day, and last frost day by region. And the climatic trends of these results were identified by performing the Mann-Kendall trend test and Sen's slope estimator. In South Korea, a lot of frost occurs in the inland area to the west of the Taebaek and Sobaek Mountains. Relatively closer to the coastal area, the number of frost days is small, the first frost day is slow, and the last frost day is early. The east coast region has fewer frost days, the first frost day is later, and the last frost day is earlier than the west coast region. The southern sea, the southeastern sea region, and the island region rarely experience frost. As a result of the annual time series trend analysis, although South Korea is a country where climate warming is progressing, there was no trend in reducing the number of frost days and slowing the first frost day, and it was found that the last frost day is delayed by 0.5 days per year.

• Journal of the Korean Geographical Society
• /
• v.43 no.6
• /
• pp.824-842
• /
• 2008

The main objective of the study was to analyze the weather conditions of frost occurrence for protecting crops against frost damage in Korea. The primary data used for the analysis of meteorological characteristics of frost occurrence days are the airmass pattern, minimum temperature, grass minimum temperature, daily temperature range, relative humidity, minimum relative humidity, mean wind speed in autumn and spring. Frost often occurs when the migratory anticyclone passes the southwest of Korea. The importance of grass minimum temperature measurements for agricultural purposes has previously been recognized. The grass minimum thermometer is capable of detecting ground frosts which are often not recorded by the minimum thermometer. The minimum temperature of frost occurrence days is above $0^{\circ}C$ in the coastal area, but the grass minimum temperature of frost occurrence days is below $0^{\circ}C$ in the whole area. The daily temperature of frost occurrence days is about 9 to $12^{\circ}C$ in the coastal area and is over $14^{\circ}C$ in the inland area. The minimum relative humidity of frost occurrence days is about 30 to 50%. The mean wind speed of frost occurrence days is less than 2m/sec.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.19 no.3
• /
• pp.86-92
• /
• 2017

In this study, a model to classify frost occurrence and frost free day was developed using the digital weather forecast data provided by Korea Meteorological Administration (KMA). The minimum temperature, average wind speed, relative humidity, and dew point temperature were identified as the meteorological variables useful for classification frost occurrence and frost-free days. It was found that frost-occurrence date tended to have relatively low values of the minimum temperature, dew point temperature, and average wind speed. On the other hand, relatively humidity on frost-free days was higher than on frost-occurrence dates. Models based on machine learning methods including Artificial Neural Network (ANN), Random Forest(RF), Support Vector Machine(SVM) with those meteorological factors had >70% of accuracy. This results suggested that these models would be useful to predict the occurrence of frost using a digital weather forecast data.

• Journal of Sericultural and Entomological Science
• /
• v.32 no.2
• /
• pp.89-93
• /
• 1990

Mulberry, damaged by a late frost on April 27, received a foliar spray of 0.5% Urea or Jambi(a foliar fertilizer). The sprays were applied once every 3 days from May 16 for 12 days. The results were : 1. Approximately 70 auxillary buds per tree sprouted after the frost. The difference between the lowest and highest number of buds was 3 to fold. 2. For the 25 days beginning May 18, shoot length increased 5.4 fold, leaf number 2.1 fold, and shoot weight 9.6 fold. 3. Yield from latent buds was about 5% of total yield. 4. Length of new shoots decreased 26% due to frost. Urea spray and Jambi increased shoot length 10% and 1%, respectively, compared to control. 5. Leaf number decreased by 5.2 due to frost. Urea spray and Jambi increased leaf number 18% and 5%, respectively, compared to control. 6. Weight of new shoot was decrease 43% by frost. Urea and Jambi spray increased shoot weight 7% and 6%, respectively, compared to control. 7. Net leaf yield decreased 47% due to frost. Urea spray and Jambi increased left yield 7% and 5%, respectively, compared to control. 8. Yield in an undamaged plot was 1,587kg, damaged 932kg, 1,070kg in urea spray, and 1,033kg/10a in Jambi spray. Yield of frost damaged plots was 59% of undamaged. Foliar spray increased this to 68% of undamaged. 9. Yield increased 3.5% daily by with postponement of harvest. Yield of four days postponement was expected 73% of undamaged. Urea spray with 4 days postponement expected to be increased this to 82% of undamaged.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.19 no.3
• /
• pp.110-119
• /
• 2017

The agroclimatic indices are produced by statistical analysis based on primary climate data (e.g., temperature, precipitation, and solar irradiance) or driving agronomic models. This study was carried out to evaluate how selection of daily temperature for a climate normal (1983-2012) affected the precision of the agroclimatic indices. As a first step, averaged daily 0600 and 1500 LST temperature for a climate normal were produced by geospatial schemes based on topo-climatology ($365days{\times}1$ set, EST normal year). For comparison, 30 years daily temperature data were generated by applying the same process ($365days{\times}30sets$), and calculated mean of daily temperature (OBS normal year). The flowering date of apple 'Fuji' cultivar, the last frost date, and the risk of late frost were estimated based on EST normal year data and compared with the results from OBS normal year. The results on flowering date showed 2.9 days of error on average. The last frost date was of 11.4 days of error on average, which was relatively large. Additionally, the risk of the late frost was determined by the difference between the flowering and the last frost date. When it was determined based on the temperature of EST normal year, Akyang was classified as a risk area because the results showed that the last frost date would be the same or later than the flowering date in the 12.5% of area. However, the temperature of OBS normal year indicated that the area did not have the risk of a late frost. The results of this study implied that it would be necessary to reduce the error by replacing the EST method with the OBS method in the future.

• Journal of Korea Water Resources Association
• /
• v.54 no.9
• /
• pp.667-680
• /
• 2021

We investigated changes in frost days and frost-free periods and to comparatively assess frost event prediction models developed using logistic regression (LR), random forest (RF), and long short-term memory (LSTM) networks. The meteorological variables for the model development were collected from the Suwon, Cheongju, and Gwangju stations for the period of 1973-2019 for spring (March - May) and fall (September - November). The developed models were then evaluated by Precision, Recall, and f-1 score and graphical evaluation methods such as AUC and reliability diagram. The results showed that significant decreases (significance level of 0.01) in the frequencies of frost days were at the three stations in both spring and fall. Overall, the evaluation metrics showed that the performance of RF was highest, while that of LSTM was lowest. Despite higher AUC values (above 0.9) were found at the three stations, reliability diagrams showed inconsistent reliability. A further study is suggested on the improvement of the predictability of both frost events and the first and last frost days by the frost event prediction models and reliability of the models. It would be beneficial to replicate this study at more stations in other regions.

• Journal of The Korean Society of Grassland and Forage Science
• /
• v.20 no.3
• /
• pp.147-154
• /
• 2000

This study was conducted to increase the utility and productivity of forage crops at high altitued areas. For that purpose, 21 cultivars of corn and 2 cultivars of rye were cultivated for 3 years using a randomized block designed with 3 replications. The results obtained were as follows. 1. In the Taekwallyong area (800m above sea level) which has a short frostless period, all the seeding and harvest of corn must be finished within about 135 days between mid May, the time of the last frost, and late September, the time of the first frost 2. It was relatively safe for the early maturity cultivar(ll0days) and the medium maturity cultivar(l20days), compared to the late maturity cultivar(l30days) which might have had the possibility of an overlapping period between the time of harvest and the first frost in high altitude areas 3. The productivity of forage corn, which is the most efficient crop for capturing solar energy, varied significantly with the climate circumstances but the productivity of Taekwallyong showed similar results of 19 M/T/ha, compared with 20 M/T/ha in Suwon from the '96-'98 study 4. Rye could be cultivated in high altitude areas and when corn was raised as a second crop after rye in the same year, it was possible to increase the productivity of dry matter yield by 20% through double cropping(P < 0.05). (Key words : Cropping system, Corn, Rye, Forage production)

• Proceedings of The Korean Society of Agricultural and Forest Meteorology Conference
• /
• 2019.08a
• /
• pp.212-212
• /
• 2019

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.6 no.3
• /
• pp.164-169
• /
• 2004

A spatial interpolation scheme incorporating local geographic potential for cold air accumulation (TOPSIM) was used to test the feasibility of operational frost warning in Chatancheon basin in Yeoncheon County, where the introduction of new crops including temperate zone fruits is planned. Air temperature from April to June 2003 was measured at one-minute intervals at four locations within the basin. Cold-air accumulation potentials (CAP) at 4 sites were calculated for 3 different catchment scales: a rectangular area of 65 x 55 km which covers the whole county, the KOWACO (Korea Water Corporation) hydrologic unit which includes all 4 sites, and the sub-basins delineated by a stream network analysis of the digital elevation model. Daily minimum temperatures at 4 sites were calculated by interpolating the perfect prognosis (i.e., synoptic observations at KMA Dongducheon station) based on TOPSIM with 3 different CAPs. Mean error, mean absolute error, and root mean square error were calculated for 45 days with no precipitation to test the model performance. For the 3 flat locations, little difference was detected in model performance among 3 catchment areas, but the best performance was found with the CAPs calculated for sub-basins at one site (Oksan) on complex terrain. When TOPSIM loaded with sub-basin CAPs was applied to Oksan to predict frost events during the fruit flowering period in 2004, the goodness of fit was sufficient for making an operational frost warning system for mountainous areas.