• Proceedings of the Korean Association of Geographic Inforamtion Studies Conference
• /
• 2008.10a
• /
• pp.153-161
• /
• 2008

Agriculture and forest are basis for livelihood in Nepal while both sectors constitute around 40 percent of the national product and over two-thirds of the economically active population is dependent on agriculture. However, radical changes in land use, depletion in crops production and food availability are major threats due to loss of soil fertilityand severe environmental degradation. In this study, we used time series data from 1986/87 to 2005/06 about food crop production and population published by Government of Nepal, Ministry of Agriculture and Cooperatives and Central Bureau of Statistics. Descriptive statistics and ArcGIS were used to assess and map the food security status of Nepalese Terai based on the local food demand and supply system. Food supply to demand ratio(FSDR) was the main idea of assessment. Our results showed that out of 20 districts, only 8 districts were categorised under secured food districts whereas 5 districts were still under food unsecured situation. The analysis further revealed that 7 districts had faced food deficit more than 8-16 times during the last 20 year periods. Data further showed that there was surplus food supply relative to the requirements dictated by FSDR. However, the average FSDR was less than 1.2(less than 20% surplus) exploring fact that most of the districts were not producing sufficient food to cope up the food shock and after 1995 it was relatively stagnant. Our prediction reveals that food supply in Terai even in the future would remain at almost the same level as now, and there will not more than 16-17% surplus by 2021 considering medium vibrant population growth. The findings thus, indicate that Terai may not be a food secure region in the future, even though the region is considered as a food storage house of Nepal. In addition, this paper suggests ways to make future comprehensive case studies more widely comparable in Terai, Nepal.

• Ecology and Resilient Infrastructure
• /
• v.4 no.2
• /
• pp.97-104
• /
• 2017

For the restoration of lateral connectivity between channel and floodplain, it is important to find the former floodplain and to characterize its land use in streams which were channelized by the levee construction for the flood protection. The aim of this study is to map the former floodplains and to assess its land use pattern in the Cheongmi-cheon Stream, Korea. The former floodplains were explored by being overlapped on a digital elevation model (DEM), digital topographic map and design flood level using a geographical information system (GIS) in the Cheongmi-cheon Stream basin. The land use of the identified former floodplains was classified by land-use map. The total number of the former floodplains was 104 and their total area was $11.9km^2$ in the Cheongmi-cheon Stream. The land use pattern of the former floodplains was mostly farmland (87.1%). The former floodplains were usually surrounded by mountain forest in the downstream of the Cheongmi-cheon Stream. These former floodplains are probably suitable for restoration of lateral connectivity because of lower ratio of urban area but higher ratio of farmland. The results of delineation and land use analysis of the former floodplain can be used as a baseline data for planning stream restoration in the Cheongmi-cheon Stream.

• Ecology and Resilient Infrastructure
• /
• v.2 no.4
• /
• pp.324-329
• /
• 2015

For the restoration of lateral connectivity between rivers and floodplains, it is important to find the isolated former floodplain (IFF) and to characterize its land use in Korean rivers which were channelized by levee constructions for flood protection. The aim of this study is to map the IFF and to assess its land use pattern in the Nakdong River, Korea. The isolated former floodplain was explored by being overlapped on a digital elevation model (DEM), digital topographic map and design flood level using a geographical information system (GIS) in the Nakdong River basin. The land use of the identified IFF was classified by land-use map. The total number of IFFs was 384 and their total area was $291km^2$. While IFFs were usually surrounded by mountain forest in the upper river area, they tended to be located on wide plain areas in the downstream area of Nakdong River. The land use pattern of IFFs was mostly farmland (73.9%) and urban areas (12.7%) in the river. The results of delineation and land use analysis of isolated former floodplain in the Nakdong River will be used as a base line data for planning stream restoration.

• Spatial Information Research
• /
• v.17 no.3
• /
• pp.261-268
• /
• 2009

This study accomplished to draw a soil erosion map and a grade map of soil loss hazard in Korea. RUSLE and Rainfall-runoff (R) factor, which was estimated by using the rainfall data observed in 59 meteorological stations from 1977 to 2006 (for 30 years). FARD was used to analyze the frequency, and the whole country R factor was estimated according to the frequency. In the analysis of estimating the whole country R factor, Nakdong river has the smallest vaule, but Han river has the biggest value. According to the result of analyzing soil loss, soil loss occurred in a grass land, a bare land and a field in size order, and also approximately 17.2 ton/ha soil loss happened on the whole area. The average soil loss amount by the unit area takes place in a bare land and a grass land a lot. The total amount of soil loss in 5-year-frequency rainfall yields 15,000 ton and, what is more, a lot of soil loss happens in a paddy field, a forest and a crop field. The grade map of soil loss hazard is drawn up by classifying soil loss hazard grade by 5. As a result of analyzing soil loss, the moderate area which is the soil loss hazard grade 2 takes up the largest part, 72.8% of the total soil loss hazard area, on the contrary, the severe soil loss hazard area takes up only $1,038km^2$ (1.1%) of the whole area. The severe soil loss hazard area by land cover shows $93.5km^2$ in a bare land, $168.1km^2$ in a grass land and $327.4km^2$ in a crop field respectively.

• Journal of Korean Society of Forest Science
• /
• v.88 no.2
• /
• pp.157-168
• /
• 1999

This paper analyzed vegetation and land use classification, slope and permanent preservation and legal reserves on the farm Jangada and Jamaica-Mato Grosso do Sul, Brazil, using satellite image for assisting the planning of planting Eucalyptus spp. This part of the State of Mato Grosso do Sul represents an important geopolitcal area, since it is located on the borders of Bolivia and Paraguay. Also exportation of goods can be achieved through hydrovias extending to Buenos Aires, Argentina-through the Paraguay River. Also there are road and railroad connection which link the soutreastern part of Brazil to the Andean countries. The vegetation map from sheet SF 21-Campo Grande of the RADAMBRASIL Project was used as the basis for the preliminary interpretation of coverage, and complemented by a visit of the field. After the initial interpretation of the image, definition of classes of use and land occupation were made, and files of spectral signatures were created. On the farms Jamaica and Jangada Open Arboreal Savanna and Grass Savanna are the predominant physiognomies occupying 68% of total area. In spite of the results being satisfactory at the present moment, the development of this project should be revised and adjusted based on the evaluations already made, including a greater detailing of environmental components, principally with respect to soil and topography.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.2
• /
• pp.456-477
• /
• 2024

With information technology's rapid development, the Internet faces serious security problems. Studies have shown that malware has become a primary means of attacking the Internet. Therefore, adversarial samples have become a vital breakthrough point for studying malware. By studying adversarial samples, we can gain insights into the behavior and characteristics of malware, evaluate the performance of existing detectors in the face of deceptive samples, and help to discover vulnerabilities and improve detection methods for better performance. However, existing adversarial sample generation methods still need help regarding escape effectiveness and mobility. For instance, researchers have attempted to incorporate perturbation methods like Fast Gradient Sign Method (FGSM), Projected Gradient Descent (PGD), and others into adversarial samples to obfuscate detectors. However, these methods are only effective in specific environments and yield limited evasion effectiveness. To solve the above problems, this paper proposes a malware adversarial sample generation method (PixGAN) based on the pixel attention mechanism, which aims to improve adversarial samples' escape effect and mobility. The method transforms malware into grey-scale images and introduces the pixel attention mechanism in the Deep Convolution Generative Adversarial Networks (DCGAN) model to weigh the critical pixels in the grey-scale map, which improves the modeling ability of the generator and discriminator, thus enhancing the escape effect and mobility of the adversarial samples. The escape rate (ASR) is used as an evaluation index of the quality of the adversarial samples. The experimental results show that the adversarial samples generated by PixGAN achieve escape rates of 97%, 94%, 35%, 39%, and 43% on the Random Forest (RF), Support Vector Machine (SVM), Convolutional Neural Network (CNN), Convolutional Neural Network and Recurrent Neural Network (CNN_RNN), and Convolutional Neural Network and Long Short Term Memory (CNN_LSTM) algorithmic detectors, respectively.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.18 no.3
• /
• pp.127-134
• /
• 2016

Land suitability assessment for apples and pears was conducted with soil and climate information in South Korea. In doing so, we intended to preserve land and increase the productivity by providing valuable information regarding where more suitable areas for apples or pears are located. We used soil classification driven by soil environmental information system developed by National Institute of Agricultural Science, RDA, and also used climate classification in digital agro-climate map database for which is made by National Institute of Horticultural and Herbal Science. We combined both soil and climate classification results using a most-limiting characteristic method. The combined results showed very similar patterns with the results by classification based on soil information. Such results seem to come from the fact that the classification results by soil relatively lower than those by climate information. The results by soil classification seem to be too downgraded and checking if the final classification ranges in soil are reasonably made is strongly required. Although the most limiting characteristic method had been used widely in land suitability assessment, adapting the method based on results by soil and climate can be influenced by one downgraded factor. Therefore, alternative ways should be carefully considered for increasing the accuracy.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.16 no.4
• /
• pp.389-395
• /
• 2014

This technical note describes about the base stages of technology implementation for establishing "Early Warning System for Weather Hazard Management in Climate-smart Agriculture" to national onsite service. First of all, a special weather report service at catchment was represented sequential risk of 810 units of catchment by spatial statistical methods to existing 150 counties units special weather report released in KMA. The second, chronic hazard alarm service based on daily data of 76 Synoptic stations was monitor about 810 Catchment of mid-long term lapse weather and represented as a relative risk index chronic hazard risk of this time in preparation for the climatological normal conditions in the same period. Finally, we establish the foundation for delivering individually calculated field specific in hazard risk about volunteer farmer of early warning service demonstration area in seomjin downstream watershed. These three types of information were built a near real-time map service on the VWORLD background map of Ministry of Land as superposed layers nationwide catchment and demonstration areas within the farm unit weather hazard.

• Journal of Korean Society of Forest Science
• /
• v.81 no.1
• /
• pp.40-52
• /
• 1992

The use of meteorological information is essential in the industrial society. More specialized weather services are required to perform better industrial activities including forestry. A topoclimatological technique, in this study, which makes use of empirical relationships between the topography and the weather in Cheju Island was applied to produce reasonable estimates of monthly air temperatures over remote land area where routine observations are rare. Altitude values of the 250m grid points were first read from a 1 : 25000 topographic map. The mean altitude and other valuable topographical variables were then determined for each $1km^2$ land area. Daily minimum, maximum and mean air temperature data were collected from 19 points in Cheju Island from June 1987 to September 1988. The data were analyzed and grouped into 36 sets by type of air temperature and by month. Each of data set was regressed to the topographical variables to delineate empirical relationships between the local air temperature and the site topography. The total of 36 regression equations were finally selected and the equations were used to calculate the monthly air temperature for each $1km^2$ land area. The outputs were presented in a fine-mesh grid map with a 6-level contour capability.

• Korean Journal of Remote Sensing
• /
• v.36 no.6_1
• /
• pp.1367-1377
• /
• 2020

The purpose of this study is to assess the seismic vulnerability of buildings in Gyeongju city starting with the earthquake that occurred in the city on September 12, 2016, and produce a seismic vulnerability map. 11 influence factors related to geotechnical, physical, and structural indicators were selected to assess the seismic vulnerability, and these were applied as independent variables. For a dependent variable, location data of the buildings that were actually damaged in the 9.12 Gyeongju Earthquake was used. The assessment model was constructed based on random forest (RF) as a mechanic study method and support vector machine (SVM), and the training and test dataset were randomly selected with a ratio of 70:30. For accuracy verification, the receiver operating characteristic (ROC) curve was used to select an optimum model, and the accuracy of each model appeared to be 1.000 for RF and 0.998 for SVM, respectively. In addition, the prediction accuracy was shown as 0.947 and 0.926 for RF and SVM, respectively. The prediction values of the entire buildings in Gyeongju were derived on the basis of the RF model, and these were graded and used to produce the seismic vulnerability map. As a result of reviewing the distribution of building classes as an administrative unit, Hwangnam, Wolseong, Seondo, and Naenam turned out to be highly vulnerable regions, and Yangbuk, Gangdong, Yangnam, and Gampo turned out to be relatively safer regions.