• Proceedings of the KSRS Conference
• /
• 2008.10a
• /
• pp.204-207
• /
• 2008

The purpose of this study is to identify how much the MODIS NDVI (Normalized Difference Vegetation Index) can explain the soil moisture simulated from SWAT (Soil and Water Assessment Tool) continuous hydrological model. For the application, ChungjuDam watershed (6,661.3 $km^2$) was adopted which covers land uses of 82.2 % forest, 10.3 % paddy field, and 1.8 % upland crop respectively. For the preparation of spatial soil moisture distribution, the SWAT model was calibrated and verified at two locations (watershed outlet and Yeongwol water level gauging station) of the watershed using daily streamflow data of 7 years (2000-2006). The average Nash and Sutcliffe model efficiencies for the verification at two locations were 0.83 and 0.91 respectively. The 16 days spatial correlation between MODIS NDVI and SWAT soil moisture were evaluated especially during the NDVI increasing periods for forest areas.

• Proceedings of the KSRS Conference
• /
• 2009.03a
• /
• pp.101-104
• /
• 2009

기후 시스템에서 지구온난화는 세계적으로 매우 중요한 문제이고 이는 기후변화, 이상기온, 폭우, 가뭄 등의 문제를 초래한다. 특히 사막화는 전 세계적으로 10억 명 이상의 사람들에게 영향을 미치고 있다. 건조한 상태의 식생은 사막화되기 쉽기 때문에 식생의 수분상태는 사막화의 중요한 지표이다. 본 논문에서는 중국과 몽골 사막 주변영역에 대해 식생의 수분상태를 탐지하였다. 식생의 수분상태를 탐지하기 위해 1999년부터 2006년까지의 SPOT/VEGETATION 위성 이미지를 이용하여 정규화 수분지수(NDWI: Normalized Difference Water Index)를 산출하였다. 그 결과 1999년부터 2006년까지의 NDWI는 사막주변영역에서 감소하는 경향을 보였고, 그 영역은 몽골 고비사막 북동지역과 중국 타클라마칸 사막의 남동지역에 위치해 있었다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.140-140
• /
• 2019

본 연구에서는 Terra MODIS 위성자료와 Tensorflow를 활용해 1 km 공간 해상도의 토양수분을 산정하는 알고리즘을 개발하고, 국내 관측 자료를 활용해 검증하고자 한다. 토양수분 모의를 위한 입력 자료는 Terra MODIS NDVI(Normalized Difference Vegetation Index)와 LST(Land Surface Temperature), GPM(Global Precipitation Measurement) 강우 자료를 구축하고, 농촌진흥청에서 제공하는 1:25,000 정밀토양도를 기반으로 모의하였다. 여기서, LST와 GPM의 자료는 기상청의 종관기상관측지점의 LST, 강우 자료와 조건부합성(Conditional Merging, CM) 기법을 적용해 결측치를 보간하였고, 모든 위성 자료의 공간해상도를 1 km로 resampling하여 활용하였다. 토양수분 산정 기술은 인공 신경망(Artificial Neural Network) 모형의 딥 러닝(Deep Learning)을 적용, 기계 학습기반의 패턴학습을 사용하였다. 패턴학습에는 Python 라이브러리인 TensorFlow를 사용하였고 학습 자료로는 농촌진흥청 농업기상정보서비스에서 101개 지점의 토양수분 자료(2014 ~ 2016년)를 활용하고, 모의 결과는 2017 ~ 2018년까지의 자료로 검증하고자 한다.

• Korean Journal of Remote Sensing
• /
• v.30 no.3
• /
• pp.375-381
• /
• 2014

Soil moisture is an essential satellite-driven variable for understanding hydrologic, pedologic and geomorphic processes. The European Space Agency (ESA) has endorsed soil moisture as one of Climate Change Initiates (CCI) and had merged multi-satellites over 30 years. The $0.25^{\circ}$ coarse resolution soil moisture satellite data showed correlations with variables of a water stress index, Temperature-Vegetation Dryness Index (TVDI), from a stepwise regression analysis. The ancillary data from TVDI, Land Surface Temperature (LST) and Normalized Difference Vegetation Index (NDVI) from MODIS were inputted to a multi-regression analysis for estimating the surface soil moisture. The estimated soil moisture was validated with in-situ soil moisture data from April, 2012 to March, 2013 at Andong observation sites in South Korea. The soil moisture estimated using satellite-based LST and NDVI showed a good agreement with the observed ground data that this approach is plausible to define spatial distribution of surface soil moisture.

• Korean Journal of Remote Sensing
• /
• v.37 no.5_1
• /
• pp.1207-1214
• /
• 2021

In thisletter, we developed technology which can exclude effect of cloudsto perform remote waterbody detection based on Sentinel-2 optical satellite imagery to calculate the area of ungauged reservoirs and applied to the Hwanggang dam reservoir, a representative ungauged reservoir, to verify usability. The remote waterbody detection technology calculates the cloud blocking ratio by comparing the cloud boundary in the Sentinel-2 imagery and the reservoir boundary first. Next, itselects data whose cloud blocking ratio does not exceed a specific value and calculates NDWI (Normalized Difference Water Index) with selected imagery. In last, it calculatesthe area of the reservoir by counting the number of grids which have NDWI value considered as waterbody within the boundary of the target reservoir and correcting with cloud blocking ratio. To determine cloud blocking ratio threshold forselecting image, we performed the area calculation of Hwanggang dam reservoir from July 2018 to October 2021. As a result, when the cloud blocking ratio threshold wasset 10%, we confirmed that the result with large error due to clouds were filtered well and obtained 114 results that can show changes in Hwanggang dam reservoir area among 220 images.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.230-230
• /
• 2023

딥러닝 알고리즘 중 과거의 정보를 저장하는 문제(장기종속성 문제)가 있는 단순 RNN(Simple Recurrent Neural Network)의 단점을 해결한 LSTM(Long short-term memory)이 등장하면서 특정한 유역의 강우-유출 모형을 구축하는 연구가 증가하고 있다. 그러나 하나의 모형으로 모든 유역에 대한 유출을 예측하는 지역화 강우-유출 모형은 서로 다른 유역의 식생, 지형 등의 차이에서 발생하는 수문학적 행동의 차이를 학습해야 하므로 모형 구축에 어려움이 있다. 따라서, 본 연구에서는 국내 12개의 유역에 대하여 LSTM 기반 분포형 지역화 강우-유출 모형을 구축한 이후 강우 이외의 보조 자료에 따른 정확도를 살펴보았다. 국내 12개 유역의 7년 (2012.01.01-2018.12.31) 동안의 49개 격자(4km²)에 대한 10분 간격 레이더 강우, MODIS 위성 이미지 영상을 활용한 식생지수 (Normalized Difference Vegetation Index), 10분 간격 기온, 유역 평균 경사, 단순 하천 경사를 입력자료로 활용하였으며 10분 간격 유량 자료를 출력 자료로 사용하여 LSTM 기반 분포형 지역화 강우-유출 모형을 구축하였다. 이후 구축된 모형의 성능을 검증하기 위해 학습에 사용되지 않은 3개의 유역에 대한 자료를 활용하여 Nash-Sutcliffe Model Efficiency Coefficient (NSE)를 확인하였다. 식생지수를 보조 자료를 활용하였을 경우 제안한 모형은 3개의 검증 유역에 대하여 하천 흐름을 높은 정확도로 예측하였으며 딥러닝 모형이 위성 자료를 통하여 식생에 의한 차단 및 토양 침투와 같은 동적 요소의 학습이 가능함을 나타낸다.

• Journal of Astronomy and Space Sciences
• /
• v.36 no.4
• /
• pp.225-234
• /
• 2019

We explore the associations between the total sunspot area, solar north-south asymmetry, and Southern Oscillation Index and the physical characteristics of clouds by calculating normalized cross-correlations, motivated by the idea that the galactic cosmic ray influx modulated by solar activity may cause changes in cloud coverage, and in turn the Earth's climate. Unlike previous studies based on the relative difference, we have employed cloud data as a whole time-series without detrending. We found that the coverage of high-level and low-level cloud is at a maximum when the solar north-south asymmetry is close to the minimum, and one or two years after the solar north-south asymmetry is at a maximum, respectively. The global surface air temperature is at a maximum five years after the solar north-south asymmetry is at a maximum, and the optical depth is at a minimum when the solar north-south asymmetry is at a maximum. We also found that during the descending period of solar activity, the coverage of low-level cloud is at a maximum, and global surface air temperature and cloud optical depth are at a minimum, and that the total column water vapor is at a maximum one or two years after the solar maximum.

• Korean Journal of Remote Sensing
• /
• v.27 no.3
• /
• pp.235-258
• /
• 2011

The purpose of this study is to investigate climatological variations from the temporal and spatial surface particulate organic carbon (POC) estimates based on SeaWiFS spectral radiance, and to determine the physical mechanisms that affect the distribution of pac in the Gulf of Mexico. 7-year monthly mean values of surface pac concentration (Sept. 1997 - Dec. 2004) were estimated from Maximum Normalized Difference Carbon Index (MNDCI) algorithm using SeaWiFS data. Synchronous 7-year monthly mean values of remote sensing data (sea surface temperature (SST), sea surface wind (SSW), sea surface height anomaly (SSHA), precipitation rate (PR)) and recorded river discharge data were used to determine physical forcing factors. The spatial pattern of POC was related to one or more factors such as river runoff, wind-derived current, and stratification of the water column, the energetic Loop Current/Eddies, and buoyancy forcing. The observed seasonal change in the POC plume's response to wind speed in the western delta region resulted from seasonal changes in the upper ocean stratification. During late spring and summer, the low-density river water is heated rapidly at the surface by incoming solar radiation. This lowers the density of the fresh-water plume and increases the near-surface stratification of the water column. In the absence of significant wind forcing, the plume undergoes buoyant spreading and the sediment is maintained at the surface by the shallow pycnocline. However, when the wind speed increases substantially, wind-wave action increases vertical motion, reducing stratification, and the sediment were mixed downward rather than spreading laterally. Maximum particle concentrations over the outer shelf and the upper slope during lower runoff seasons were related to the Loop Current/eddies and buoyancy forcing. Inter-annual differences of POC concentration were related to ENSO cycles. During the El Nino events (1997-1998 and 2002-2004), the higher pac concentrations existed and were related to high runoffs in the eastern Gulf of Mexico, but the opposite conditions in the western Gulf of Mexico. During La Nina conditions (1999-2001), low Poe concentration was related to normal or low river discharge, and low PM/nutrient waters in the eastern Gulf of Mexico, but the opposite conditions in the western Gulf of Mexico.

• Journal of Korea Water Resources Association
• /
• v.33 no.4
• /
• pp.471-482
• /
• 2000

Surface energy balance components were evaluated by Landsat TM data and GIS with meteorological data. Calibration and validation for the applicability of this methodology were made through the estimating of the large-scale evapotranspiration (ET). In addition, sensitivity and error analysis was conducted to see the effects of the surface energy balance components on ET and the accuracy of each components. Bochong-chon located on the upper part of Guem River basin was selected as the case study area. Spatial distribution map of ET were produced for five dates: Jan. 1, Apr. 3, May. 10, and Nov. 27, 1995. The study results showed tat ET was greatly varied with the aspect and theland use type on the surface. In the case of having northeast and southeast in the aspect, ET was linearly increased depending on growing net radiation. While surface temperature has a high value, NDVI(Normalized Difference Vegetation Index) has a low value in the vegetated area. Therefore, ground heat flux was increased but ET was relatively decreased. The results of sensitivity and error analysis showed that net radiation is most sensitive and effective, ranging from 12.5% to 23.6% of sensitivity. Furthermore, the surface temperature, air temperature, and wind speed have the significant effects on ET estimation using remotely sensed data.

• Korean Journal of Remote Sensing
• /
• v.28 no.6
• /
• pp.597-609
• /
• 2012

Leaf area index (LAI) is important in explaining the ability of the crop to intercept solar energy for biomass production and in understanding the impact of crop management practices. This paper describes a procedure for estimating LAI as a function of image-derived vegetation indices from temporal series of IKONOS, Landsat TM, and MODIS satellite images using empirical models and demonstrates its use with data collected at Missouri field sites. LAI data were obtained several times during the 2002 growing season at monitoring sites established in two central Missouri experimental fields, one planted to soybean (Glycine max L.) and the other planted to corn (Zea mays L.). Satellite images at varying spatial and spectral resolutions were acquired and the data were extracted to calculate normalized difference vegetation index (NDVI) after geometric and atmospheric correction. Linear, exponential, and expolinear models were developed to relate temporal NDVI to measured LAI data. Models using IKONOS NDVI estimated LAI of both soybean and corn better than those using Landsat TM or MODIS NDVI. Expolinear models provided more accurate results than linear or exponential models.