• Journal of Wetlands Research
• /
• v.21 no.1
• /
• pp.1-8
• /
• 2019

In this study, the multi-objective optimization method is attemped to optimize the hydrological model to estimate the runoff through two hydrological processes. HL-RDHM, a distributed hydrological model that can simultaneously estimate the amount of snowfall and runoff, was used as the distributed hydrological model. The Durango River basin in Colorado, USA, was selected as the watershed. MOSCEM was used as a multi-objective optimization method and parameter calibration and hydrologic model optimization were tried by selecting 5 parameters related to snow melting and 13 parameters related to runoff. Data from 2004 to 2005 were used to optimize the model and verified using data from 2001 to 2004. By optimizing both the amount of snow and the amount of runoff, the RMSE error can be reduced from 7% to 40% of the simulation value based on the initial solution at three SNOTEL points based on the RMSE. The USGS observation point of the outflow is improved about 40%.

• Journal of the Korean Society of Industry Convergence
• /
• v.24 no.3
• /
• pp.343-350
• /
• 2021

Unmanned aerial vehicles(UAV) photogrammetry provides a cost-effective option for collecting high-resolution 3D point clouds compared with UAV LiDAR and aerial photogrammetry. The main objectives of this study were to (1) validate the accuracy of 3D site model generated, and (2) determine the differences between Digital Elevation Model(DEM) and Digital Surface Model(DSM). In this study, DEM and DSM were shown to have varying degree of accuracy from observed data. The results indicated that the model predictions were considered tend to over- and under-estimated. The range of RMSE of DSM predicted was from 8.2 and 21.3 when compared with the observation. In addition, RMSE values were ranged from 10.2 and 25.8 to compare between DEM predicted and field data. The predict values resulting from the DSM were in agreement with the observed data compared to DEM calculation. In other words, it was determined that the DSM was a better suitable model than DEM. There is potential for enabling automated topography evaluation of the prior-harvest areas by using UAV technology.

• Journal of Korean Society of Water and Wastewater
• /
• v.35 no.6
• /
• pp.417-424
• /
• 2021

Algal bloom is an ongoing issue in the management of freshwater systems for drinking water supply, and the chlorophyll-a concentration is commonly used to represent the status of algal bloom. Thus, the prediction of chlorophyll-a concentration is essential for the proper management of water quality. However, the chlorophyll-a concentration is affected by various water quality and environmental factors, so the prediction of its concentration is not an easy task. In recent years, many advanced machine learning algorithms have increasingly been used for the development of surrogate models to prediction the chlorophyll-a concentration in freshwater systems such as rivers or reservoirs. This study used a light gradient boosting machine(LightGBM), a gradient boosting decision tree algorithm, to develop an ensemble machine learning model to predict chlorophyll-a concentration. The field water quality data observed at Daecheong Lake, obtained from the real-time water information system in Korea, were used for the development of the model. The data include temperature, pH, electric conductivity, dissolved oxygen, total organic carbon, total nitrogen, total phosphorus, and chlorophyll-a. First, a LightGBM model was developed to predict the chlorophyll-a concentration by using the other seven items as independent input variables. Second, the time-lagged values of all the input variables were added as input variables to understand the effect of time lag of input variables on model performance. The time lag (i) ranges from 1 to 50 days. The model performance was evaluated using three indices, root mean squared error-observation standard deviation ration (RSR), Nash-Sutcliffe coefficient of efficiency (NSE) and mean absolute error (MAE). The model showed the best performance by adding a dataset with a one-day time lag (i=1) where RSR, NSE, and MAE were 0.359, 0.871 and 1.510, respectively. The improvement of model performance was observed when a dataset with a time lag up of about 15 days (i=15) was added.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1125-1139
• /
• 2022

In an unmanned aerial vehicles (UAVs) system, a physical offset can be existed between the global positioning system/inertial measurement unit (GPS/IMU) sensor and the observation sensor such as a hyperspectral sensor, and a lidar sensor. As a result of the physical offset, a misalignment between each image can be occurred along with a flight direction. In particular, in a case of multi-sensor system, an observation sensor has to be replaced regularly to equip another observation sensor, and then, a high cost should be paid to acquire a calibration parameter. In this study, we establish a precise sensor model equation to apply for a multiple sensor in common and propose an independent physical offset estimation method. The proposed method consists of 3 steps. Firstly, we define an appropriate rotation matrix for our system, and an initial sensor model equation for direct-georeferencing. Next, an observation equation for the physical offset estimation is established by extracting a corresponding point between a ground control point and the observed data from a sensor. Finally, the physical offset is estimated based on the observed data, and the precise sensor model equation is established by applying the estimated parameters to the initial sensor model equation. 4 region's datasets(Jeon-ju, Incheon, Alaska, Norway) with a different latitude, longitude were compared to analyze the effects of the calibration parameter. We confirmed that a misalignment between images were adjusted after applying for the physical offset in the sensor model equation. An absolute position accuracy was analyzed in the Incheon dataset, compared to a ground control point. For the hyperspectral image, root mean square error (RMSE) for X, Y direction was calculated for 0.12 m, and for the point cloud, RMSE was calculated for 0.03 m. Furthermore, a relative position accuracy for a specific point between the adjusted point cloud and the hyperspectral images were also analyzed for 0.07 m, so we confirmed that a precise data mapping is available for an observation without a ground control point through the proposed estimation method, and we also confirmed a possibility of multi-sensor fusion. From this study, we expect that a flexible multi-sensor platform system can be operated through the independent parameter estimation method with an economic cost saving.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.20 no.4
• /
• pp.376-385
• /
• 2018

To estimate the hourly temperature distribution due to solar radiation during the day, on slope in complex terrain, an empirical formula was developed including the hourly deviation in the observed temperature following solar radiation deviation, at weather stations on the east-facing and west-facing slopes. The solar radiation effect was simulated using the empirical formula to estimate hourly temperature at 11 weather observation sites in mountainous agricultural areas, and the result was verified for the period from January 2015 to December 2017. When the estimated temperature was compared with the control, only considering temperature lapse rate, it was found that the tendency to underestimate the temperature from 9 am to 3 pm was reduced with the use of an empirical formula in the form of linear expression; consequently, the estimation error was reduced as well. However, for the time from 5 pm to 6 pm, the estimation error was smaller when a hyperbolic equation drawn from the deviation in solar radiation on the slope, which was calculated based on geometric conditions, was used instead of observed values. The reliability of estimating the daytime temperature at 3 pm was compared with existing estimation model proposed in other studies; the estimation error could be mitigated up to an ME (mean error) of $-0.28^{\circ}C$ and RMSE (root mean square error) of $1.29^{\circ}C$ compared to the estimation error in previous models (ME $-1.20^{\circ}C$, RMSE $2.01^{\circ}C$).

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.28 no.4
• /
• pp.469-476
• /
• 2022

This study developed a Recurrent Neural Network model implemented through Long Short-Term Memory (LSTM) that generates long-term tidal level data at Busan Port using tide observation data. The tide levels in Busan Port were predicted by the Korea Hydrographic and Oceanographic Administration (KHOA) using the tide data observed at Busan New Port and Tongyeong as model input data. The model was trained for one month in January 2019, and subsequently, the accuracy was calculated for one year from February 2019 to January 2020. The constructed model showed the highest performance with a correlation coefficient of 0.997 and a root mean squared error of 2.69 cm when the tide time series of Busan New Port and Tongyeong were inputted together. The study's finding reveal that long-term tidal level data prediction of an arbitrary port is possible using the deep learning recurrent neural network model.

• Atmosphere
• /
• v.33 no.4
• /
• pp.423-440
• /
• 2023

The updated version of Global Ocean Data Assimilation and Prediction System (GODAPS) in the NIMS/KMA (National Institute of Meteorological Sciences/Korea Meteorological Administration), which has been in operation since December 2021, is being introduced. This technical note on GODAPS2 describes main progress and updates to the previous version of GODAPS, a software tool for the operating system, and its improvements. GODAPS2 is based on Forecasting Ocean Assimilation Model (FOAM) vn14.1, instead of previous version, FOAM vn13. The southern limit of the model domain has been extended from 77°S to 85°S, allowing the modelling of the circulation under ice shelves in Antarctica. The adoption of non-linear free surface and variable volume layers, the update of vertical mixing parameterization, and the adjustment of isopycnal diffusion coefficient for the ocean model decrease the model biases. For the sea-ice model, four vertical ice layers and an additional snow layer on top of the ice layers are being used instead of previous single ice and snow layers. The changes for data assimilation include the updated treatment for background error covariance, a newly added bias scheme combined with observation bias, the application of a new bias correction for sea level anomaly, an extension of the assimilation window from 1 day to 2 days, and separate assimilations for ocean and sea-ice. For comparison, we present the difference between GODAPS and GODAPS2. The verification results show that GODAPS2 yields an overall improved simulation compared to GODAPS.

• Journal of Astronomy and Space Sciences
• /
• v.37 no.2
• /
• pp.105-115
• /
• 2020

The Korean Institute of Technology Satellite (KITSAT-1) is the first satellite developed by the Satellite Technology Research Center and the University of Surrey. KITSAT-1 is orbiting the Earth's orbit as space debris with a 1,320 km altitude after the planned mission. Due to its relatively small size and altitude, tracking the KITSAT-1 was a difficult task. In this research, we analyzed the tracking results of KITSAT-1 for one year using the Midland Space Radar (MSR) in Texas and the Poker Flat Incoherent Scatter Radar (PFISR) in Alaska operated by LeoLabs, Inc. The tracking results were analyzed on a weekly basis for MSR and PFISR. The observation was conducted by using both stations at an average frequency of 10 times per week. The overall corrected range measurements for MSR and PFISR by LeoLabs were under 50 m and 25 m, respectively. The ionospheric delay, the dominant error source, was confirmed with the International Reference of Ionosphere-16 model and Global Navigation Satellite System data. The weekly basis orbit determination results were compared with two-line element data. The comparison results were used to confirm the orbital consistency of the estimated orbits.

• Atmosphere
• /
• v.24 no.1
• /
• pp.101-110
• /
• 2014

This study estimates and evaluates the daily January temperature from 2003 to 2012 with 30 m-resolution over South Korea, using a modified Parameter-elevation Regression on Independent Slopes Model (K-PRISM). Several factors in K-PRISM are also adjusted to 30 m grid spacing and daily time scales. The performance of K-PRISM is validated in terms of bias, root mean square error (RMSE), and correlation coefficient (Corr), and is then compared with that of inverse distance weighting (IDW) and hypsometric methods (HYPS). In estimating the temperature over Jeju island, K-PRISM has the lowest bias (-0.85) and RMSE (1.22), and the highest Corr (0.79) among the three methods. It captures the daily variation of observation, but tends to underestimate due to a high-discrepancy in mean altitudes between the observation stations and grid points of the 30 m topography. The temperature over South Korea derived from K-PRISM represents a detailed spatial pattern of the observed temperature, but generally tends to underestimate with a mean bias of -0.45. In bias terms, the estimation ability of K-PRISM differs between grid points, implying that care should be taken when dealing with poor skill area. The study results demonstrate that K-PRISM can reasonably estimate 30 m-resolution temperature over South Korea, and reflect topographically diverse signals with detailed structure features.

• Journal of Korean Society for Atmospheric Environment
• /
• v.25 no.4
• /
• pp.304-315
• /
• 2009

Numerical simulations were carried out to understand the effect of Sea Surface Temperature (SST) spatial distribution on regional circulation. A three-dimensional non-hydrostatic atmospheric model RAMS, version 6.0, was applied to examine the impact of SST forcing on regional circulation. New Generation Sea Surface Temperature (NGSST) data were implemented to RAMS to compare the results of modeling with default SST data. Several numerical experiments have been undertaken to evaluate the effect of SST for initialization. First was the case with NGSST data (Case NG), second was the case with RAMS monthly data (Case RM) and third was the case with seasonally averaged RAMS monthly data (Case RS). Case NG showed accurate spatial distributions of SST but, the results of RM and RS were $3{\sim}4^{\circ}C$ lower than buoy observation data. By analyzing practical sea surface conditions, large difference in horizontal temperature and wind field for each run were revealed. Case RM and Case RS showed similar horizontal and vertical distributions of temperature and wind field but, Case NG estimated the intensity of sea breeze weakly and land breeze strongly. These differences were due to the difference of the temperature gradient caused by different spatial distributions of SST. Diurnal variations of temperature and wind speed for Case NG indicated great agreement with the observation data and statistics such as root mean squared error, index of agreement, regression were also better than Case RM and Case RS.