1 AIT Asian Institute of Technology

Evaluation of global land use/land cover and precipitation products for hydrologic simulation in the Yom River Basin, Thailand

AuthorChirayut Chirachawala
Call NumberAIT Thesis no.WM-19-06
Subject(s)Climatic changes--Thailand--Yom River Basin
Hydrologic models--Computer simulation
Land use--Thailand--Yom River Basin

NoteA thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Water Engineering and Management
PublisherAsian Institute of Technology
AbstractThe types of data scarcity can be classified into many cases such as scarcity of climate data and scarcity of geospatial data (soil and land use data). Those are the main input to hydrological model for streamflow prediction. This study aims to evaluate global land use/ land cover and global precipitation products for hydrologic simulation and to understand more about uncertainty of global data sources. The Yom River Basin which located in the Northern part of Thailand was selected as a case study to investigate the performance of global products. Three different sources of global land cover products such as ESACCI, MCD12Q1, and GLC2000 were selected for this study. In terms of climate datasets, 9 sources of global precipitation and 3 sources global temperature were selected for assessment. The Soil and Water Assessment Tool (SWAT) hydrological model and the Hydrological Engineering Centre Hydrologic Modelling System (HEC-HMS) were used to simulate the streamflow in this research. The assessment of global land cover products was considered by (i) the similarity of land use classification, (ii) model performances, and (iii) hydrological indices. The results of similarity showed that ESACCI can capture well with local land use (LDD) which has the area of forest 60.1% and agriculture 20.5%. However, when land use/land cover products were assessed by using the confusion matrix (comparing the land use/land cover pixel by pixel), MCD12Q1 performed quite well which has the percentage of homogeneity 37.65%. The following is ESACCI and GLC2000 land cover product which has the percentage homogeneity 19.06% and 13.90%, respectively. In terms of model performance showed that for different sources of land use/ land cover products had similar results in terms of R2 and NSE both calibration and validation period. The model performance at the outlet point had R2 value of 0.72 and NSE of 0.70 in calibration period and R2 value of 0.73 and NSE of 0.73 in validation period. The hydrological indices showed that both of ESACCI and MCD12Q1 performed quite well in different purposes by ESACCI captured well with annual discharge, discharge in wet season (Qwet) and baseflow and MCD12Q1 captured well with high flow (Q5) and discharge in dry season (Qdry). The assessment of global climate products such as precipitation and temperature, the results of global precipitation showed that the statistical indicators (R2, SD, RMSE) for each product are very low (R2 < 0.5) in daily analysis. However, in monthly and annual analysis showed that TRMM 3B42 V7 performed better than other products which has an average R2 value of 0.78, SD of 100.2 mm, and RMSE of 17.4 mm in monthly analysis and an average R2 value of 0.52, SD of 136.4 mm, and RMSE of 101.6 mm in annual analysis. In addition, when considering the rainy and dry season, TRMM 3B42 V7 also showed good performance to capture heavy precipitation (P5), median precipitation (P50), and very less precipitation (P90). For the global temperature products showed that CPC global temperature performed very well compared to other products in both minimum and maximum temperature which has the less percentage difference from observed data in almost every month. The average R2 of iv CPC global temperature is 0.82 and 0.78 and RMSE is equal to 0.8 and 0.93˚C for minimum and maximum temperature, respectively. After finished the analysis of global precipitation and global temperature products, 5 sourced of global precipitation (TRMM3B42 V7, PERSIANN-CDR V1R1, CHIRPS V2.0, APHRODITE, and CNRM-CM5) and one source of global temperature which is CPC global temperature were selected an input to SWAT model with different land use/ landcover products. The results showed that TRMM 3B42 V7 performed very well compared to other products which has R2 value of 0.70, NSE of 0.70 during calibration period and R2 value of 0.63 and NSE of 0.62 during validation period. Considering hydrological indices indicated that ESACCI land cover product performed very well with TRMM 3B42 V7 satellite data which can capture the annual discharge, low flow (Q90), and Qdry by having the percentage difference from observed data 1.66%, 11.95%, and 3.01%, respectively. In addition, HEC-HMS model and its performance were compared with SWAT model. The results showed that model performances of HEC-HMS when (i) running by ground basegauged stations are very good in both calibration and validation period. ESACCI seemed to perform better than other land use/ land cover products. The results of model performance when (ii) running by global precipitation product (TRMM 3B42 V7), it showed the similar model performances in different land use/ land cover products in both periods. In calibration period, R2 value was about to 0.73 and NSE of 0.72. In validation period, R2 value was about to 0.70 and NSE of 0.68. ESACCI also showed the good performance with TRMM 3B42 V7 product which captured the annual discharge, low flow(Q90), high flow(Q5), Qwet, and base flow by showing percentage difference of 3.4%, 83.3%, 8.7%, 0.1%, and 3.4%, respectively. The results of this research are very useful for anyone who need to conduct their own research but they do not have any data available. They can use this research as a way to decide the source of global data products to address the data scarce-problem.
Year2019
TypeThesis
SchoolSchool of Engineering and Technology (SET)
DepartmentDepartment of Civil and Infrastucture Engineering (DCIE)
Academic Program/FoSWater Engineering and Management (WM)
Chairperson(s)Shrestha, Sangam;
Examination Committee(s)Babel, Mukand Singh;Virdis, Salvatore G.P.;Supattana Wichakul;
Scholarship Donor(s)His Majesty the King’s Scholarship (Thailand) ;
DegreeThesis (M.Eng.) - Asian Institute of Technology, 2019


Usage Metrics
View Detail0
Read PDF0
Download PDF0