The future flow (2020 to 2099) of the Khowai River has been simulated by the calibrated and validated model driven by bias-corrected CANESM2 climate datasets. This study can play a crucial role in assessing the future flow of the Khowai River Basin for proper planning and water management in this basin.
Background and Present State of the Problem
According to the locals, the heavy downpour and water rolling down the hills caused breaches at various points of the protective embankment of the Khowai River (The Daily Star, 2017). Residents of Habiganj town have been suffering from flooding for a long time due to an unplanned drainage system and illegal dredging of the Khowai river (The Daily Star, 2019).
Scope of the Study
This study intends to establish a HBV hydrologic model to assess the effect of climate change on the flow of the Khowai River Basin. Future changes in the frequency and magnitude of peak annual flow will be assessed through frequency analysis of the future flow data of the Shaistaganj station of the Khowai River.
Objectives of the Study
Possible Outcome
Organization of the Thesis
This chapter also includes previous studies on the Meghna River Basin and finally on the Khowai River Basin. The suitability of two global gridded datasets (NOAA versus ERA5) has been discussed in this chapter for streamflow modeling of the data-scarce Khowai River Basin using the HBV hydrological model.
The River System of Bangladesh
Khowai is one of the rivers connected to the Meghna river system which is a tributary of the Kushiyara river.
A Brief History of Hydrological Modeling
However, sometimes this physics-based model definition is reduced to only the laws of conservation of mass. For example, SWAT (Tool and Water Assessment Tools), one of the world's most widely used catchment models, is often called a physics-based model even though it does not strictly consider momentum and energy conservation (Spruill et al., 2000). .
Necessity of Catchment-Scale Hydrological Modeling
Physically based rainfall-runoff models express major hydrological processes in the form of fundamental mathematical equations of conservation of mass, momentum and energy (Freeze and Harlan, 1969; Valerity Y. Ivanov et al., 2004; Kavvas et al., 2004; Krysanova and Arnold, 2008; Meselhe et al., 2009). On the contrary, most global hydrological models (GHM) are usually applied for impact studies with a global parameterization, which compromises the quality of local performance for assumed good performance globally, that is, using a priori estimates of individual process parameters ( e.g. Vörösmarty et al. 2000), or after calibration only for selected large catchments (e.g. Döll et al. 2003, Widén-Nilsson et al. 2007), or combinations of these approaches (e.g. Nijssen et al. 2001 ).
Previous Studies Using HBV Model
Estimation of the variation of snowmelt contribution to streamflow at increased temperatures had been undertaken in the research. The HBV-light model was used to evaluate model performance in response to climate change in the snowy and ice-covered catchment of the Hunza River Basin.
Previous Studies Using Other Models in The Meghna Basin and The Khowai Basin
The model gives unsatisfactory results due to the lack of rainfall data in the upper catchment. Also, ignoring the effects of the upstream Chakmaghat dam on the catchment hydrology may be partially responsible for the model's unrealistic water balance. 2017) conducted a study on channel migration and its impact on the Khowai River in Tripura, Northeast India.
Climate Change Modeling for Hydrological Impact Assessment
Climate change scenarios
Arbitrary climate change scenarios are changes in key variables selected to test a system's sensitivity to possible climate change. These scenarios are most useful for testing the sensitivity of systems to changes in individual variables and combined changes (Narzis, 2020).
Emission Scenarios
The representative concentration pathways (RCPs)
CO2 concentration continues to rise, albeit at a slower rate in the latter parts of the century, reaching 620 ppm by 2100. This is the nightmare scenario where emissions continue to rise rapidly through the early and middle parts of the century.
Regional Climate Model Data Portal-CORDEX
CORDEX South Asia
The keys to success of this initiative in South Asia will be in developing a means of analyzing and translating CORDEX data into terms relevant to South Asia's knowledge needs, and in developing the internal capacity to carry out analyzes and thereby create expertise at regional levels in South Asia (CCCR, 2021).
Regional model: REGCM4 - a regional climate model system
General
Data Collection
Precipitation, temperature and evaporation data were collected from both ERA5 (Fifth Generation ECMWF - European Center for Medium-Range Weather Forecasts Reanalysis) Land Hourly Dataset under the Copernicus Climate Change Service (C3S), as well as from NOAA (National Oceanic and Atmospheric Administration) under NOAA Climate Prediction Center (CPC) Global Unified Gauge-Based Analysis. Water level and discharge data at Shaistaganj station (SW 158.1, NTQ) were collected from BWDB (Bangladesh Water Development Board). Prospective data on precipitation, temperature and evaporation from 1979 to 2099 were collected via CORDEX (Coulated Regional Climate Downscaling Experiment).
Data Pre-processing
Stream burning and watershed delineation
Precipitation, temperature and evaporation data preparation
A separate rainfall dataset was made by averaging the ERA5, NOAA and BMD (Habiganj) rainfall datasets for further use. Future ERA5, NOAA, and CORDEX datasets are downloaded gridded datasets for an area near the Khowai Basin. Using the Theissen polygon method in ArcGIS software, the nearest gridded datasets from the basin are weighted averaged and prepared as ERA5, NOAA and CORDEX (precipitation, temperature and evaporation) datasets.
Discharge data preparation
In this research, the cubic spline method was applied to interpolate the missing water level value. From the water level data and the estimation curve equations, the missing discharge is generated.
Calibration and Validation of HBV Model
- Description of HBV model
- General structure of HBV model
- Model parameters
- Description of four routines
- Input climate data
- Model setup
When each of the snow routine parameters (TT, CFMAX, SFCF, CFR and CWH), response function parameters (PERC, Alpha, UZL, K0, K1 and K2) and CET. As each of the Soil Routine parameters (FC, LP, and BETA) and the Routing Routine parameter (MAXBAS) were increased, the efficiency of the model output increased. The long-term average of evaporation can be corrected using temperature deviations from its long-term average.
Evaluation Criteria Based on Goodness of Fit Functions
Bias Correction of The Future Climate Data Set
Linear scaling method
To overcome these biases, future precipitation, temperature, and evaporation data from the CORDEX CANESM2 model were corrected to mean precipitation, ERA5 temperature, and evaporation using the linear scaling (LS) bias correction method. The linear scaling method calculates a ratio of the observed data (In this study, combined precipitation, ERA5 temperature and evaporation) and the corresponding raw data from CORDEX for the historical period. Where, correction factor (C.F.) for month i = Average of observed data for month i Average of raw data from climate model for month i.
The quantile mapping method
Then, the forward data is corrected by multiplying the forward data by this ratio (Narzis, 2020).
Simulation and Analysis of Future Change of Flow
Flow duration curve
Frequency analysis
General
Watershed Delineation
Elevation Wise Zone Distribution
Land Use Data Processing
The land use map is shown in Figure 4.2 and height distribution of the land use area is shown in Appendix (Table A 3).
Filling Gaps in Discharge Data
These discharge data gaps were filled by an estimation curve generated from Shaistaganj station WL and Q data. If data were missing in WL, the missing values were interpolated using the cubic splicing method of the NUM XL software. This data set of gap-filled discharges was used in the calibration and validation of the HBV model.
Model Calibration and Validation
- Results from calibration and validation with ERA5 datasets (trial 1)
- Results from calibration and validation with NOAA datasets (trial 2)
- Results from calibration and validation with merged datasets (trial 3)
- Comparison of the ERA5 and NOAA gridded datasets and merged dataset
The coefficient of determination (R2) for the observed and simulated discharge for the calibration and the validation period is 0.48 and 0.40 respectively (Figure 4.7). The coefficient of determination (R2) for the observed and simulated discharge for the calibration and validation period is 0.67 and 0.63 respectively on a daily time scale (Table 4.1). The Nash-Sutcliff coefficient (NSE) for the calibration and the validation period is 0.65 and 0.56 respectively (Table 4.1).
Model Parameter Uncertainty Analysis
IWFM (2020) developed a hydrological model to predict flash flood in the northeastern region of Bangladesh. Lack of observations of rainfall in the upper catchments outside Bangladesh also made their research challenging. Optimal parameters and parameter uncertainty are estimated by allowing single or multiple (up to all) parameters to vary within the limit stated in Table 3.4.
Bias Correction of The Future Climate Data Set
Linear scaling method
The quantile mapping method
The results of the quantile mapping bias correction for the temperature data are shown in Figure 4.16 and Figure 4.17. The results of the quantile mapping bias correction for the evaporation data are shown in Figure 4.18 and Figure 4.19. Since quantile mapping can more effectively correct the bias (Table 4.3) regarding the pooled rainfall, ERA5 temperature and evaporation, this approach will be used to correct the future climate data set.
Climate Change Impact Assessment
- Change in precipitation, temperature and evaporation due to climate change
- Future flow simulation
- Analysis of future change of flow
- Flow duration curve
- Frequency analysis
Monthly average daily AET and PET plots are given in Figure 4.25 and Figure 4.26 respectively. From April to December (Figure 4.31 to Figure 4.39), the highest flow for all the future periods increases compared to the base period. For August to December (Figure 4.35 to Figure 4.39), the lowest flow for all the future periods decreases compared to the base period.
Effect of Chakmaghat Barrage
A schematic diagram showing the effect of Chakmaghat dam on Shaistaganj station discharge is presented in Figure 4.43. The changes in mean monthly flow for the period 2020–2040 at Shaistaganj station for different flow diversion scenarios are shown in Table 4.8. Since flow diversion is activated only during the lean season (October-April), the monsoon flow remains the same for all diversion scenarios.
Discussion and Comparison with Past Studies
The difference in the findings on temperature changes between this study and Narzis (2020) may be due to choosing a different bias correction method and a different baseline period. Rahman (2016) has modeled the impacts of climate change on the water regime of the river-wetland systems in the transboundary Upper Meghna River basin (Bangladesh and India) and has shown that precipitation will increase by 20-30% in the future with respect to the base period for the Tripura region, where most of the Khowai Basin is located. 2014) shows that average evaporation in Bangladesh has decreased between January and April in recent decades.
Conclusion
Future precipitation, temperature and evaporation from the CANESM2 climate model under the RCP8.5 scenario were used to simulate future flow. The future flow (2020 to 2099) of the Khowai Basin is simulated by the calibrated and validated model, driven using bias-corrected future meteorological datasets as input. In addition, because the Khowai River is a transboundary river, the results of this study can also be used for the.
Recommendation
EFFECT ON WATER BALANCE OF TEESTA RIVER BASIN DUE TO VARIOUS CLIMATE CHANGES AND STREAM DEVELOPMENT. Impacts of climate change and upstream intervention on the hydrology of the Meghna River basin using SWAT. Modeling the impacts of climate change on the water regime of river-wetland systems in the transboundary Upper Meghna River Basin (Bangladesh and India) where data are scarce.
