Assessment of climate change impacts on the surface water quality of Sewa River in Sierra Leon

다국어 초록 (Multilingual Abstract)

The effects of climate change on water resources management in Sierra Leone have significantly impacted the country's socio-economic situation. This study provides a context-specific quantitative assessment of climate change impacts on the surface water quality of Sewa River. The river serves as an important source of domestic, agricultural, and industrial water for the southern and eastern regions in Sierra Leone. The study integrates observed water quality and rainfall data spanning 2020- 2024 alongside downscaled KMA precipitation data of CMIP6 climate projections under the SSP1-2.6, and SSP5-8.5 emission pathways. Karl Pearson’s correlation and multiple linear regression (MLR) model were applied to establish rainfall- turbidity relationship, and generate future turbidity projections for early-century (2026-2050), mid-century (2051-2075), and end-century (2076-2100) periods. The results revealed that under SSP1-2.6, rainfall is projected to decline slightly within the remaining part of the 21st century with an average variability of 23%, while SSP5-8.5 suggests an early increase in rainfall but a slight decline in mid and late century with an average variability of 35%. The extreme event of 400mm/day in the high-emission scenario highlights high risk of floods and soil detachment. The all-season MLR model explained approximately 50% of turbidity (R2 = 0.5, adj- R2 = 0.43, p<0.001). Karl Pearson’s correlation indicated a strong rainfall-turbidity relationship (r = 0.7). Model validation showed good performance (RPD = 2.5%, Bias = 0.95 NTU, RMSE = 12 NTU, MAE = 10.47 NTU and R2 = 0.43). This result indicates reliable long-term performance which projects annual turbidity means of about 32.2-32.8 NTU for SSP1-2.6 and 33.4-33.8 NTU for SSP5-8.5 with high variability through early to end-century periods. These findings offer scientific evidence to guide NWRMA/SALWACO catchment measures: erosion control, early-warning system, turbidity-triggered treatment protocols, climate-resilient infrastructure design, and water quality management. Keywords: Climate change; Rainfall variability; Turbidity; CMIP6; SSP1-2.6; SSP5-8.5; Water quality; Sewa River; Sierra Leone

국문 초록 (Abstract)

기후 변화가 시에라리온의 수자원 관리에 미치는 영향은 국가의 사회·경제적 상황에 상당한 파급효과를 가져왔다. 본 연구는 시에라리온 남부와 동부 지역의 가정용·농업용·산업용 수자원으로 중요한 역할을 하는 세와강의 지표수 수질에 대한 기후 변화 영향을 정량적으로 평가하는 맥락 특화 연구이다. 이를 위해 2020–2024년 기간의 관측 수질 및 강수 자료와 CMIP6 기후 시나리오(SSP1-2.6, SSP5-8.5)에 대해 KMA가 다운스케일링한 일 강수 자료를 통합하고, Karl Pearson 상관분석과 다중선형회귀(MLR) 모형을 적용하여 강수–탁도 관계를 규명하고 세기 초(2026–2050년), 세기 중반(2051–2075년), 세기 후반(2076–2100년)의 미래 탁도 변화를 예측하였다.

SSP1-2.6 하에서는 21세기 잔여 기간 동안 강수가 평균 변동성 23% 수준에서 소폭 감소할 것으로 나타난 반면, SSP5-8.5 하에서는 초기에는 강수가 증가하나 세기 중·후반에는 다소 감소하는 경향을 보이며 평균 변동성은 35%로 나타났다. 고배출 시나리오에서 나타난 일 강수량 400 mm의 극한사상은 홍수 및 토양 분리(침식) 위험이 매우 높음을 시사한다. 전 계절(all-season)을 대상으로 구축된 MLR 모형은 강수에 따른 탁도 변동의 약 50%를 설명하였으며(R² = 0.50, adj. R² = 0.43, p<0.0002), Karl Pearson 상관계수 (r = 0.70)는 높은 양의 상관성을 보여주었다. 모형 검증 결과 RPD = 2.5%, Bias = 0.95, RMSE = 12 NTU, MAE = 10.47, R² = 0.43으로 높은 예측 성능을 확인하였다. 이는 장기 예측 성능이 신뢰할 만함을 의미하며, SSP1-2.6에서 연평균 탁도는 32.2–32.8 NTU, SSP5-8.5에서는 33.4–33.8 NTU 범위로, 세기 초·중·후반 전 기간에 걸쳐 높은 변동성을 동반할 것으로 전망된다.

이러한 결과는 침식 저감, 조기경보 시스템, 탁도 기반 정수처리 운전 프로토콜, 기후 회복력 기반 인프라 설계, 수질 관리 등 유역 관리 대책을 수립하는 데 있어 NWRMA–SALWACO에 과학적근거를 제공한다.

핵심어: 기후 변화, 강수 변동성, 탁도, CMIP6, SSP1-2.6, SSP5-8.5, 수질, 세와강, 시에라리온

목차 (Table of Contents)

• CHAPTER ONE 1
• INTRODUCTION 1
• 1.1 Background 1
• 1.2 Problem Statement 4
• 1.3 Justification of the Study 4
• 1.4 Research Objectives 5
• 1.5 Scope of the Study 6
• 1.6 Research Structure 6
• CHAPTER TWO 8
• LITERATURE REVIEW 8
• 2.1 Introduction. 8
• 2.2 Global Perspective on Climate Change 9
• 2.3 Climate Change from a Western African Regional Perspective 11
• 2.4 Evidence of Climate Change in Sierra Leone Over the Years 14
• 2.5 Relationship Between Climate Change and Water Resources 16
• 2.6 Climate Change Effect on the Hydrologic Cycle 17
• 2.7 Climate Change Effects on Water Resources in Africa 18
• 2.8 Climate Change and Water Quality in Africa 19
• 2.9 Impacts of Climate Change on Water Resources in Sierra Leone 20
• 2.10 Correlation and Linear Regression Model in Water Quality Analysis and Future Projections 21
• 2.11 Research Gap obtained from the Review 22
• 2.12 Methodological Approach 23
• CHAPTER THREE 24
• METHODOLOGY 24
• 3.1 Introduction 24
• 3.2 Study Plan 25
• 3.3 Study Area 25
• 3.4 Data Collection 27
• 3.5 Data Processing 30
• 3.6 Statistical Correlation Analysis 31
• 3.7 Multiple Linear Regression Model 32
• 3.8 Variable Selection 32
• 3.9 Goodness of Fit and Predictive Model Performance 33
• 3.10 Analytical Tools and Software 34
• 3.11 Research Considerations 34
• 3.12 Research Limitations 34
• 3.13 Synthesis 34
• CHAPTER FOUR 35
• RESULTS AND DISCUSSIONS 35
• 4.0 Introduction 35
• 4.1 Precipitation Projections for the Study Area Using CMIP6 Climate Model. 35
• 4.1.1 Projected Precipitation for the Year 2026-2100 Under CMIP6: SSP1-2. 6 & SSP5-8.5 Scenario. 36
• 4.1.2 Early-Century (2026-2050) Projected Precipitation under CMIP6: SSP1-2.6 & SSP5-8.5. 37
• 4.1.3 Mid-Century (2051-2075) Projected Precipitation under CMIP6: SSP1-2.6 & SSP5-8.5. 41
• 4.1.4 End-Century (2076-2100) Precipitation Projections under CMIP6: SSP1-2.6 & SSP5-8.5 scenarios. 45
• 4.2 Historical Rainfall in Southern Province of Sierra Leone. 48
• 4.3 Correlation and Multiple Linear Regression Analysis on Surface Water Quality 50
• 4.3.1 Karl Pearson’s Correlation Coefficient and Multiple Linear Regression (MLR) Model 51
• 4.3.2 Historical Turbidity and Flow of the Sewa River Watershed 57
• 4.3.3 Turbidity Forecasting Using CMIP6: SSP1-2.6 and SSP 5-8.5 Scenarios for Precipitation. 60
• CHAPTER FIVE 64
• CONCLUSION 64
• 5.1 Summary of Key Findings 64
• 5.2 Conclusion 66
• 5.3 Contributions of the Research to Reduce the Knowledge Gap 67
• 5.4 Limitations of the Study 68
• 5.5 Future Research Considerations 69
• 6.0 Reference List 70
• 7.0 Appendix Tables 76
• Korean Abstract (요약) 78