Rainfall-Runoff Modelling

Keith Beven is Professor of Hydrology and Fluid Dynamics at Lancaster University. He has published over 300 scientific papers many of which apply uncertainty estimation techniques to different environmental problems in the areas of surface and groundwater hydrology, hydraulics, ecology, water quality, pollutant transport in different environments, and flood risk assessment and forecasting. He is also the author of Rainfall-Runoff Modelling: The Primer and six other edited volumes.

• Preface to the Second Edition xiii About the Author xviiList of Figures xix1 Down to Basics: Runoff Processes and the Modelling Process 11.1 Why Model? 11.2 How to Use This Book 31.3 The Modelling Process 31.4 Perceptual Models of Catchment Hydrology 61.5 Flow Processes and Geochemical Characteristics 131.6 Runoff Generation and Runoff Routing 151.7 The Problem of Choosing a Conceptual Model 161.8 Model Calibration and Validation Issues 181.9 Key Points from Chapter 1 21Box 1.1 The Legacy of Robert Elmer Horton (1875–1945) 222 Evolution of Rainfall–Runoff Models: Survival of the Fittest? 252.1 The Starting Point: The Rational Method 252.2 Practical Prediction: Runoff Coefficients and Time Transformations 262.3 Variations on the Unit Hydrograph 332.4 Early Digital Computer Models: The Stanford Watershed Model and Its Descendants 362.5 Distributed Process Description Based Models 402.6 Simplified Distributed Models Based on Distribution Functions 422.7 Recent Developments: What is the Current State of the Art? 432.8 Where to Find More on the History and Variety of Rainfall–Runoff Models 432.9 Key Points from Chapter 2 44Box 2.1 Linearity, Nonlinearity and Nonstationarity 45Box 2.2 The Xinanjiang, ARNO or VIC Model 46Box 2.3 Control Volumes and Differential Equations 493 Data for Rainfall–Runoff Modelling 513.1 Rainfall Data 513.2 Discharge Data 553.3 Meteorological Data and the Estimation of Interception and Evapotranspiration 563.4 Meteorological Data and The Estimation of Snowmelt 603.5 Distributing Meteorological Data within a Catchment 613.6 Other Hydrological Variables 613.7 Digital Elevation Data 613.8 Geographical Information and Data Management Systems 663.9 Remote-sensing Data 673.10 Tracer Data for Understanding Catchment Responses 693.11 Linking Model Components and Data Series 703.12 Key Points from Chapter 3 71Box 3.1 The Penman–Monteith Combination Equation for Estimating Evapotranspiration Rates 72Box 3.2 Estimating Interception Losses 76Box 3.3 Estimating Snowmelt by the Degree-Day Method 794 Predicting Hydrographs Using Models Based on Data 834.1 Data Availability and Empirical Modelling 834.2 Doing Hydrology Backwards 844.3 Transfer Function Models 874.4 Case Study: DBM Modelling of the CI6 Catchment at Llyn Briane, Wales 934.5 Physical Derivation of Transfer Functions 954.6 Other Methods of Developing Inductive Rainfall–Runoff Models from Observations 994.7 Key Points from Chapter 4 106Box 4.1 Linear Transfer Function Models 107Box 4.2 Use of Transfer Functions to Infer Effective Rainfalls 112Box 4.3 Time Variable Estimation of Transfer Function Parameters and Derivation of Catchment Nonlinearity 1135 Predicting Hydrographs Using Distributed Models Based on Process Descriptions 1195.1 The Physical Basis of Distributed Models 1195.2 Physically Based Rainfall–Runoff Models at the Catchment Scale 1285.3 Case Study: Modelling Flow Processes at Reynolds Creek, Idaho 1355.4 Case Study: Blind Validation Test of the SHE Model on the Slapton Wood Catchment 1385.5 Simplified Distributed Models 1405.6 Case Study: Distributed Modelling of Runoff Generation at Walnut Gulch, Arizona 1485.7 Case Study: Modelling the R-5 Catchment at Chickasha, Oklahoma 1515.8 Good Practice in the Application of Distributed Models 1545.9 Discussion of Distributed Models Based on Continuum Differential Equations 1555.10 Key Points from Chapter 5 157Box 5.1 Descriptive Equations for Subsurface Flows 158Box 5.2 Estimating Infiltration Rates at the Soil Surface 160Box 5.3 Solution of Partial Differential Equations: Some Basic Concepts 166Box 5.4 Soil Moisture Characteristic Functions for Use in the Richards Equation 171Box 5.5 Pedotransfer Functions 175Box 5.6 Descriptive Equations for Surface Flows 177Box 5.7 Derivation of the Kinematic Wave Equation 1816 Hydrological Similarity, Distribution Functions and Semi-Distributed Rainfall–Runoff Models 1856.1 Hydrological Similarity and Hydrological Response Units 1856.2 The Probability Distributed Moisture (PDM) and Grid to Grid (G2G) Models 1876.3 TOPMODEL 1906.4 Case Study: Application of TOPMODEL to the Saeternbekken Catchment, Norway 1986.5 TOPKAPI 2036.6 Semi-Distributed Hydrological Response Unit (HRU) Models 2046.7 Some Comments on the HRU Approach 2076.8 Key Points from Chapter 6 208Box 6.1 The Theory Underlying TOPMODEL 210Box 6.2 The Soil and Water Assessment Tool (SWAT) Model 219Box 6.3 The SCS Curve Number Model Revisited 2247 Parameter Estimation and Predictive Uncertainty 2317.1 Model Calibration or Conditioning 2317.2 Parameter Response Surfaces and Sensitivity Analysis 2337.3 Performance Measures and Likelihood Measures 2397.4 Automatic Optimisation Techniques 2417.5 Recognising Uncertainty in Models and Data: Forward Uncertainty Estimation 2437.6 Types of Uncertainty Interval 2447.7 Model Calibration Using Bayesian Statistical Methods 2457.8 Dealing with Input Uncertainty in a Bayesian Framework 2477.9 Model Calibration Using Set Theoretic Methods 2497.10 Recognising Equifinality: The GLUE Method 2527.11 Case Study: An Application of the GLUE Methodology in Modelling the Saeternbekken MINIFELT Catchment, Norway 2587.12 Case Study: Application of GLUE Limits of Acceptability Approach to Evaluation in Modelling the Brue Catchment, Somerset, England 2617.13 Other Applications of GLUE in Rainfall–Runoff Modelling 2657.14 Comparison of GLUE and Bayesian Approaches to Uncertainty Estimation 2667.15 Predictive Uncertainty, Risk and Decisions 2677.16 Dynamic Parameters and Model Structural Error 2687.17 Quality Control and Disinformation in Rainfall–Runoff Modelling 2697.18 The Value of Data in Model Conditioning 2747.19 Key Points from Chapter 7 274Box 7.1 Likelihood Measures for use in Evaluating Models 276Box 7.2 Combining Likelihood Measures 283Box 7.3 Defining the Shape of a Response or Likelihood Surface 2848 Beyond the Primer: Models for Changing Risk 2898.1 The Role of Rainfall–Runoff Models in Managing Future Risk 2898.2 Short-Term Future Risk: Flood Forecasting 2908.3 Data Requirements for Flood Forecasting 2918.4 Rainfall–Runoff Modelling for Flood Forecasting 2938.5 Case Study: Flood Forecasting in the River Eden Catchment, Cumbria, England 2978.6 Rainfall–Runoff Modelling for Flood Frequency Estimation 2998.7 Case Study: Modelling the Flood Frequency Characteristics on the Skalka Catchment, Czech Republic 3028.8 Changing Risk: Catchment Change 3058.9 Changing Risk: Climate Change 3078.10 Key Points from Chapter 8 309Box 8.1 Adaptive Gain Parameter Estimation for Real-Time Forecasting 3119 Beyond the Primer: Next Generation Hydrological Models 3139.1 Why are New Modelling Techniques Needed? 3139.2 Representative Elementary Watershed Concepts 3159.3 How are the REW Concepts Different from Other Hydrological Models? 3189.4 Implementation of the REW Concepts 3189.5 Inferring Scale-Dependent Hysteresis from Simplified Hydrological Theory 3209.6 Representing Water Fluxes by Particle Tracking 3219.7 Catchments as Complex Adaptive Systems 3249.8 Optimality Constraints on Hydrological Responses 3259.9 Key Points from Chapter 9 32710 Beyond the Primer: Predictions in Ungauged Basins 32910.1 The Ungauged Catchment Challenge 32910.2 The PUB Initiative 33010.3 The MOPEX Initiative 33110.4 Ways of Making Predictions in Ungauged Basins 33110.5 PUB as a Learning Process 33210.6 Regression of Model Parameters Against Catchment Characteristics 33310.7 Donor Catchment and Pooling Group Methods 33510.8 Direct Estimation of Hydrograph Characteristics for Constraining Model Parameters 33610.9 Comparing Regionalisation Methods for Model Parameters 33810.10 HRUs and LSPs as Models of Ungauged Basins 33910.11 Gauging the Ungauged Basin 33910.12 Key Points from Chapter 10 34111 Beyond the Primer:Water Sources and Residence Times in Catchments 34311.1 Natural and Artificial Tracers 34311.2 Advection and Dispersion in the Catchment System 34511.3 Simple Mixing Models 34611.4 Assessing Spatial Patterns of Incremental Discharge 34711.5 End Member Mixing Analysis (EMMA) 34711.6 On the Implications of Tracer Information for Hydrological Processes 34811.7 Case Study: End Member Mixing with Routing 34911.8 Residence Time Distribution Models 35311.9 Case Study: Predicting Tracer Transport at the Gåardsjön Catchment, Sweden 35711.10 Implications for Water Quality Models 35911.11 Key Points from Chapter 11 360Box 11.1 Representing Advection and Dispersion 361Box 11.2 Analysing Residence Times in Catchment Systems 36512 Beyond the Primer: Hypotheses, Measurements and Models of Everywhere 36912.1 Model Choice in Rainfall–Runoff Modelling as Hypothesis Testing 36912.2 The Value of Prior Information 37112.3 Models as Hypotheses 37212.4 Models of Everywhere 37412.5 Guidelines for Good Practice 37512.6 Models of Everywhere and Stakeholder Involvement 37612.7 Models of Everywhere and Information 37712.8 Some Final Questions 378Appendix A Web Resources for Software and Data 381Appendix B Glossary of Terms 387References 397Index 449