Samenvatting Class notes - Catchment and Climate Hydrology

Vak
- Catchment and Climate Hydrology
- Teuling
- 2020 - 2021
- Wageningen University (Wageningen University, Wageningen)
- Earth and Environment
104 Flashcards en notities
1 Studenten
  • Deze samenvatting

  • +380.000 andere samenvattingen

  • Een unieke studietool

  • Een oefentool voor deze samenvatting

  • Studiecoaching met filmpjes

Onthoud sneller, leer beter. Wetenschappelijk bewezen.

Samenvatting - Class notes - Catchment and Climate Hydrology

  • 1615762800 1. Course Info and Context

  • What is hydrology?
    Branch of science concerned with the occurrence, distribution, circulation and properties of water on Earth
  • What is catchment hydrology?
    The study of hydrology in drainage basins. Typically at smaller scales of a hillslope or catchment. Focus on processes and models
  • What is climate hydrology?
    Study of interaction between terrestrial water cycle and the climate system. Typically larger spatial and temporal scales and across climate gradients. Focus on extremes and trends.
  • Give the mass balance for catchments?
    Eq: dS/dt = P-Q-ET 
    Rate of water storage change = Fluxin - Fluxout
  • 1615935600 2. Hydrological models and model evaluation

  • Is a complex model better because it represents more processes?
    No, generally not and especially not when following XX law.
  • Model optimization is an objective process. True/false?
    False, the modeller has to make many choices. Small changes in the manner of calculation or even starting values can have large effects in the results.
  • Give the definition for catchments and endorheic drainage basins and their difference?
    Catchment: area of land from which all runoff converges to a single point (outlet) at a lower elevation.
    Endorheic: do not drain to see, hence rivers might dry up. Endorheic drainage basins are inland basins that don't drain to an ocean. Around 18% of all land drains to endorheic lakes/seas/sinks.
  • Name 3 types of catchments and their characteristics
    • Endorheic: inland basins that do not drain to the ocean. 
    • Headwater: small catchments which don't contain significant confluences (higher mountains)
    • Mesoscale: area 1-1000 km2
    
  • Name two types of discharges and their difference?
    Discharge: volume flux of water leaving catchment via streamflow (l/s or m3/s)
    Specific discharge: normalized by catchment area (mm/h or mm/d)
  • Name one main catchment characteristic
    Catchments are a control volume and thus have boundaries.
  • Name 3 catchment functions
    Partition: interception, infiltration, percolation
    Storage: vegetation, depression, channel, lake, bank, detention, GW, retention, snow
    Release, ET, streamflow, GW

    Task modeler: represent these functions and their relations mathematically
  • Describe what a hydrograph is?
    • Discharge vs time graph past a specific point in space.
    • Typically measured at outlet
    • Limbs: rising/falling
    • Direction of time becomes visible (in contrast to precipitation time series that is very erratic)
  • What is the difference between plotting Q on a log scale vs a normal scale in a hydrograph? How can you use this to differentiatie between quickflow and baseflow?
    It shows linear and non-linear areas in hydrograph.
    Log is non-linear transformation. 

    Drawing linear line/inflection point, from this point, all Q is assumed to originate from (slow) linear GW reservoir. 
    Plot baseflow separation line. 
    Everything that behaves more non-linear than that (above this line), should be fast non-linear runoff processes.   

    All flow above this line is quickflow (=fast nonlinear runoff processes).
    Baseflow: maybe GW, should behave like linear reservoir, so should decay exponentially in time, so plotting on log scale it should be straight.
  • Why are hydro models generally poor when predicting extremes? Why is this a problem?  
    Models are too often build during normal conditions and we expect them to work well during extreme conditions, but the conditions might be very different.  

    Bc those are conditions that cause damage and are scientifically interesting.
  • Models have different levels. Assumptions/preconceptions/theories/physical laws can already be considered models. They are a form of ... Model? And what could they grow into?
    Perceptual model (qualitative). Often first building blocks of numerical models.
  • What is the main difference between conceptual and physically-based model and what do they have in common? 
    Conceptual models (reservoirs) do not have a differential equation, it's more a mass counting model. They are easier to work with. 
    Physically-based model have differential equations, which does not make them necessarily better.

    They both solve the same mass balance, and are in that sense both equally physical. 

    We'll mainly use conceptual models - lumped version.
  • Name the 5 steps of model development
    1. System conceptualization: determine lens of research question that you'll look through with your model
    2. Numerical implementation: details very important for model outcome
    3. Code verification: does your model reflect concepts you had in mind? During step 2 your model concept might change. 
    4. Model parameter calibration or optimization: optimize your parameters (based on training dataset)
    5. Model output validation: what's the value of your model outcome (rest of splitted dataset used for validation)
  • With which factors can an optimum in model complexity be achieved?
    • Predictive performance
    • Data availability
    • Model complexity


    >Optimum: more complexity reduces predictive performance 
  • Explain the difference between accurate and precise (and inaccurate and imprecise)? Use dartboards in your explanation. Which statistical concepts do they represent?
    Bias = accuracy
    Variance = precision
  • What is Occam's razor (law of parsimony)?
    Problem-solving principle at core of many theories in science, law of parsimony.
    Entities should not be multiplied without necessity.

    When presented with competing hypotheses that make the same predictions, one should always select the solution with the fewest assumptions. Note: the razor is not meant to be a way of choosing between hypotheses that make different predictions(!).
  • Name 3 implications of Ockhams' razor?
    • Determine how many assumptions and conditions are necessary for each explanation to be correct.
    • If an explanation requires extra assumptions or conditions, demand evidence proportional to the strength of each claim.
    • Extraordinary claims require extraordinary evidence.        
  • Name the 3 inputs and two outputs of a model simulation
    Input:
    • model parameters/variables
    • initial conditions (for all model state variables, typically have a differential equation behind them)
    • Model/atmospheric forcing (e.g. P, T, RH)

    Output:
    • Model variables (e.g. Q, ET)
    • State variables (e.g. S - left side differential eq)
  • What would leaf area index be in model input?
    Depends on model and purpose of model!
    In many models, it would be an input, e.g. Parameter that varies per month and influences ET. Vegetation could also be a state variable (as the thing you want to predict), co-evolving with water storage.
  • How to quantify how well a model fits the data?
    2nd half of lecture
  • What is the difference between Nash-Sutcliffe and Coefficient of determination/R^2?
    Mathematically same thing. 
    Nash-Sutfcliffe goes from - infinity to +1 in same equation

    If you use R^2 you do that on a linear fit (so you've already optimized the fit, the fit can never be worse than optimal/average?) 

    Min 70!!
  • Explain the drawbacks of NSE, based on slide 23 and 25 (paper by Schaefli and X) 
    See min 75. 

    25: Seasonality - maybe not using the mean as benchmark anymore due to seasonality. Nash values would take only average over the whole year.
  • What do Schaefli and Gupta suggest to improve the precipiration-based benchmark efficiency? Explain this in 4 steps
    Benchmark calendar day > includes seasonality. 
    4 steps as zero hypothesis
  • What is KGE and how does it differ from NSE?
    Range is similar, but interpretation of values is different. KGE solves 3rd issue with Nash. KGE takes correlation, bias and relative contr... And weighs them equally into the NSE equation. That's main reason why KGE is gaining popularity
  • How can NSE be deconstructed?
    -

    Smt with unequal weighing
  • Name and summarize the 3 issues with NSE and their current solutions.
    1. ? 
    2. Seasonality - solution suggested by Gupta and Schaefli(?)
    3. Unequal weighing of bias/variability and correlation - KGE proposes equal weighing.
  • What's a constrain of modelling all non-linear flow as runoff processes in a hydrograph, as is often done in models?
    You only look at catchment response and not at processes that are going on. Maybe diff explanations that would come up the same hydrograph.
  • Explain the main hydrological modeling concepts (learning goal)
    -
  • Explain something abou tthe common model performance metrics (learning goal)
    -
  • Give several trade-offs in model complexity and model evaluation (learning goal)
    -
  • What happens when you do a log transform on a hydrograph (Q-axis)? Give a mathematical example of a non-linear process.
    Log is non-linear transformation, thus if a certain process behaves non-linear, it will become a straight line on the log-scaled axis. 
    A non-linear process would be exponential decay.
  • How do accuracy (bias) and precision (var) behave in relation to model complexity and why? How can this be practically used?
    Variance increases, bias decreases, due to growing model flexibility. Superposition forms a minimum of optimal model complexity.
Lees volledige samenvatting
Deze samenvatting. +380.000 andere samenvattingen. Een unieke studietool. Een oefentool voor deze samenvatting. Studiecoaching met filmpjes.

Laatst toegevoegde flashcards

How do accuracy (bias) and precision (var) behave in relation to model complexity and why? How can this be practically used?
Variance increases, bias decreases, due to growing model flexibility. Superposition forms a minimum of optimal model complexity.
Explain the difference between accurate and precise (and inaccurate and imprecise)? Use dartboards in your explanation. Which statistical concepts do they represent?
Bias = accuracy
Variance = precision
What is the main difference between conceptual and physically-based model and what do they have in common? 
Conceptual models (reservoirs) do not have a differential equation, it's more a mass counting model. They are easier to work with. 
Physically-based model have differential equations, which does not make them necessarily better.

They both solve the same mass balance, and are in that sense both equally physical. 

We'll mainly use conceptual models - lumped version.
Models have different levels. Assumptions/preconceptions/theories/physical laws can already be considered models. They are a form of ... Model? And what could they grow into?
Perceptual model (qualitative). Often first building blocks of numerical models.
Why are hydro models generally poor when predicting extremes? Why is this a problem?  
Models are too often build during normal conditions and we expect them to work well during extreme conditions, but the conditions might be very different.  

Bc those are conditions that cause damage and are scientifically interesting.
What is the difference between plotting Q on a log scale vs a normal scale in a hydrograph? How can you use this to differentiatie between quickflow and baseflow?
It shows linear and non-linear areas in hydrograph.
Log is non-linear transformation. 

Drawing linear line/inflection point, from this point, all Q is assumed to originate from (slow) linear GW reservoir. 
Plot baseflow separation line. 
Everything that behaves more non-linear than that (above this line), should be fast non-linear runoff processes.   

All flow above this line is quickflow (=fast nonlinear runoff processes).
Baseflow: maybe GW, should behave like linear reservoir, so should decay exponentially in time, so plotting on log scale it should be straight.
What happens when you do a log transform on a hydrograph (Q-axis)? Give a mathematical example of a non-linear process.
Log is non-linear transformation, thus if a certain process behaves non-linear, it will become a straight line on the log-scaled axis. 
A non-linear process would be exponential decay.
What's a constrain of modelling all non-linear flow as runoff processes in a hydrograph, as is often done in models?
You only look at catchment response and not at processes that are going on. Maybe diff explanations that would come up the same hydrograph.
Describe what a hydrograph is?
  • Discharge vs time graph past a specific point in space.
  • Typically measured at outlet
  • Limbs: rising/falling
  • Direction of time becomes visible (in contrast to precipitation time series that is very erratic)
Model optimization is an objective process. True/false?
False, the modeller has to make many choices. Small changes in the manner of calculation or even starting values can have large effects in the results.