Cold Regions Hydrological Model (CRHM) overview

C:\Program Files\CRHM\
QdroD
QdfoD
Qdro
Qdfo
SunMax
global
CalcHr
Qsi
calcsun
hru_t
hru_rh
hru_ea
hru_u
hru_p
hru_rain
hru_snow
hru_SunAct
hru_tmax
hru_tmin
hru_tmean
hru_eamean
hru_umean
hru_rhmean
hru_newsnow
t
rh
ea
u
p
ppt
Qsi
Qso
Qn
Qln
SunAct
form_data
obs
net_rain
net_snow
intcp_evap
intcp
net
cum_net
netall
Rn
Qg
Qs
net_rn
hru_evap
transp_on
evap
pb
Cold Regions Hydrological Model
Platform, CRHM:
Brief Introduction
Chris Marsh and John Pomeroy
Centre for Hydrology
University of Saskatchewan
Saskatoon
Cold Regions Hydrological Cycle
Snowfall
Sublimation
Blowing Snow
Ice
Snowmelt
Infiltration Frozen
Ground
Permafrost
Evaporation
Rainfall
Evapotranspiration
Sublimation
Runoff
Interflow
Cold Regions Hydrological Model
Platform: CRHM




Modular – purpose built from C++ modules
Modules based upon +45 years of hydrology research at
Univ. of Saskatchewan
No provision for calibration or optimization; parameters set
by knowledge
Hydrological Response Unit (HRU) basis







landscape unit with characteristic hydrological processes
single parameter set
horizontal interaction along flow cascade matrix
Model tracks state variables and flows for HRU
HRUs assumed to represent one response type, basis for
coupled energy and mass balance
HRUs connected aerodynamically for blowing snow and
via dynamic drainage networks for streamflow
Incorporate wetlands directly using fill and spill algorithm
Building Physically-based,
Distributed Hydrological Models
with CRHM




uses a library of physically-based
hydrological and energy balance process
modules;
handles the following aspects of the
modelling process:
 data pre-processing,
 module and model building,
 results analysis
is easy to use: employs a Windows
environment with pull-down menus.
has an extensive HELP file and open
module code
Cold Regions Hydrological Model
DATA COMPONENT
Preparation of spatial and meteorological data.






Spatial data (e.g. basin area, elevation, location, cover type) is
analyzed using an external Geographic Information System (GIS)
that assists the user in basin delineation, characterization and
parameterization of Hydrological Response Units (HRU). CRHM
takes in ArcGIS files with this information. The user can also
simply enter this information in a menu for less complex basins
Time-series meteorological data include air temperature,
humidity, wind speed, precipitation and radiation.
Adjustments for elevation (lapse rate), snowfall versus rainfall,
interpolation between input observations (stations)
Filters permit adjustment to data, changing time interval, creating
synthetic data using mathematical functions, interpolating data
Unit conversions to consistent SI units
Visualization of input data allows checking for quality
Observation Files






ASCII file created using weather station, atmospheric
model or other outputs from Excel, Matlab, IDL, etc.
Obs file sets model time step interval
Possible to have multiple obs files with varying time
steps
Interpolation, synthesis tools to fill in missing data or
create synthetic data
Observations are interpolated onto HRU by CRHM
Visualization tools useful in assessing reliability of
observations
Parameters






Parameters describe basin, define HRU and set
operation of process modules (examples: basin area,
elevation, tree height, soil type, fresh snow albedo)
Parameters set based upon understanding of basin and
HRU (limits imposed in model)
Assumption that substantial transferability of process
parameters exists for similar HRU
If not known from basin surveys then can be looked up
or guessed at
No formal facility to calibrate parameters exists in CRHM
Calibration possible using external programs such as
DDS.
Cold Regions Hydrological Model
MODEL COMPONENT

Utilizes Windows-based series of pull-down menus
linked to the system features.

Modules, (process algorithms) are selected from the
library and grouped together by the CRHM processor.

Modules have a set order of execution with a common
set of variables and parameters.

Modules are created in C++ programming language.

Macro modules can be created from within the model
using a simple macro language.
Hydrological Response Units


A HRU is a spatial unit in the basin that has 3 groups of
attributes

biophysical structure - soils, vegetation, drainage, slope,
elevation, area (determine from GIS, maps)

hydrological state – snow water equivalent, snow internal
energy, intercepted snow load, soil moisture,
depressional storage, lake storage, water table (track
using model)

hydrological flux - snow transport, sublimation,
evaporation, melt discharge, infiltration, drainage, runoff.
Fluxes are determined using fluxes from adjacent HRU
and so depend on location in a flow sequence.
HRU need not be spatially continuous but must
have some approximate geographical location or
location in a hydrological flow sequence
HRU Delineation





Driving meteorology:
temperature, humidty,
wind speed, snowfall,
rainfall, radiation
Blowing snow,
intercepted snow
Snowmelt and
evapotranspiration
Infiltration &
groundwater
Stream network
HRU Delineation
CRHM Routing





HRU routing conceptualizes more complex reality in
characteristic sequences
HRU to HRU
Can include lake, wetland, forest stand
Blowing snow and groundwater routed separately
from near surface and surface water
Flexible – HRU can route sequentially or
accumulate in an outflow HRU
HRU2
HRU5
HRU3
HRU4
HRU1
HRU0
HRU Routing Example
Fallow
Stubble
Grassland
Woodland
Wetland
Open
Water
(a)
River
Channel
RB outlet
Groups

A collection of modules executed in sequence.
module
Define group:
Model
incorporation:
module
module
Group
Group ‘A’
Module n
Module i
Group ‘B’
Groups

E.g. estimate sub-canopy SWE for forests of
differing leaf-area-index
Using Groups
to define many
sub-basins
Representative Basins (RB)
RB1
RB1
RB2
RB3
RB2
RB4
RB1






Permits upscaling of CRHM to large, complex basins
using groups to describe sub-basins
RB sub-basins determined in detail as assemblies of
HRU
RB types repeated with identical module structure
(Groups), similar parameters but differing geometry
Many RB types allowed in the larger basin
Muskingum routing module routes RBs through streams,
lakes, wetlands
Basin model is a network of RBs linked by a routing
module.
Representative Basin Routing
Example
RB 1
RB 2
RB 3
RB 5
RB 4
(b)
Smith Creek
basin outlet
Cold Regions Hydrological Model
ANALYSIS COMPONENT

Used to display, analyze and export results (Excel,
ASCII, Obs).

Statistical and graphical tools are used to analyze model
performance, allowing for decisions to be made on the
best modelling approach.

Sensitivity-analysis tools are provided to optimize
selected model parameters and evaluate the effects of
model parameters on simulation results.

Mapping tools use ArcGIS files to map outputs for
geographical visualization.
CRHM Modules
PROCESSES
DATA
ASSIMILATION


Data from multiple sites
Interpolation to the HRUs






SPATIAL
PARAMETERS

Basin and HRU parameters
are set. (area, latitude,
elevation, ground slope,
aspect)







Infiltration into soils (frozen and
unfrozen)
Snowmelt (open & forest)
Radiation – level, slopes
Wind speed variation – complex topo
Evapotranspiration
Blowing snow transport
Interception (snow & rain)
Sublimation (dynamic & static)
Soil moisture balance
Pond/depression storage
Surface runoff
Sub-surface runoff
Routing (hillslope & channel)
Radiation






α, Ts
Diffuse and Direct
Slopes
Longwave
Forest canopy
Albedo estimation
Limited data requirements
References: Brunt, Brutsaert,
Garnier & Ohmura,
Granger & Gray, Gray and Landine,
Pomeroy et al., Satterlund,
Sicart et al.
Forest Cover
Modules
New integrated
“canopy” module
Snow Interception & Sublimation
Snow Water Equivalent mm
140
Forest
120
Clearing
100
80
60
40
20
0
North
South
East
Jim MacDonald, MSc research
West
Interception: snow and rain
Sublimation
Evaporation
Rainfall
Snow
Snowfall
interception
Rainfall
Interception
Storage
Throughfall
Throughfall
Unloading
Drip
References: Liu, Granger & Pomeroy,
Hedstrom & Pomeroy, Parviainen & Pomeroy, Pomeroy et al 1998
Snowmelt
dU
,
Qm + Qn + Q H + Q E + QG + Q D =
dt
α
Qm
M=
,
ρw B hf
-Daily EB
-Hourly EB
-Advection
-SCA Depletion
-Meltwater Flow
-Degree Day
-Radiation Index
References: Gray & Landine, Kustas et al., Essery, Shook, Marks et al. 1999
Infiltration into Frozen Soils
Snowmelt
Water
INF = (1 − θ p ) SWE 0.584 or Parametric Equation
Unlimited
Infiltration
Limited
Infiltration
Soils
References: Granger et al., Gray et al., Zhao & Gray
Restricted
Infiltration
Runoff
Infiltration into Unfrozen Soils
Green Ampt Infiltration
Depends on Ponding Time
Iterative Solution
New module considers
Ayers infiltration rates
References: Green & Ampt; Ogden and Saghafian; Pietroniro in Pomeroy et al; Gray
Evaporation
Granger-Gray QN
or
Priestly-Taylor
or
Penman-Monteith
or
ShuttleworthWallace
G[ s (Q * −QG ) + C vdda / ra ]
QE =
sG + γ
Ta, ea, uz
QE
QH
QA=QN-QG
es = f (ea, Ta-Ts)
Ts, es, z0
Surface Vegetation & Soil
QG
Water drawn from
1) canopy
2) recharge zone
3) deep soil
References: Granger & Gray, Granger & Pomeroy,
Priestly & Taylor
Soil

- Depressional storage
- sub-HRU
- can form subsurface runoff or ground water recharge or fill and
spill.
- transfer of flows between HRUs
- Pond storage
all of HRU water covered.
- parameterization of maximum pond storage.
- possible to: (i) leak to subsurface flow or groundwater recharge
(ii) fill and spill.
Interflow between HRU
- subsurface flow can enter downhill HRU as surface or
subsurface flow.
-
-
“Soil”: permits dryland or pond water
balance and accounts for sub-HRU
depressional storage
Routing - Muskingum
The Muskingum method is based on a variable
discharge-storage relationship and is used to route the
runoff between HRUs in the RB. The routing storage
constant is estimated from the averaged length of HRU
to main channel and averaged flow velocity; the
average flow velocity is calculated by Manning’s
equation based on averaged HRU length to main
channel, average change in HRU elevation, overland
flow depth and HRU roughness;
Reference: Chow (1959, 1964)
CRHM Test - Yukon
Modular structure
HRU based
INF = 5 ⋅ (1 − θp ) ⋅ SWE 0.584
INF = C ⋅ S
2.92
0
⋅ (1 − S I )
1.64
 273.15 − TI 
⋅

 273.15 
−0.45
⋅ t0
0.44
Modelling Approach
Aggregated vs. Distributed
Basin discharge
2002
2003
1.0
Obs
Aggregated
Distributed
0.4
Q [m3/s]
Q [m3/s]
0.8
0.5
0.6
0.4
Obs
Aggregated
Distributed
0.3
0.2
0.2
0.1
0.0
5/01/02 5/09/02 5/17/02 5/25/02 6/02/02 6/10/02
0.0
4/17/03 4/26/03 5/05/03 5/14/03 5/23/03 6/01/03
Time [days]
Time [days]
CRHM Mountain Structure
Model Structure
Model Tests - SWE
Streamflow Prediction 2006
Mean Bias = -0.13
all parameters estimated from basin data
Streamflow Prediction 2007
Mean Bias = -0.068
all parameters estimated from basin data
Conclusions




CRHM has been a useful modelling platform to
explore the influence of model structure, basin
discretization, and process parameterisation on
hydrological cycle prediction.
The physical basis and parameter identifiability of
many modules in CRHM permits hydrological
prediction with minimal calibration.
Recent research has applied CRHM at larger
scales and to environments in South America,
Europe and Asia as well as agricultural, boreal,
mountain and arctic basins throughout Canada.
http://www.usask.ca/hydrology/CRHM.php

Download Report