6.15. jules_rivers.nml

This file sets the river routing options. It contains two namelists called JULES_RIVERS and JULES_OVERBANK.

River routing introduces two more grids to a JULES run: the river routing input grid and the river routing model grid. The river routing input grid must always be specified as a 2D grid in JULES_RIVERS_PROPS. This is not required to be identical to the input or the model grid. Internally the model compresses this to the river routing model grid, which is a 1D grid with length np_rivers, which is the number of valid routing points in the river routing input grid. All river routing output will be on the river routing model grid, or will be regridded to the model grid.

Note

The river routing code in JULES is still in development. Users should ensure that results are as expected, and provide feedback where deficiencies are identified.

6.15.1. JULES_RIVERS namelist members

JULES_RIVERS::l_rivers

Type:

logical

Default:

F

Switch for enabling river routing.

TRUE

Use the river routing algorithm specified by i_river_vn to route runoff along river pathways.

FALSE

No river routing.

JULES_RIVERS::i_river_vn

Type:

integer

Default:

None

Switch to select the river routing algorithm to use for river routing.

1

Use a UM-coupled JULES implementation of the TRIP model (see Oki et al. 1999). This value is not allowed in standalone JULES

2

Use a standalone JULES implementation of the RFM kinematic wave model (see Dadson and Bell 2010, Bell et al. 2007).

3

Use a standalone JULES implementation of the TRIP model (see Oki et al. 1999).

JULES_RIVERS::l_riv_overbank

Type:

logical

Default:

F

Switch for enabling river overbank inundation. Only used if l_rivers is TRUE.

TRUE

Calculate frac_fplain_lp, i.e. overbank inundation area as a fraction of gridcell area.

FALSE

No overbank inundation calculations

Note

If l_riv_overbank = FALSE, then optional namelist JULES_OVERBANK is not required.

JULES_RIVERS::nstep_rivers

Type:

integer

Permitted:

> 0

Default:

None

The number of model timesteps per routing timestep.

For example, nstep_rivers = 5 means that runoff will be accumulated for 5 model timesteps before being routed on the 5th timestep.

Warning

The river routing parameter values can be highly dependent on model resolution, so care is required by the user to ensure that appropriate values are selected, tested and adjusted as required.

Suggested values for global and high-resolution runs are listed below, however these should be treated as a starting point only.

RFM parameters - used if i_river_vn = 2

JULES_RIVERS::a_thresh

Type:

integer

Default:

None

Suggested:

1 (spatial resolution coarser than 20 km gridcells), ~10 (high-resolution)

The threshold drainage area (specified in number of cells) draining to a gridbox, above which the grid cell is considered to be a river point (see a_T in Bell et al. 2007:541).

Remaining points are treated as land (drainage area = 0) or sea (drainage area < 0). See Bell et al. (2007).

JULES_RIVERS::cland

Type:

real

Permitted:

> 0

Default:

None

Suggested:

0.20 m/s (global), 0.40 m/s (1 km resolution, Bell et al. 2007)

The land wave speed (kinematic wave speed for surface flow in a land grid box on the river routing grid, m s^-1). This is the speed at which water moves through surface soil in a non-river grid cell (even without major rivers, there are always minor water courses so these cells do still contribute flow to neighbouring cells).

JULES_RIVERS::criver

Type:

real

Permitted:

> 0

Default:

None

Suggested:

0.62 m/s (global), 0.50 m/s (1 km resolution, Bell et al. 2007)

The river wave speed (kinematic wave speed for surface flow in a river grid box on the river routing grid, m s^-1). This value should be close to the rivers_speed used by TRIP, but not identical because RFM makes different assumptions about e.g. meandering.

JULES_RIVERS::cbland

Type:

real

Permitted:

> 0

Default:

None

Suggested:

<= cland. 0.10 m/s (global), 0.05 m/s (1 km resolution, Bell et al. 2007)

The subsurface land wave speed (kinematic wave speed for subsurface flow in a land grid box on the river routing grid, m s^-1).

JULES_RIVERS::cbriver

Type:

real

Permitted:

> 0

Default:

None

Suggested:

<= criver. 0.15 m/s (global), 0.05 m/s (1 km resolution, Bell et al. 2007)

The subsurface river wave speed (kinematic wave speed for subsurface flow in a river grid box on the river routing grid, m s^-1).

JULES_RIVERS::retl

Type:

real

Permitted:

-1 to 1

Default:

None

Suggested:

0.005 (1 km resolution, Bell et al. 2007)

The (resolution dependent) land return flow fraction. Bell et al. (2007:Table1) suggested value 0.005. On non-river grid cells in the land mask: if retl>0 then fraction retl of the subsurface flow moves to the surface per routing timestep; if retl<0 then fraction retl of the surface flow moves to the subsurface per routing timestep.

JULES_RIVERS::retr

Type:

real

Permitted:

-1 to 1

Default:

None

Suggested:

0.005 (1 km resolution, Bell et al. 2007)

The (resolution dependent) river return flow fraction. On river grid cells in the land mask: if retr>0 then fraction retr of the subsurface flow moves to the surface per routing timestep; if retr<0 then fraction retr of the surface flow moves to the subsurface per routing timestep.

JULES_RIVERS::runoff_factor

Type:

real

Permitted:

> 0

Default:

None

Values !=1.0 are generally used to correct biases in precipitation when the model is forced with observed data It is highly recommended that this is set to 1.0 (i.e. no runoff adjustment).

TRIP parameters - used if i_river_vn = 1,3

JULES_RIVERS::rivers_speed

Type:

real

Permitted:

> 0

Default:

None

The effective river velocity (m s^-1). See Oki et al. (1999). rivers_speed should equal (river flow velocity / rivers_meander). A value of 0.4 can be used, while Oki et al. (1999) used a value of 0.5.

JULES_RIVERS::rivers_meander

Type:

real

Permitted:

> 0

Default:

None

The ratio of the actual to calculated river lengths in a river routing gridbox. See Oki et al. (1999). Oki & Sud (1998) called this the Meandering Ratio r_M and suggested an average global value of 1.4.

TRIP parameters - used if i_river_vn = 1

JULES_RIVERS::lake_water_conserve_method

Type:

integer

Default:

1

Selects different fields for use in water conservation of lake evaporation

1

fqw_surft: This is the moisture flux on each tile, in which case the inland water tile is used. Snow sublimation has already been removed from fqw_surft at the point in the code that this is used.

2

elake_surft: This is the lake evaporation component of fqw_surft. This avoids the impact that snow melt has on modifying fqw_surft.

See also

References:

• Arora VK & Boer GJ (2012). A variable velocity flow routing algorithm for GCMs. Journal of Geophysical Research D 104:30965-30979.

• Bell, V.A. et al. (2007) Development of a high resolution grid-based river flow model for use with regional climate model output. Hydrology and Earth System Sciences. 11 532-549

• Dadson, S.J. and Bell, V.A. (2010) Comparison of Grid-2-Grid and TRIP runoff routing schemes. Centre for Ecology & Hydrology Internal Report http://nora.nerc.ac.uk/10890/1/dadson_etal_2010_g2gtrip.pdf

• Dadson S.J. et al. (2011) Evaluation of a grid-based river flow model configured for use in a regional climate model. Journal of Hydrology. 411 238-250

• Falloon, P.D. et al (2007) New global river routing scheme in the Unified Model. Hadley Centre Technical Note 72, available from the Met Office Library.

• Jones R., Dadson, S. and Bell, V.A. (2007) Report on European grid-based river-flow modelling for application to Regional Climate Models. Met Office Hadley Centre deliverable report.

• Oki, T. and Sud, Y.C. (1998) Design of Total Runoff Integrating Pathways (TRIP)—A Global River Channel Network. Earth Interactions, 2: 1-37.

• Oki, T., et al (1999) Assessment of annual runoff from land surface models using Total Runoff Integrating Pathways (TRIP). Journal of the Meteorological Society of Japan. 77 235-255

6.15.2. JULES_OVERBANK namelist members

Warning

The overbank inundation parameter values can be highly dependent on model resolution, so care is required by the user to ensure that appropriate values are selected, tested and adjusted as required.

Suggested values for global and high-resolution runs are listed below, however these should be treated as a starting point only.

JULES_OVERBANK::l_riv_overbank

Type:

logical

Default:

F

Switch for enabling river overbank inundation. Only used if l_rivers is TRUE.

TRUE

Calculate overbank inundation area as a fraction of gridcell area.

FALSE

No overbank inundation calculations

Note

If l_riv_overbank = FALSE, no further variables are needed from this namelist.

JULES_OVERBANK::overbank_model

Type:

integer

Permitted:

1, 2, 3

Default:

none

Choice of model of overbank inundation.

1. Simple model using an allometric (scaling) relationship to estimate river width, without use of topographic data.

2. Simple model using allometric relationships to estimate river width and depth, and the Rosgen (1994) entrenchment ratio, without use of topographic data. When river flow rates are higher than the estimated bankfull flow, river width is constrained so that when river depth = 2 x bankfull depth then width = ent_ratio * bankfull width.

3. The inundated area is calculated using a hypsometric integral based on a lognormal area-altitude distribution and an allometric relationship to estimate river depth. The parameters of the lognormal distribution are specified via JULES_OVERBANK_PROPS. (This is the recommended approach.)

River depth allometry (used if overbank_model = 2 or 3)

Allometry is: (DEPTH in m) = riv_c * ( (SURFACE RIVER INFLOW in m3 s^-1) ^ riv_f) (Leopold & Maddock 1953:eqn2)

JULES_OVERBANK::riv_c

Type:

real

Default:

none

Permitted:

>=0 and <=(1/riv_a)

Suggested:

0.27 (global, from Andreadis et al. 2013)

Coefficient in the allometry for river depth (units are (m / ((m3/s)^riv_f)), i.e. dependent on the value of riv_f)

JULES_OVERBANK::riv_f

Type:

real

Default:

none

Permitted:

>=0 and <=(1-riv_b)

Suggested:

0.30 (global, from Andreadis et al. 2013)

Exponent in the allometry for river depth (dimensionless)

River width scaling (used if overbank_model = 1 or 2)

River width allometry

Allometry is: (WIDTH in m) = riv_a * ( (SURFACE RIVER INFLOW in m3 s^-1) ^ riv_b) (Leopold & Maddock 1953:eqn1)

JULES_OVERBANK::riv_a

Type:

real

Default:

none

Permitted:

>=0 and <=(1/riv_c)

Suggested:

7.20 (global, from Andreadis et al. 2013)

Coefficient in the allometry for river width (units are (m / ((m3/s)^riv_b)), i.e. dependent on the value of riv_b)

JULES_OVERBANK::riv_b

Type:

real

Default:

none

Permitted:

>=0 and <=(1-riv_f)

Suggested:

0.50 (global, from Andreadis et al. 2013)

Exponent in the allometry for river width (dimensionless)

Bankfull flow allometry (used if overbank_model = 2)

Allometry is: (BANKFULL DISCHARGE RATE QBF in m3 s^-1) = coef_b * ( (CONTRIBUTING AREA in km2) ^ exp_c ) (see e.g. Andreadis et al. 2013)

JULES_OVERBANK::coef_b

Type:

real

Default:

none

Suggested:

0.08 (for “several drainages in western Washington State, USA”, Cragun 2005)

Coefficient in the allometry for bankfull flow (see Sen 2018:eqn3.33).

JULES_OVERBANK::exp_c

Type:

real

Default:

none

Suggested:

0.95 (for “several drainages in western Washington State, USA”, Cragun 2005)

Exponent in the allometry for bankfull flow (see Sen 2018:eqn3.33).

JULES_OVERBANK::ent_ratio

Type:

real

Default:

none

The Rosgen entrenchment ratio (single value for all water courses in the simulation): when river depth = 2 x bankfull depth then width = ent_ratio * bankfull width (i.e. ent_ratio can be used to specify how wide floodplains are allowed to be).

See also

References:

• Andreadis KM, Schumann GJ & Pavelsky T (2013). A simple global river bankfull width and depth database. Water Resources Research 49:7164-7168

• Cragun WS (2005). Discharge-Area relations from Selected Drainages on the Colorado Plateau: A GIS Application. Utah State University, http://hydrology.usu.edu/giswr/archive05/scragun/termproject/

• Leopold LB & Maddock T (1953). The Hydraulic Geometry of Stream Channels and Some Physiographic Implications. United States Geological Survey Professional Papers 252:1-57

• Rosgen DL (1994). A classification of natural rivers. Catena 22:169-199.

• Sen Z (2018). Flood Modeling, Prediction and Mitigation. Springer.