Transcript New Results on Snow and Runoff Research

Modifications and Experiments
Default CLM2: precipitation occurs over the entire grid
cell
Interception Scheme: Convective precipitation occurs
only (RE Dickinson)
Pconv
f

conv
Plarge = 0. fconv = 10%
Pl arg e  10 * Pconv
Pconv = 0. fconv = 100%
Simplified Topmodel : more closely following TOPMODEL
but improved schemes for water table, saturated area,
surface runoff and baseflow; super-saturated water
recharged to unsaturated soil layers.
Frozen Soil: Relax frozen soil control on hydraulic
conductivity ( hsat increased in cold regions)
The NCAR CLM2-GSWP2
The CLM2 run with the GSWP2 3-hr forcing data at 1-hr time step
(1983-1995)
10-year monthly runoff outputs (1986-1995)
Improvement index =
12
 ( EXP  OBS) /
2
imon 1
12
2
(
Default

OBS
)

imon 1
<1
improvement
>1
degradation
TOPMODEL vs. CONTROL
Degrade
Improve
Frozen Soil Effects
Degrade
0
Improve
Runoff over snow-covered areas
UNH-GRDC Observed Runoff (mm/day)
The NCAR CLM2 simulated Runoff (mm/day)
CLM2 – USAF/ETAC observed Snow Depth (m)
The simulated runoff
in May is much
higher than the
GRDC observed in
North Europe and
West Siberia, because
the CLM2 produced
much more snow than
the USAF/ETAC
observed
Regional Averages (Arctic and Boreal Regions)
North Europe (55-70N, 5-60E)
West Siberia (50-70N, 60-90E)
Model
OBS
East Siberia (50-70N, 90-140E)
North Canada (50-70N, 90-170W)
Regional Averages (Tropical Regions)
Central America (10-25N, 60-110W)
Amazon Basin (10S-0N, 50-70W)
Congo (10S-5N, 10-30E)
India (10-30N, 70-90E)
Regional Averages (Midlatitudes)
Western U.S. (30-50N, 110-130W)
Central U.S. (30-50N, 90-110W)
Eastern U.S. (30-50N, 70-90W)
Central Europe (40-55N, 0-40E)
Regional Averages (Arid Regions)
Sahara+Arabian Peninsula(10-30N, 0-50E)
Southern South America (25-60S, 50-80W)
South Africa (10-35S, 10-40E)
Australia (10-40S, 110-160W)
Soil Moisture
(mm/day)
Surface Runoff
(mm/day)
Canopy Evaporation
(mm/day)
Amazon Region
The
new
interception
scheme
intercepts less water, allowing more
water to reach the ground and
increase surface runoff instead of
increasing soil moisture.
The simplified TOPMODEL largely
decreases the surface runoff, allowing
more water to infiltrate into deeper
soil and increase soil moisture.
Infiltration and the Water Table
• After a storm event, water infiltrates through the unsaturated zone,
and recharges the ground water reservoir
• This raises the elevation of the water table, increasing the gradient
and the flow of ground water away from the recharge point. This
results in increased flow at the discharge point
• Increased evaporation, or withdrawal by pumping, lowers the water
table by increasing the discharge from that reservoir
Aquifers
• Unconfined Aquifers have
an upper boundary defined
by the water table
• Confined aquifer
(Artesian): The upper
boundary is a confining
layer (Aquitard)
– The water level (head) in a
well penetrating confined
aquifer rises above the level
of the confining layer
– The surface defined by the
water level is the piezometric
(or potentiometric) surface,
and is not the water table
– Some are also flowing
artesian wells
Water Table
Simulations over the Amazon Basin
Topographic Index λ
Simulations over the Amazon Basin
Simulations over the Amazon Basin