Lecture 5 - Water Budget 3: Stream Flow

Report
Water Budget III:
Stream Flow
P = Q + ET + G + ΔS
Why Measure Streamflow?
• Water supply planning
– How much water can we take out (without
harming ecosystems we want to protect)
• Flood protection
– How much water will come down the channel if X
storm happens? Who’ll be flooded?
• Water quality
– What are the fluxes (flow x concentration) of
contaminants to a lake or estuary?
• What are the effects of land use change on
water delivery to downstream systems?
Stream Flow Network
(http://water.usgs.gov)
Extreme Low Flows
Rainfall between Oct 1. 2010 and
January 21, 2011 is 2.68 inches
Hydrograph- graph of flow over time
Flow Over Time – Santa Fe River
Seasonality of Flow
Watersheds as Filters
• Rain falls
• Storages “buffer” the rainfall signal, letting
water out slowly
– More storage = more buffering
• The result is that the rainfall signal looks
stochastic, the flow looks more “organized”
• Watershed properties AND size affect the
filtering effect
Rainfall Filtering – Santa Fe River
Rainfall Filtering – Finer View
Consider the “Filter” Effects of:
•
•
•
•
•
•
•
Watersheds with steep vs. shallow slopes
Watersheds with deep vs. shallow soils
Watersheds with intense vs. extended rainfall
Watersheds with forests vs. parking lots
Watersheds with dams vs. not
Big vs. little watersheds
Watersheds with big shallow aquifers
Basin Filtering Creates Lags –
Hatchet Creek
Where is Stream Flow From?
• At any time, flow is a composite of water with
different sources and residence times
– Some water is stored in the watershed for a very
long time, some very short
– During low flow conditions, water is mostly old
– During storms, the contribution of new water
increases
– How does an aquifer affect this?
2%
8%
23%
Flow and Rainfall Intensity
• If Rainfall Intensity
> Infiltration
Capacity then
surface runoff
occurs
• Stream flows are
composites of
– Surface runoff
– Subsurface flows
Variable Source Area
(Stormflow Generation in Florida)
Variable Source Area Makes
Antecedent Rainfall IMPORTANT
Land Use (Cover) Affects Runoff
Generation
• Impervious surfaces preclude infiltration
• Less infiltration means more runoff
• Runoff also MOVES faster
– Less “filtering”
• Compare land uses…
More stormflow, higher peak flow, sooner.
Forest
Ag
Urban
The importance of storage – the
basis of filtering
Same total flow (area under the
curve), lower peak flow.
Water Storage in the Forest
Wetland Hydrological Services
Depressions and vegetation (swamps) slow runoff.
Upper watershed wetland storage delays runoff
and reduces peak flows.
Wetland flood plain has a dominant influence on
downstream peak flow and solute transport.
Why Does Storage Matter?
3
200,000 m of Stormwater Runoff;
3
Channel Peak Flow capacity of 1m /s
All in one day
3
Peak Flow =2.3 m /s
Spread over 3 days
3
Peak Flow = 0.8m /s
How Big a Flood Can We Expect?
• The size of the flood is inversely proportional
to it’s frequency
–
–
–
–
Big event happen rarely
Big events shape the landscape
Medium events maintain the landscape
Small events control the biology
• How would we predict the size of a flood that
happens roughly once in 25 years?
– Think back to the rainfall lab…
Rainfall Recurrence Series
Flow (Santa Fe River Station 3)
Daily Flow Recurrence Series
The 100-yr Floodzone Map
How Do We Measure Streamflow?
• Funny you should ask…basis of Lab #4.
• Basis is to estimate:
– Cross-sectional area (A; through which water flows)
– Water flow velocity (V)
• Q=A*V
Measuring Surface Flow
Typical
stream
velocity
profile
Where to measure
mean velocity?
Float Velocity * 0.8
for natural channels
Float Velocity * 0.9
for concrete channels
Velocity
Instruments
$2k
Turn-cup
0.500 ft/s
$4k
$7k
Electromagnetic
0.050 ft/s
Sonic Doppler
0.005 ft/s
Sect
Width
(m)
Depth
(m)
1
2
1
1
0.7
2.0
Vel@.6 Flow
m/s
(m3/s)
0.20
0.14
0.25
0.50
3
Total
1
1.3
0.15
0.20
0.84m3 /s
Discharge is HARD to Measure
• We want:
– Daily (or sub-daily)
measurements
– Multiple stations per
river
– Real time updating
(detect changes in
flow as they are
happening)
Rating equations (stage vs. discharge)
allow continuous flow monitoring
Stage-Discharge Relation
• Water stage (elevation) is EASY to measure
• Stage is related to dischage via a mathematical
relationship
• Applying that relationship to measured stage gives
estimates of discharge
H
H
t
Stage Hydrograph
Q
Q
Stage-Discharge Curve
or Rating Curve
t
Discharge Hydrograph
Stage-Discharge Relation
• Typical relationship: Q = a(H +b)c
• The relationship between H & Q has to be
calibrated locally for different stations
• At low stage, positive
relationship between
stage and discharge
• At high stage,
negative relationship
• Why?
Discahrge
Stage Discharge Relationship for
the Ichetucknee River
Stage
Stage Measurements
Float-pulley
Staff gage
Pressure
Ultrasonic
Weirs
Flumes
Weir vs Flume
Type
The Good
The Bad
Weir
Low cost
Easy installation
Won’t work on low gradient
streams
Upstream flooding
Clogs
Changes WQ
Wildlife barrier
Flume Works ok in low gradient
streams
Better for WQ and
wildlife Self cleaning
High cost
Difficult to install
What if there’s no rating curve?
• New watershed, new conditions
• Areas where it’s hard to develop rating curves
– For example, the Everglades
Manning’s Equation - Flow Estimation
without a rating equation
Q= 1/n * A * r 2/3 * s1/2
Q = estimated flow m3/s
n = Manning’s roughness number
(0.02 smooth to 0.15 rough or weedy, 0.5 dense vegetation)
A = cross sectional area (m2)
r = Hydraulic Radius (wetted perimeter = WD/(W + 2D)
W > 10D, R → D)
s = Hydraulic Gradient ΔH/L
Predicting Flow in the Everglades
• Dense vegetation
channel (n = 0.4)
• Shallow slope (s = 3 cm
per km = 0.00003)
• Wide channel (100 m
wide, 0.3 m deep, A = 30
m2, r = 30 m2 / 100.6 m =
0.3 m)
• What is Q? What is flow
velocity (u)?
• Q = (1/n) * A * r0.67 * s0.5
• V=Q/A
• Q = (1/0.4) * 30 m2 * 0.3
m0.67 * 0.000030.5 = 0.183
m3/s
• V = 0.183 m3/s / 30 m2 =
0.006 m/s = 0.6 cm/s
Next Time…
• Groundwater

similar documents