BEGIN

1
BEGIN

• Precipitation as the Input

2
Some Huge Rainfalls
3
Precipitation As Input

• Precipitation is generally pre-processed
• Uniform in space and time never!
• Gages - Recording non-recording
• Radar
• Satellite Derived
• QPF

4
The Basic Process.
Necessary for a single basin
Focus on Precipitation
Excess Precip. Model
Excess Precip.
Basin Routing UHG Methods
Runoff Hydrograph
Excess Precip.
Stream and/or Reservoir Routing
Downstream Hydrograph
Runoff Hydrograph
5
From A Basin View
Excess Precip.
Excess precip. is uniformly distributed!
Excess Precip. Model
Basin Routing Unit Hydrograph
Runoff Hydrograph
Stream Routing
6
Precipitation

• ... primary "input" for the hydrologic cycle (or
  hydrologic budget).
• The patterns of the precipitation are affected
  by large scale global patterns, mesoscale
  patterns, "regional" patterns, and
  micro-climates.
• In addition to the quantity of precipitation,
  the spatial and temporal distributions of the
  precipitation have considerable effects on the
  hydrologic response.

7
Precipitation

• In lumped models, the precipitation is input in
  the form of average values over the basin. These
  average values are often referred to as mean
  aerial precipitation (MAP) values.
• MAP's are estimated either from 1)
  precipitation gage data or 2) NEXRAD
  precipitation fields (MAPX).

8
Precipitation (cont.)

• If precipitation gage data is used, then the
  MAP's are usually calculated by a weighting
  scheme.
• a gage (or set of gages) has influence over an
  area and the amount of rain having been recorded
  at a particular gage (or set of gages) is
  assigned to an area.
• Thiessen, isohyetal, and the inverse-distance
  squared are some of the more popular methods.

9
Precipitation Issues for the Hydrologist

• Characteristics of precipitation in or on my
  basin(s)!
• Quantity How much are we getting?
• Space Where will it fall?
• Time When will it fall (and where)?
• Integrity of the Data Is this data valid?

10
Characteristics

• Convective, Frontal, Orographic, etc

11
Convectional Storms....

• Thunderstorms are the classic example.
• Warm moist air is rapidly lifted - making it
  unstable.
• As the air lifts it cools and precipitation
  forms.
• As the precipitation falls - it cools the air
• This is why you may feel very cool bursts of air
  during those hot summer days when a thunderstorm
  kicks up.

12
Urban Areas Thunderstorms...

• It has been reported that urban areas may
  contribute to the development of thunderstorms
  due to the presence of a heat source and the
  typically darker areas.

13
Orographic Effects.....

• Terrain can also cause lifting - which is a major
  component in the precipitation mechanism.
• The mountains provide a lifting mechanism for the
  warm advecting moist air.

14
Orographic effects
15
Local Effects e.g. the Great Lakes...
Do lake effect events alter the volume of Lake
Superior?
16
Ice....

• Hail, Rime, Sleet, and Graupel
• Very difficult to measure
• Antifreeze or heated gages

17
Snow, A Few Brief Points .....

• Snow or snowfall reaches the ground to form the
  snowpack. Snowpack is generally reported as snow
  depth.
• We must also consider the snow water equivalent
  or SWE - WHY?

NOAA Photo Library
18
SWE....

• SWE is reported as a ratio - i.e. 101
• Meaning 10 inches of snow equal 1 inch of water -
  when melted.
• We also report this as density.
• 101 would be a density of 10 or 0.1.
• When is the snowfall most dense and least dense.
• When is the pack most or least dense?

NOAA Photo Library
19
Measuring Snow and SWE...

• Snow gages
• Snow tubes
• Radar - VERY difficult!! - WHY?????

20
Quantity

• Measuring the Precipitation

21
Rainfall.....

• Rainfall varies in both space and time
• This is referred to as spatial and temporal
  variability.
• Rainfall amounts vary considerably

22
Measuring Precipitation....

• Generally use rain gages
• Measure depth
• What are the problems with rain gages?
• Point coverage...
• Interference - wind, trees, etc...
• How many others can you name?
• Radar

23
Standard Gage(non-recording)
24
Fisher Porter Tipping Bucket
25
Universal
26
Precipitation Gage Networks

• A system of gages
• Design Issues
• density
• location
• quality (of data)
• collection transmission
• processing, filing, managing

27
Factors Affecting Density

• Purpose of Network Desired Quality/Precision/Acc
  uracy
• Finances Installation and UPKEEP!
• Nature of Precipitation rain, rain snow,
  orographic, convective, etc..
• Accessibility
• to name a few.....

28
Network Densities

• Many studies
• Brakensiek et al., 1979 Brakensiek, D. L., H.
  B. Osborn, and W. J. Rawls, cooridnators. 1979.
  Field Manual for research in Agricultural
  Hydrology. USDA, Agricultural Handbook, 224, 550
  pp, illustrated.

29
Spatial Characteristics

• Where will it fall and
• how will I use it?

30
Precipitation in Models

• In lumped models, the precipitation is input in
  the form of average values over the basin. These
  average values are often referred to as mean
  aerial precipitation (MAP) values.
• MAP's are estimated either from
• 1) precipitation gage data or
• 2) NEXRAD precipitation fields (MAPX).

31
Precipitation (cont.)

• The MAP's are usually calculated by a weighting
  scheme.
• a gage (or set of gages) has influence over an
  area and the amount of rain having been recorded
  at a particular gage (or set of gages) is
  assigned to an area.
• Thiessen, isohyetal, and the inverse-distance
  squared are some of the more popular methods.

32
Calculating Areal Averages....

• Arithmetic
• Isohyetal
• Theissen
• Inverse Distance

33
Arithmetic....
34
Thiessen

• Thiessen method is a method for areally weighting
  rainfall through graphical means.

35
Isohyetal

• Isohyetal method is a method for areally
  weighting rainfall using contours of equal
  rainfall (isohyets).

36
Inverse-Distance Squared
Used to compute average precipitation at any
point based on nearby gages. The weight of the
nearby gages is dependant on the distance from
the point to each of the nearby gages.
Gage A
Gage B
dA
dB
dC
Gage C
37
Radar Precip. as Input

• Radar gives a good picture of where it is raining
  - may indicate how to adjust the Unit Hydrograph
  for moving and partial area storms!
• May also give good estimate of how much, BUT
• Will differ from gages in total basin average.
• Historical records are based on gages!
• This makes calibration rather difficult.

38
WSR-88D

• Weather Surveillance Radars - 1988 Doppler
• 1st WSR-88D sites installed in 1991
• At the present time, there are more than 160
  radars in place.
• Should optimally provide coverage for a large
  percentage of the United States.
• Optimally used because under many circumstances,
  the useful range of the radars varies
  considerably.

39
Locations
40
NEXRAD

• Nexrad is a method of areally weighting rainfall
  using satellite imaging of the intensity of the
  rain during a storm.

41
Temporal

• When will fall and where?

42
Temporal Distributions

• Gages record data at intervals - 10 min., 15
  min., 1 hour, 24 hour, etc....
• Models use the data at 1-hour, 6-hour, etc...
• Must either aggregate or disaggregate
  precipitation amounts....
• i.e. Combine 1 hour values into a 6-hour value...
  Not a problem!
• Or... Break a 24-hour value into 6 hour values...
  Much more difficult!

43
Temporal Disaggregation
24-hour gage 3.6 inches total
1 hour gage with 2.2 total inches and the
following distribution
Distribute the 3.6 inches using the breakdown of
the hourly gage
44
Intensity, Duration, Frequency

• Intensity, duration, frequency
• Duration - the length of time over which the rain
  falls.
• Intensity - the rate at which the rain falls or
  the amount / duration.
• Frequency - the frequency of occurrence - i.e.
  How rare is this storm? - Well get back to
  this.....
• General relationships
• the greater the duration, the greater the amount
• the greater the duration, the lower the intensity
• the more frequent the storm, the the shorter the
  duration, and
• the more frequent the storm, the less the
  intensity

45
Lets Look at at an Example

• First
• Lets compute the Rainfall/Runoff ratios for the
  Little J at Spruce Creek.

46
The Situation.
47
1996 Totals
48
Some Issues

• How to handle the missing data
• Which basin averaging technique to use.
• Gage Average
• Thiessen
• Isohyetal
• Inverse Distance Weighting

49
Missing Data

• Filling in missing data is a major issue.
• In this case, we are filling it in space not
  time.
• There are many ways to fill in this data
• Averaging nearby stations
• Weighting (averaging is a special case)
• Isohyetal

50
The Missing Data

• Averaging 57.06 inches
• Weighting would depend on local knowledge and
  would require creation of historical
  relationships between all of the local gages.
• Isohyetal would imply that the value is closer to
  62 to 63 inches see next slide
• For this exercise we will use 60 inches.

51
Isohyetal
52
Now Lets Find Basin Average

• Arithmetic Averaging
• Thiessen
• Isohyetal
• IDW

53
Gage Average
I used Excel to average the gages. The small
worksheet is shown at the right -
54
Thiessen Polygons
55
Thiessen Wts. ()
56
Combine w/ Totals
Replace w/ 60.0
57
Thiessen - Final Computations
58
Isohyetal Approach
59
Isohyetal Areas
60
Combine and Precip. Values
61
Inverse Distance Weighting

• Need coordinates of each gage
• Need coordinates of basin centroid or point of
  interest.
• Then Calculate gage weights

62
Measure 4 Distances
63
The Computations
64
In Summary
65
What if this had been a 6-hour storm instead of
yearly totals?

• What would we do?

66
Use Thiessen Weights
Just average each incremental contribution using
the pre-calculated Thiessen weights!
67
Area-Depth (amount) Relationship....
Indeed we should get less basin average
precipitation than for a single gage
68
Use this Chart
A gage in the middle of a 200 square mile basin
records 5 inches of rain in 3 hours. Estimate
the basin average rainfall For 200 square miles,
the basin average is 80 of the gage total or
0.8 5 4 inches!
69
Temporal Distributions

• Understanding Temporal Distributions is very
  important,as this greatly affects runoff timing
  and volumes.

70
Temporal Distributions

• Precipitation is a continuous process.
• Intensities vary depending on amount and duration
• Gages record data at intervals - 10 min., 15
  min., 1 hour, 24 hour, etc....
• Models may use the data at 1-hour, 6-hour, etc...
• Must either aggregate or disaggregate
  precipitation amounts....
• i.e. Combine 1 hour values into a 6-hour value...
  Not a problem! Or... Break a 24-hour value into
  6 hour values... Much more difficult!

71
Understanding Intensities
72
Intensities Durations

• A 5-minute recording gage
• Recorded a storm for 40 minutes
• Calculate
• Total Rainfall
• Cumulative Rainfall Curve
• Max. 5,10, 30 minute intensities
• The average intensity

73
The Data
74
Solutions

• Total rainfall simply sum the precipitation
  values 56.16 mm or 2.21 inches

• Cumulative data is shown and plotted below

75
Solutions, cont.

• The maximum 5 minute intensity was 15.6 mm
  between 10-15 minutes at 187.2 mm/hr or 7.3
  inches/hr. This is illustrated in the data below

76
Solutions, cont

• The maximum 10 minute intensity was found by
  aggregating sequential 5-minute periods. The
  maximum 10-minute intensity is illustrated below,
  between 10-20 minutes with 22.56 mm or 135.36
  mm/hr or 5.29 inches/hr.

77
Solutions, cont

• The maximum 30 minute intensity was found by
  aggregating sequential 5-minute periods. The
  maximum 30-minute intensity is illustrated below,
  between 5-35 minutes with 52.8 mm or 105.6 mm/hr
  or 4.125 inches/hr.

78
Solutions, cont
The total rainfall was 56.16 mm over a duration
of 40 minutes for an average intensity of 84.24
mm/hr or 3.29 inches/hr. In summary
79
Temporal Aggregation
Simply aggregate values to desired periods.
80
The Previous 40-minute Storm

• Recombine into 10, 20, and 40 minute hyetographs.
• What are the issues here?

81
The Graphs
82
Temporal Disaggregation
Basin gage records 66.2 mm total
5-minute gage with 56.16 mm total precip. and the
following distribution
Distribute the 66.2 mm using the breakdown of the
5 minute gage
83
The Solution
We made a very large assumption about the 66.2 mm
total duration what was it ?
84
END

• Precipitation as the Input