Multiple Sensor Precipitation Estimation over Complex Terrain

1
Multiple Sensor Precipitation Estimation over
Complex Terrain

• AGENDA
• I. Paperwork
• A. Committee member signatures
• B. Advisory conference requirements
• II. Bulk of meeting
• A. Importance of rainfall monitoring in the West
• B. Goals of M.S. research
• C. Problems with precipitation estimates in
  complex terrain
• D. Multi-sensor approach
• E. Methodology
• 1. Case study (events)
• 2. G vs. R comparisons
• 3. Application of Methods 1 and 2
• III. Timeline
• IV. Discussion

2
Multiple Sensor Precipitation Estimation over
Complex Terrain

• Importance of Rainfall Monitoring in the West
• Water management in the West is becoming
  increasingly important as heavily populated
  cities demand more water.
• Today, companies involved in dam management, such
  as the Salt River Project (SRP), rely on limited
  rain gauge measurements to predict streamflow
  from a given watershed.
• Peak flows from a basin often result from a
  combination of snowmelt and rainfall. Clearly,
  there is a need to accurately estimate basin-wide
  precipitation in the West to predict the
  resultant streamflow.
• Since hydrologic models of the near future will
  rely on WSR-88D measurements, we need to be able
  to IMPROVE our monitoring capabilities of both
  snow and rainfall in the West.

3
Multiple Sensor Precipitation Estimation over
Complex Terrain

• Goals of M.S. Research
• Discover possible error sources within the PPS in
  current precipitation estimates made by the
  WSR-88D.
• IMPROVE basin-wide precipitation measurements on
  a realistic time scale by
• A) Utilizing rainfall measurements from radar,
  rain gauges, and satellite.
• B) Blending radar data for regions which fall
  underneath the umbrellas of 2 or more radars.
• C) Explore ways to use satellite IR imagery to
  apply a model of reflectivity structure to
  blocked regions.

4
Multiple Sensor Precipitation Estimation over
Complex Terrain

• Study Region showing AZ terrain

5
Multiple Sensor Precipitation Estimation over
Complex Terrain

• Problems with Radar Precipitation Estimates in
  Complex Terrain
• 1. Sampling Issues

• 2. PPS Issues
• Z-R relationship
• Grid transformation
• Gauge correction

6
Multiple Sensor Precipitation Estimation over
Complex Terrain

• Problems with Rain Gauge Measurements in Complex
  Terrain
• Spatial coverage of rain gauge network is often
  limited and fails to capture large precipitation
  gradients which are common in areas of high
  relief.
• Gauges provide data on an hourly basis, but they
  report at varying times after the hour.
• Most sensors often are not equipped to handle
  mixed-phase precipitation events (i.e. tipping
  bucket gauges).
• Splash-out, underestimating the catch, wind
  effects, debris blockages, just to name a few.

7
Multiple Sensor Precipitation Estimation over
Complex Terrain

• Problems with Satellite Precipitation Estimates
• Derived rainfall accumulations are too coarse for
  basin-wide precipitation estimations.
• Estimates are subject to the cirrus problem.
  The vertical depth and cloud top temperature of
  certain cloud features are not always
  well-correlated with rainfall reaching the
  surface.
• Some precipitating systems produce rain via warm
  rain processes.
• In the West, the lower troposphere can be quite
  dry and result in significant sub-cloud
  evaporation.

8
Multiple Sensor Precipitation Estimation over
Complex Terrain

• Dual Radar Multi-Sensor Approach
• Many precipitating systems in the West during the
  winter can be of a stratiform type. Perhaps it
  is best to sample these clouds at the lowest
  available tilt from either radar (optimal height
  sampling).
• Many regions which are blocked from one radars
  perspective are quite visible from another radar.
  
• By utilizing data from 2 radars, we can obtain
  precipitation estimates which were sampled at
  differing heights.
• The dual radar approach allows radar data to be
  blended from adjacent radars.

9
Multiple Sensor Precipitation Estimation over
Complex Terrain

• Single Radar and Satellite Multi-Sensor
  Approach
• Due to the complexity of the terrain and
  associated radar blockages in AZ, precipitation
  estimates for many regions are sampled at a
  height near or above the tropopause!!!
• This problem worsens in the case of shallow,
  precipitating clouds which are not uncommon in
  the winter.
• The multi-sensor approach attempts to improve
  precipitation estimates in regions which are
  inadequately sampled by the WSR-88D network.
• In essence, a model will be applied which
  correlates reflectivity structures (profiles) to
  cloud top temperatures.
• This relationship can then interpolated to areas
  which are shadowed by mountains.

10
Multiple Sensor Precipitation Estimation over
Complex Terrain

• Single Radar and Satellite Multi-Sensor
  Approach
• Obtain the vertical reflectivity structure of a
  precipitating system in a region which is
  well-sampled by the WSR-88D (every 15 minutes).
• Similarly, determine the associated average cloud
  top temperature of the system by GOES 9 IR
  imagery.
• Next, obtain the average cloud top temperature of
  cloud features in the problem area.
• With this, derive a vertical reflectivity profile
  for this region. This model will need to be
  adjusted in the case of contrasting average cloud
  top temperatures.
• Utilizing optimal height sampling, retrieve the
  reflectivity value in the problem area and
  convert to a rainfall rate (standard Z-R).
• Update the REF profile-IR relationship every 15
  mins.

11
Multiple Sensor Precipitation Estimation over
Complex Terrain

• Multi-Sensor Approach (Single Radar and
  Satellite)

12
Multiple Sensor Precipitation Estimation over
Complex Terrain

• Methodology
• Collect archive level II radar data, GOES 9 IR
  imagery and rain gauge accumulations for 2-3
  cases of widespread precipitation affecting a
  large area in central AZ.
• For each event, objectively determine the
  performance of both radars using rain gauge data
  as ground truth.
• Next, identify the problem areas, and attempt
  to ascertain why the radar-derived precipitation
  estimates are in error.
• Apply the multi-sensor methods and determine the
  magnitude of the improvements made upon the
  basin-wide precipitation estimates.

13
Multiple Sensor Precipitation Estimation over
Complex Terrain

• Methodology - Case Study (Events)
• How were cases chosen?
• Major watershed event in the Salt and Verde River
  Basins - determined by average precipitation
  amounts.
• Thus far, we have identified 9 cases which fit
  the above criteria.
• These cases are mixed-phase, widespread
  precipitation events which occur over 1-2 days.
• Will narrow these cases down to 2-3 case events
  for detailed study.

14
Multiple Sensor Precipitation Estimation over
Complex Terrain

• Methodology - G vs. R comparisons
• Using GIS software, I am now able to objectively
  analyze the radar-derived precipitation data
  versus the rain gauge accumulations.
• Next, I will segregate the G-R statistics by the
  elevation of the rain gauge and by the height
  from which the reflectivity value was obtained
  (sampling height).
• Where are the problem areas and why are they
  there?
• Is the error a function of gauge location w.r.t.
  the radar?

15
Multiple Sensor Precipitation Estimation over
Complex Terrain

• Methodology - Application of Methods 1 and 2

16
Multiple Sensor Precipitation Estimation over
Complex Terrain

• Timeline
• April 8 - Have all data collected and formatted
• April 17 - Identify all problem areas based on
  G-R comparisons
• May 1 - Begin implementing methods to improve
  basin-wide precipitation estimates
• May 15 - Determine how well the methods worked
• June 1 - Begin writing thesis
• August 3 - Goal Draft of thesis
• August 28 - Finalize revisions
• September 8 - Give thesis seminar
• September 10 - Defend thesis