Solid Precipitation Daqing Yang, Barry Goodison, Paul Joe, others ??

1
Solid PrecipitationDaqing Yang, Barry Goodison,
Paul Joe, others ??

• Role of snowfall
• Status of observations gauge network, satellite,
  and radar
• Research examples
• Recommendations

2
1. Role of Solid Precipitation

• Significant portion of yearly precipitation in
  the cold regions (including the polar regions)
  important indicator of climate change and
  variation
• Input to winter snowpack and spring snowmelt
  runoff in mountain and polar regions critical
  element of basin water cycle and regional water
  resources
• Influence on large-scale land surface radiation
  and energy budget particularly during
  accumulation and melt seasons
• Effect on glacier/ice sheet accumulation/mass
  balance, lake/river and sea ice, seasonal
  frozen-ground and permafrost
• Impact to human society and activity, such as
  air/ground transportation, disaster prevention,
  agriculture, water resources management, and
  recreation

3
2. Status of Observations - gauges, satellite and
radar

• Gauge network
• Global coverage with various operational,
  national/regional networks.
• Manual and automatic gauges, measuring water
  equivalent (amount), not snow particle size.
• Manual gauges can measure snowfall (rate) at
  6-hour to daily time intervals, and auto gauges
  can provide hourly (or sub-hourly) snowfall
  (rate).
• Snow rulers are also used for snowfall
  observations at the national/regional networks,
  providing snow depth info, not SWE.
• Snow pillow/snowboard/snow depth sensor record
  snow accumulation changes over time - (in)direct
  info of snowfall.
• Gauge networks/data are long-term and
  fundamental, defining global snowfall/climate
  regimes and changes.

4

• Satellites
• Global coverage with merging data / products from
  IR, MV sensors and satellite radars
• Rain rate info (TRMM), also snowfall rate ???,
  challenge with mixed precip
• Particle size info from radars
• Operational products - GPCP blended version 2
  monthly/global, 1987-present, and others????
• Problems of MV data over land, need systematical
  evaluation particularly over the high latitudes
• Limited validations show GPCP v2 data are not
  better than atmospheric reanalysis precip over
  northern regions (Serreze et al., 2005)
• Statement of importance Key to advance our
  capability of monitoring and observing
  (liquid/solid) precipitation globally???

5
Examples of RS Precip Dadasets

• Operational products - GPCP blended version 2,
  monthly/2.5x2.5 grid, global, 1987-present

6
Summary Table current/planned capabilities and
requirements for space-based remote sensing of
snowfall parameters (adopted from xxx, not done
yet)
C Current Capability L Low end of
measurement range U Unit T Threshold
Requirement (Minimum necessary) H High end of
measurement range V Value O Objective
Requirement (Target)
7
Radar network

• Only cover very limited parts of the globe (much
  less extensive than the gauge network)
• Expensive and can be difficult to operate and
  calibrate
• Mainly designed for severe weather detection,
  with less concern for precipitation, certainly
  NOT for snowfall measurements, (although being
  used to measure snowfall with problems for light
  snowfall)
• Major limitations for operational radars
• lack of low level coverage at moderate (80 km) to
  long range for precipitation and this is even
  shorter for snowfall
• in complex terrain, the radar beam is often
  blocked by mountains and/or the radar is located
  to scan over the top of mountains and not in the
  valleys
• A new innovation is the deployment of a network
  of redundant low cost, low maintenance radars
  (CASA radars) to scan the low levels of the
  atmosphere.
• Statement of importance - key to understand
  cloud/precipitation physics and for validation of
  satellite precipitation data and products.

8
3. Research Examples

• gauge network and data
• RS snow data validation

9
Shortcomings in gauge network

• Sparseness of the precipitation observation
  networks in the cold regions.
• Uneven distribution of measurement sites, i.e.
  biased toward coastal and the low-elevation
  areas, less stations over mountains and oceans.
• Spatial and temporal discontinuities of
  precipitation measurements induced by changes in
  observation methods and by different observation
  techniques used across national borders.
• Biases in gauge measurements, such as
  wind-induced undercatch, wetting and evaporation
  losses, underestimate of trace and low amount of
  precipitation, and blowing snow into the gauges
  at high winds
• Data access is also difficult or costly for some
  regions and countries
• Decline of the networks in the northern
  regions/countries

10
Synoptic/climate stations on land above 45?N and
the Arctic Ocean drifting stations
11
NRCS SNOTEL / Wyoming gauge network
NRCS National Water and Climate Center
www.wcc.nrcs.usda.gov/snotel/Alaska/alaska.html
12
NOAA US CRN
http//www.ncdc.noaa.gov/oa/climate/uscrn/
13
National standard gauges tested in Barrow
Russian Tretyakov
Canadian Nipher
US 8
Hellmann
14
Biases in Gauge Meaurements (mentioned 3 times
in IGOS Water Cycle Report)

• Wind-induced gauge under-catch
• Wetting and evaporation losses
• Underestimate of trace precipitation events
• Blowing snow into gauges at high winds
• Uncertainties in auto gauge systems

15
WMO Solid Precipitation Intercomparison
CRN modified DFIR
Goodison, B.E., P.Y.T. Louie, and D. Yang, 1998
WMO solid precipitation measurement
intercomparison, final report, WMO/TD-No. 872,
WMO, Geneva, 212pp.
WMO double fence intercomparison reference
(DFIR) in Barrow, AK
16
Wind-induce undercatchWMO intercomparison
results
17
Overall mean for the NP drifting stations,
1957-90 (Yang, 1999)
Overall mean for 61 climate stations in Siberia,
1986-92 (Yang and Ohata, 2001)
18
Bias corrections of daily precipitation data,
Barrow, 1982-83 (Yang et al., 1998)
19
(No Transcript)
20
Mean Gauge-Measured (Pm) and Bias-Corrected (Pc)
Precipitation, and Correction Factor (CF) for
January
Yang et al., 2005, GRL
a) Pm (mm)
b) Pc (mm)
c) CF

• Total 4827 stations located north of 45N, with
  data records longer-than 15 years during
  1973-2004.
• Similar Pm and Pc patterns corrections did
  not significantly change the spatial
  distribution.
• CF pattern is different from the Pm and Pc
  patterns, very high CF along the coasts of the
  Arctic Ocean.

21
Mean Gauge-Measured (Pm) and Bias-Corrected (Pc)
Precipitation, and Correction Factor (CF) for
July
Yang et al., 2005, GRL
a) Pm (mm)
b) Pc (mm)
c) CF

• Total 4802 stations with records longer-than 15
  years during 1973-2004.
• Similar Pm and Pc patterns.
• Small CF variation for rainfall over space.
• CF pattern is different from the Pm and Pc
  patterns.

22
Impact of Bias-Corrections on Precip Trend Pm
Pc Trend Comparison, Selected Stations with Data
gt 25 Yrs during 1973-04
Jan.
Jul.
Yang et al., 2005, GRL
23
RS snow data validation
- Comparison with in-situ snow data (scale
issue) - Regional / basin water budget
calculations to assess moisture budget closure
       Basin/region winter snow mass balance
SWE Snowfall Sublimation       
Basin spring water budget Runoff SWE
Precip. Evaporation Storage -
Hydrologic modeling and snow assimilation
24
9
15
Large Arctic rivers their annual discharge to
the Arctic Ocean/marginal seas
17
5
11
25
Snow Water Equivalent (SWE) Information
Streamflow inter- annual variation Basin
extreme (weekly-mean) discharge (m3/s). Data
source UNH/SHI
26
Snow Water Equivalent (SWE) Information

• Lena basin has the highest winter snow pack, and
  Yenisei basin has the lowest.
• 2. The snow pack accumulate to the highest in
  winter, week 4-12.
• 2. For study convenience, when when SWE lt0.5mm,
  the basin is considered empty.

Basin SWE inter-annual variation Extreme (SSM/I)
snowcover water equivalent (SWE, mm),
1988-2000. Data source NSIDC/UNH
27
Basin SWE (mm) vs. weekly discharge (m3/s), Lena
R., 1988-99
28
Basin SWE vs. winter precip (mm), Lena R.,
1988-2001
29
Basin SWE vs. winter precip (mm), Ob R.,
1988-2001
30
4. Recommendations
31
Gauge networks and observations

• Network
• continue conventional point precipitation
  measurements against declining networks in many
  countries
• sustain and enhance the gauge network in the cold
  regions
• develop guidelines on the minimum station density
  required for climate research studies on solid
  precipitation in cold climate regions
• Data
• undertake bias analysis and corrections of
  historical precipitation gauge data at regional
  to global scale
• ensure regular monitoring of the snowfall
  real-time data, quality control and transmission
• examine the impact of automation on precipitation
  measurement and related QA/QC challenges,
  including compatibility between national data,
  and manual vs. auto gauge observations
• develop digitized metadata for regional and
  national networks
• Test facility/new technology
• identify and establish intercomparison sites for
  standardized testing of new technology, such as
  polarization radar, CASA radar networks, hot
  plate, pressure, or blowing snow sensors
• encourage national research agencies to establish
  programs to provide support for the development
  of new instruments to measure solid precipitation
  in high latitude regions
• use of wind shields and direct measurement of
  winds at emerging auto gauge sites/networks

32
Satellites

• Need GPM ASAP and strongly encourage the EGPM
  mission to measure global rain/snowfall data,
  including major parts of the N regions
• Need to blend (combine) data from different
  sources (in-situ, model, satellite)
• Need to systematically evaluate RS snow data /
  products over cold regions via direct
  comparisons, analyses of basin water budget and
  compatibility in basin/region SWE-runoff,
  SWE-snowfall
• Need to maintain reasonable expectations on what
  satellite and radar technologies are able to
  provide
• Need for further intensive field efforts to
  address scaling issues.
• Need for new technology development
• The use of combined active and passive satellite
  data for snowfall detection/retrieval should be
  further encouraged.
• Active space-borne instruments need to have a low
  detectability threshold (better than than 5 dBz)
  to detect light rainfall and snowfall.
  Deployment of rain radars with lower
  detectability threshold is encouraged.
• New passive microwave instruments and new channel
  combinations need to be studied, particularly at
  high frequency.
• The sounding channel technique proposed by the
  EGPM mission should be implemented.
• The new Meteosat Second Generation has many more
  channels than previous geostationary satellites.
  They have been able to provide information on
  particle size and phase. Exploration of these
  additional channels for precipitation estimation
  is encouraged.
• Aircraft sensors together with extended channel
  selection studies provide an excellent testbed
  for future satellite instruments. Dedicated high
  latitude aircraft campaigns for snowfall remote
  sensing are encouraged.

33
Ground Radar

• Need to expand the radar networks to the
  northern/cold regions and to obtain more useful
  radar observations of snowfall.
• The CASA radar concept should be deployed with
  high sensitivity for the detection of snow, low
  level measurements and in complex terrain.
• Need to share data and to create regional and
  global radar data sets
• international radar data quality intercomparisons
  to remove inter-radar biases of precipitation
  estimates.
• Availability of common or open source algorithms
  for generating precipitation estimates are needed
  to understand the biases and errors.
• Need for development and further refinement of
  inexpensive ground-based remote sensing
  instruments for snowfall should be encouraged,
  including vertically pointing micro radars, such
  as (Precipitation Occurrence Sensing System) POSS
  or Micro-Rain-Radar (MRR).
• Encourage use of combined active and passive
  satellite data for snowfall detection/retrieval
• Need to study new passive microwave instruments
  and new channel combinations