MEASUREMENT OF SOLID PRECIPITATION
AT AUTOMATIC WEATHER STATIONS
Challenges and opportunities
R Nitu(1), K Wong(2)
(1)
Meteorological Service of Canada, 4905 Dufferin St, Toronto, ON, M3H 5T4, Canada, tel. 1-416-739-4133, fax
1-416-739-5721, e-mail Rodica.nitu@ec.gc.ca,
(2)
Meteorological Service of Canada, 4905 Dufferin St, Toronto, ON, M3H 5T4, Canada, tel. 1-416-739-4139, fax
1-416-739-5721, e-mail Kai.Wong@ec.gc.ca
ABSTRACT
The use of a wide range of instruments and configuration has major consequences for the
measurement accuracy and consistency of local and global precipitation time series. Because of the
non-consistent data sets, it is difficult to compile local and large-scale climatology of solid
precipitation, develop accurate water budgets, or effectively validate satellite observations.
The fourteenth session of the WMO Commission for Instruments and Methods of
Observation (CIMO-XIV) recognized the critical role played by the instrument experts in
providing objective understanding of the quality of the precipitation data. For that reason it has
established the assessment of methods for measurement and observation of solid precipitation,
snowfall and snow depth, at automatic, unattended stations in cold climates (i.e. polar, alpine), as a
priority for the Expert Team on Surface-Based Instrument Intercomparisons and Calibration
Methods (ET-SBII&CM). In the first phase of this initiative, a survey was conducted in 2008 to
develop the summary of current methods and instruments for measuring solid precipitation.
The results of the survey indicate that a large variety of automatic precipitation gauges and
configurations are currently used for the measurement of precipitation amount and solid
precipitation, worldwide, including in the same country. Also a wide range of parameters are
reported, measured or derived, which could create difficulties for the users accessing the
precipitation databases.
Given the information available, it is recommended that WMO leads the organization of a
field intercomparison of automatic precipitation gauges and their configurations, in various climate
conditions, building on the significant efforts currently underway in many Member countries. The
intercomparison would evaluate the automatic gauges in their operational configuration and
provide guidance on the consistency of precipitation data. A second key objective would be the
development of comprehensive datasets to support studies of the homogeneity of records of
precipitation with special consideration given to solid precipitation.
INTRODUCTION
Reliable precipitation measurements are critical for the development of regional and global
precipitation data sets, and support scientific studies and engineering applications. Large
uncertainties in precipitation estimates, in particular over the cold climate regions have been
demonstrated [D Yang et al, 2005], which complicate the verification of model simulations of
hydrological processes, including our understanding of water balances.
Over the past decade, the transition from manual to automatic observation of precipitation
has introduced new types of challenges with respect to the quality, consistency, compatibility, and
representativeness of in-situ hydro-meteorological measurements. Although simple to be observed
by humans, the measurement of solid precipitation has remained an area where the automatic
gauges have not performed as expected. In addition to functional limitations of gauges, it has been
widely demonstrated [ e.g. Nespor et all, 1998] that any automatic-based measurement of solid
precipitation is subject to systematic errors caused by wind-induced undercatch, as well as wetting
1
and evaporation losses. The metadata describing the circumstances of measurements and the
instrumentation used, are important for the users of data. While the consistency of precipitation
data would be achievable more easily if the same or compatible gauges and siting criteria were
used throughout, the technology development of the hydro-meteorological equipment has
introduced an even higher variability of instruments used for this purpose.
Between 1987 and 1994, WMO organized a Solid Precipitation Measurement
Intercomparison (Goodison et al, 1998), which assessed the national measurement methods for
solid precipitation at the time, when most of them were based on manual observations. Since then,
automatic stations have gained an increased share in providing precipitation data. In some
countries (e.g. Canada, Germany, USA), there are attempts to derive snowfall observations from
automatic measurements, as an alternative for the significant decrease in the availability of manual
observations.
Recognizing the critical role played by the instrument experts in providing objective
understanding of the quality of the precipitation data, the fourteenth session of the WMO
Commission for Instruments and Methods of Observation (CIMO-XIV) has established the
assessment of methods for measurement and observation of solid precipitation, as a priority for
the Expert Team on Surface-Based Instrument Intercomparisons and Calibration Methods (ETSBII&CM).
The first phase of the CIMO-XIV initiative has focused on developing an objective
understanding of the variability of instruments and the circumstances of measurement of
precipitation currently in use world wide, the primary focus being the measurement of solid
precipitation. For this purpose, a survey was distributed in 2008, through CIMO. The survey
invited all WMO Member countries to share information on their current practices for measuring
precipitation, solid precipitation in particular, and their current development work towards
improving the accuracy of measurement. The survey was focused primarily on the assessment of
the variability of measurements methods and the challenges related to ensuring the expected
quality of data, in correlation with the recommendations of the 1989-1994 WMO Intercomparison
on solid precipitation (Goodison et al. 1998), and less on completing an accurate inventory of the
number of stations providing specific parameters.
The results of the 2008 CIMO survey build on the information gathered through two
previous surveys conducted by CIMO, in late eighties and late nineties. These had estimated
approximately 200,000 standards gauges being in use, worldwide, at the time (Sevruk and Klemm,
1989), with about one fifth being recording precipitation gauges (Sevruk, 2002). The extrapolation
of the current results indicates a similar number of stations measuring and reporting precipitation,
however the methods and instruments used have evolved significantly, leading to a much greater
variability of the circumstances of measurement.
SURVEY RESULTS AND RELATED CONSIDERATIONS
The WMO Member countries were asked to provide information on the extent of using
automation for measuring precipitation and the parameters monitored, on the instruments used and
their metadata, on the development and application of adjustments to measurements, and on
current development works taking place for improving the accuracy of measurement of
precipitation.
The NMHSs of 54 WMO Members, covering about 46% of the global land mass, at all
latitudes, except Antarctica, responded. Thirty five (35) of the countries, covering about 28% of
the global landmass, indicated that they monitor solid precipitation (by manual and automatic
means). They reported to operate a total of 41673 sites measuring precipitation; of those17561
sites, or 42%, measure solid precipitation.
2
The results of this survey are conservative; a known limitation is the fact that in many
countries the measurement of precipitation is configured and managed through several independent
agencies in addition to the NMHSs.
The survey indicates that total precipitation amount (solid and liquid) is measured and
reported from all 41,673 stations indicated by the participating Members. Of the total, 18% of the
sites have automatic instruments, with the balance of 82% using manual methods. The next most
widely reported parameter is the depth of snow on the ground. The use of automatic gauges for
measuring this parameter is limited to less than 7% of the sites reporting it. The other parameters
reported are the type of precipitation, the snowfall amount, snowfall water equivalent, snow
temperature/snow surface temperature, snow on the ground water equivalent.
Measurement of Total Precipitation Amount (Solid and Liquid)
There are essentially four types of automatic instruments used for measuring total
precipitation amount, tipping bucker rain gauges (TBRG), 82.9%; weighing gauges (WG), 16.2%;
optical sensors, 0.4%; and “level” gauges, 0.5%.
A remarkable result of the survey is the extent of use and the variety of TBRGs for
measuring precipitation amount, including in countries where solid precipitation is a frequent
occurrence. Collectively, the Members responding to the 2008 CIMO survey use 28 different
models of TBRGs, produced by 22 manufacturers, worldwide. The TBRG measures the amount of
liquid precipitation by recording the numbers of tips corresponding to a fixed amount of
precipitation, equal to the size of each of its two buckets, which gives the gauge sensitivity. The
TBRGs are used in a heated or non-heated configuration.
There is a significant variability in terms of functional and constructive parameters of the TBRGs
in use. Their sensitivity ranges from 0.1 mm of precipitation to 0.5 mm of precipitation, the
collecting area varies from 200 cm2 to 1000 cm2, the amount of heating applied to melt solid
precipitation varies from a few watts to several hundred of watts, applied to various components of
the gauge. In addition to that, the TBRGs in use have various constructive characteristics, e.g.
cylinder or cone shaped, with various diameters.
The WG weighs the precipitation collected in a bucket, and derives the amounts by
integrating the weight over time. Three different principles of measurement are implemented; these
are the vibrating wire load, the single point electronic load, the strain gauge. The WGs currently
operational are from six manufacturers. The collecting capacity of the WGs varies from 240 mm to
1000 mm; newer gauges are available with capacities above 1000 mm of precipitation. The WG
collecting area varies between 200 cm2 and 1000 cm2. Heating of the WGs is a feature increasingly
used to deal with ice buildup and snow capping, and the amount of heating power applied varies
from gauge to gauge.
In some cases, attempts have been made to derive the precipitation amount using optical
non-catching sensors. An optical sensor uses the scattering or obscuration by hydrometeors, in
some cases together with information gathered by other sensors such as rain sensor and
temperature, to determine the type of precipitation and to estimate its intensity and accumulation.
A “level” gauge captures and stores the precipitation in a buffer tank, and measures any
increase of amount using a floater. This device was developed and used by the Royal Netherlands
Meteorological Institute (KNMI).
Use of shields on precipitation gauges
The participating Members indicated that, overall, only 28% of their automatic precipitation
instruments are configured with wind shields. Of the automatic gauges that have wind shields, the
WGs are used in a much larger proportion in this configuration. About 82% of all WGs, or 10% of
the total of automatic instruments, are configured using single wind shields. The shields used are
3
type Alter, Nipher or Tretyakov. The National Weather Service (NWS) of the National
Oceanographic and Atmospheric Administration (NOAA) of the USA is the only Service in
process of adopting double shields, planning to install an Alter shield in addition to the Tretyakov
shield, in use. The WGs installed as part of the network operated by the National Climate Data
Centre (NCDC) of the NOAA use a double shield derived from the Double Fence Intercomparison
Reference (DFIR), recommended at the end of the 1989-1994 WMO Precipitation
Intercomparison, but with a smaller footprint.
The TBRGs are equipped with windshields in a much smaller proportion; only about a third
of them, or 18% of the total automatic instruments. It’s worth noting that over 90% of these are
operated by two services, the Japan Meteorological Administration (JMA), and NWS of NOAA,
USA. In some countries, only some of the TBRGs are protected with wind shields. These are
Australia (only the TBRG at alpine sites are equipped with shields), Norway, and Switzerland.
An assessment of automatic precipitation gauges and their configurations
In 2008, Meteorological Service of Canada (MSC) initiated a qualitative assessment of
current precipitation gauges measuring precipitation amount –liquid and solid -, covering a large
spectrum of measuring principles and constructive parameters, most of which are being used
operationally by other Member countries. The gauges were installed in various configurations,
with and without windshields, and, in some cases, using shields provided by the gauge
manufacturers.
The test was conducted at the Centre for Atmospheric Research and Experiments (CARE) of
Environment Canada, in Egbert, Ontario, about 40 Km south of the Georgian Bay, the eastern side
of the Great Lakes. The area is known for intense lake effect winter precipitation, mainly snow;
however, rain precipitations are possible during the winter.
The instruments evaluated and their configuration is given in Table 1.
Instrument
Geonor T200 (H2) in DFIR:
used as field reference
Belfort Fisher&Porter (FP)
Vaisala VRG 101 (HF)
Geonor T200 (H1)
Geonor T200 (H3)
OTT Pluvio 1, 1000 mm (HJ)
OTT Pluvio 1, 1000 mm (HK)
OTT Pluvio 2, 750 mm (HL)
Vaisala PWD22 (HI)
Vaisala 2G 13H (GE)
All Weather 6021-B (GC)
All Weather 6021-B (GI)
Type
WG
Heating
(Y/N)
No
Shield
(Y/N)
Yes
WG
WG
WG
WG
WG
WG
WG
Distrometer
TBRG
TBRG
TBRG
No
Yes
No
No
Yes
Yes
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
If Yes, Type
DFIR
Alter Shield
Double Alter Shield
Alter Shield
Alter Shield
Tretyakov Shield
Tretyakov Shield
CAE PMB22 (GG)
TBRG
Yes
No
Hydrological Services TB3 (GB) TBRG
Yes
No
Hydrological Services TB3 (GH) TBRG
Yes
No
Table 1: Precipitation Gauges and their configurations, CARE experiment 2008/2009
The cumulative precipitation amounts recorded by each gauge during the test period,
December 2, 2008 to April 15, 2009, derived from the one minute measurement data of the gauges
4
under test, is plotted in the graph below. The reference for this test was a Geonor T200 gauge
installed in a Double Fence Intercomparison Reference (DFIR). The DFIR used is that
recommended following the WMO Intercomparison of 1989-1994 (Goodison et al, 1998), however
the manual observation was replaced with a Geonor T200 gauge currently in use in the Reference
Climate Network of MSC. Manual observations were not available for this test.
Figure 1: Cumulative precipitation amounts: CARE experiment 2008/2009
The experiment results have shown that there is a distinctive difference in performance
between WGs and the TBRGs; at the same time, all WGs, and all TBRGs, respectively, have
displayed comparable performances. The total accumulation for the duration of the test, for the
Pluvio 2, the only WG without a windshield was the lowest of all WGs, however higher than all
TBRGs. The test results indicated that the overall precipitation accumulation of the optical noncatching sensor was better than all the heated TBRGs, but below all WGs.
It has been noted that the measuring differences between gauges with different principles of
operation vary with the type of precipitation. For liquid precipitation all gauges under test were in
better agreement than for reporting solid precipitation. The following two plots provide examples
to illustrate this point. During the rain event of Feb 11-12, 2009, all gauges were in relative
agreement in reporting precipitation accumulation over a 24-hour period. During the snow event of
January 28-29, 2009, the precipitation amount reported from the same gauges during a 24-hour
period, varied significantly.
5
24-hour Total Precipitation Accumulation: Feb 11-12, 2009
Ge Ge
total accumulation: 24 hours
on on
or or
T2 T2
00 00
Ge _D (H
F
1
Va ono IR )
isa r T (H
la 20 2)
AW 2G 0 (H
I 6 13H 3)
AW 021 (G
- B E)
CA I 6 0 ( G
2
E 1- C)
PM B
B2 ( GI
2 )
TB (G G
)
3
Be T ( GB
B
lfo 3 )
(
rt
F GH
VR &P )
G (F
P
Pl 101 )
uv ( H
i
Pl
uv Plu o 1 F)
io vi (H
Va 2, 7 o 1 J)
isa 50 (H K
la mm )
PW
D2 (HL
2( )
HI
)
25
20
15
10
5
0
Gauge
Figure 2: 24-hour comparative rain accumulation
eo eon
o
no
r T r T2
0
20
0_ 0 (H
D
1
G
eo F IR )
no
r T ( H2
V
ai
)
20
sa
0
la
(
H
2
A G 1 3)
W
3
I6 H
0 2 (G
E)
1A
W
B
I6
(G
CA 0 21 C)
-B
E
PM (G
B2 I)
2
(G
T B G)
3
(G
TB B)
Be
3
lfo
(G
rt
F& H)
P
V
RG (FP
10 )
Pl 1 (H
uv
io F)
Pl
1
P
uv
(H
lu
io
vi
J)
o
2,
1
75
(
H
V
0
K
ai
)
sa mm
la
(
H
PW
L)
D
22
(H
I)
12
10
8
6
4
2
0
G
G
24-hour accumulation (mm)
24-hour total precipitation accum ulation Jan 28-29, 2009
Gauge
Figure 3: 24-hour comparative snow accumulation
Qualitatively, the results of this test have shown that TBRGs are significantly
underestimating the amount of solid precipitation The same result was shown by Zweifel, A, and
Sevruk, B.
Adjustment to Measurements
In the WMO/TD - No. 872, IOM 67 report, B Goodison et al, indicated that methods to
adjust solid precipitation measurements for systematic errors should be tested and implemented on
current and archived precipitation data for use by Members, and recommended that wind speed be
monitored at the height of the gauge. As indicated in CIMO Guide #8, the gauge snowfall catch
6
efficiency depends on the wind speed at gauge height, the gauge profile, the effectiveness of the
wind shield (if any), and the snow conditions.
Among the 54 Members participating in the CIMO survey, 13 countries or 24%, indicated
that they adjust the precipitation measurements for some known errors. The adjustments applied by
the participating NMHSs are for Total Precipitation Amount (18.5% of respondents), type of
precipitation (11%), depth of snow on the ground (9.3%), snowfall amount (7%), snowfall water
equivalent (5.6%).
On the topic of the measurement of wind and temperature at sites where solid precipitation
parameters are observed, the responses indicated that wind is measured at gauge location by 44%
of the participating NMHSs, while temperature is measured by 48%; but only 3.7% measure wind
at the height of the gauge (e.g. 2 m, in Canada and USA).
The role and importance of adjustments to precipitation data sets, to account for known
systematic errors should never be underestimated. Work conducted by Climate Research Division
of Environment Canada has allowed for a better understanding of the catch efficiency of the
Geonor T-200B gauges in use in the MSC’s monitoring networks. A study conducted on the
Bratt’s Lake, Saskatchewan test site, based on 24-hour accumulations, has allowed the
development of comparative catch efficiency curves for the Geonor T-200 WG in various shield
configurations (C Smith, 2009), as a function of wind speed.
CE vs. Wind Speed Relationships
1.4
DFIR (Yang et al)
Geonor-DF
Geonor-SA
Geonor-DA*
1.2
1
CE
0.8
0.6
0.4
0.2
0
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
Figure 4: Geonor T200 catch
efficiency versus wind speed.
(Geonor-DF: Geonor in an DFIR,
Geonor-DA: Geonor in an Double Alter
shield;
Geonor-SA: Geonor in a Single Alter
Shield);
courtesy Craig Smith-Environment
Canada
Wind Speed (m/s)
The results demonstrate that the Geonor in a Double Alter shield (DA) catches more snow
precipitation than the Geonor in a single Alter (SA); the DA is more effective as the wind speed
increases. Also, it was shown that the Double Alter is not as effective as the large double fence, i.e.
WMO DFIR or Geonor-DF.
Based on the tests conducted, the following catch efficiency adjustments were proposed for
the Geonor T200 gauge in a single Alter shield:
CEGeonor-SA = e-0.22Ws
CEGeonor-SA = e-0.20Ws
r 2 = 0.61
1.40
r 2 = 0.36
Figure 1: Geonor SA catch efficiency
curve
(red: Bratt’s Lake, SK, Nov-2003
through March-2006;
blue: Jokioinen, Finland, Dec-1998
through Feb-1993) courtesy Craig
Smith, Environment Canada
CE (Geonor/DFIR)
1.20
1.00
0.80
0.60
0.40
0.20
0.00
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
Wind Speed (m/s)
7
9.0
10.0
Depth of Snow on the Ground and Snowfall Amount (Depth of Fresh Snow)
The CIMO survey results have indicated that there are a total of 823 automatic sites in 11
countries reporting the depth of snow on the ground, and a total of 689 automatic sites in 7
countries reporting snowfall amount, which is obtained mostly on the basis of snow depth
measurements using automatic instruments. The majority of the instruments are ultrasonic depth
sensors, also known as sonic ranging depth sensors. In Slovenia, this parameter is obtained using
forward scattering sensors. Newer instruments are becoming available, operating on the principle
of phase variation of visible laser when it bounces off the snow surface. All the countries except
Canada have one instrument per site, and Canada has one or three instruments per site.
The sonic ranging sensors measure the elapsed time between emission and return of an
ultrasonic pulse sent vertically down to the snow covered ground surface. The most widely used
sonic ranging sensors are manufactured by Campbell Scientific, models SR-50/ SR-50A. Several
other sensors in use are Sommer Ultrasonic snow depth sensor USH-8 (Austria), MPS System
SwS-3 (Slovakia), ultrasonic snow level meters model JMA-95-1, from Ogasawara Keiki
Seisakusho, and JMA-89, JMA-93, JMA-04-1 from Ultrasonic Kaijo Sonic Corp (Japan).
Other sensors used are the so-called weight sensors or snow pillows, determining the snow
water equivalent by measuring the weight of the snow over a specific area. The working principle
of the sensor is based on the detection of the hydrostatic pressure caused by the layer of snow on
top of the pillow. They use four tensiometric sensors, situated under each corner of a steel frame
on which the plate is mounted. Data on the weight of the overlying snow is combined with the
snow depth measured with an ultrasonic snow sensor, to derive the snow water equivalent. The
standard dimensions of a snow pillow are 3 x 3m or 4 x 4m.
Summary on the Topic of Derived Measurements
Among the 54 surveys, there are only 7 countries, or 13% of the surveyed countries, that
currently derive any parameters from automatic stations in their services. The derived parameters
are depth of freshly fallen snow (Austria, Canada, Finland and France), precipitation type (Finland,
France, Germany and The Netherlands), and snow amount (Denmark).
Meteorological Service of Canada has invested considerable effort towards developing an
algorithm to derive snowfall from snow on the ground measurements, using data from three
collocated sonic ranging sensors and the total precipitation accumulation measurement, all
available at its automatic weather stations. The algorithm is based on finding correlations between
positive changes in snow depth between concurrent snow depth time series. These concurrent
measurements are combined to construct a “consensus” SF measurement; while the output of a
WG is used as a verification to ensure that changes observed in the snow depth levels reported by
the SR50’s, had indeed occurred during a period of precipitation (Fischer, A, et all, 2008).
The testing to date of this algorithm indicate that using a triple configuration of SR50
Sensors to derive daily snowfall from changes in the snow pack levels, statistically, yields a better
answer then deriving daily snowfall values using just one SR50. The use of the output of a WG as
precipitation occurrence verification helps minimizing false reports of snowfall. That being said,
better diagnosis of snow drift will be needed to further minimize snowfall measurement errors.
While this algorithm has shown promise, more work is needed to deal with the snow drift, and
improve the measurability of very light snowfall events, often missed by the WG, as well.
Test and Development of New Instruments and Methods
Among the 54 surveys, 18 countries, or about a third of the surveyed countries, currently
test/develop new instruments and methods of measurement of solid precipitation at Automatic
Weather Stations. These are Austria, Denmark, Germany, France, Lithuania, Switzerland,
8
Morocco, Slovakia, Canada, Portugal, New Zeeland, Norway, USA, Sweden, United Kingdom,
Ukraine, The Netherlands, and Republic of Belarus. Their efforts are focused on four topics: snow
depth sensors, surfaces for the snow depth sensors, precipitation gauges (WG and TBRG) and their
configurations, and present weather sensors. The efforts include the comparison of instruments
from different manufacturers, improvement of sensor outputs with the use of measurements, and
developing new algorithms.
CONCLUSIONS AND RECOMMENDATIONS
The results of the survey indicate that a large variety of automatic precipitation gauges is
currently used for the measurements of precipitation amount and solid precipitation, worldwide,
including in use in the same country. Also there is a wide range of parameters reported, measured
or derived, which could create difficulties for the users accessing the precipitation data bases. The
gauges vary in terms of their measuring system, orifice area, capacity, sensitivity, and
configuration. This variety exceeds by far the variety of manual standard precipitation gauges
(Goodison et. al, 1998). The detailed analysis of the results will be published by WMO in a IMOP
report in 2010.
These results lead to a legitimate question on the spatial and temporal consistency of
precipitation data, including in the same country using more than one method and instrument to
measure the same parameter.
The need for increasing the accuracy and the consistency of local and global precipitation
time series and the effective support of applications relying on the ground observation of
precipitation (e.g. satellite measurement validation), require that an up-to-date body of knowledge
on the performance of measurement of the gauges in use and their relativity, in all climate
conditions, is made available. Developing the understanding of the correlation between
measurements obtained from different configurations, is a key contribution of the instrument
experts towards improving the accuracy and validity of scientific and engineering studies, using
precipitation data. This should cover maintaining the consistent performance of measurement of
gauges in their operational configuration, throughout their expected life cycle, and in all conditions
where are expected to operate.
Given the information obtained through the 2008 CIMO survey, and the significant local and
regional efforts, it is recommended that WMO leads the organization of a field intercomparison of
automatic precipitation gauges and their configurations, in various climate conditions. The
proposed intercomparison should build on the significant work currently underway in many
countries, and should aim at understanding and improving the ability to reliably measure solid
precipitation using automatic gauges, to meet the expectations of their user communities.
The following objectives are proposed for a WMO led intercomparison of methods and
instruments for automatic snowfall/snow depth/precipitation measurements:
 evaluate and report on the performance of the instruments and methods in field conditions;
 evaluate and provide guidance on the operational configuration of automatic gauges (e.g.
use of heating, use of windshields, height of installation, redundancy), to achieve data
consistency.
 assess the feasibility of developing multi-parameter algorithms to improve the quality of
precipitation data reported from an Automatic Weather Stations.
 provide datasets to support studies on the homogeneity of long-term records of
precipitation with special consideration given to solid precipitation;
 enable the development of adjustment procedures of the systematic errors of precipitation
measurements.
 establish the WMO field reference standard using recording precipitation gauges;
9


provide feedback to manufacturers to enable the development of recording precipitation
gauges, addressing known limitations.
draft recommendations for consideration by CIMO.
ACKNOWLEDGEMENTS
The authors would like to thank to Igor Zahumensky and Isabelle Rüedi (WMO CIMO
Secretariat), Michel Leroy and Eckhard Lanzinger (WMO-CIMO Expert Team on Surface Based
Instrument Intercomparisons and Calibration Methods), and Alexandre Fischer and Craig Smith of
the Environment Canada, for their support and input in developing, distributing, collecting, and
analyzing the information used in this paper.
In addition, our thanks to all those who responded to the CIMO questionnaire and provided
the information on behalf of their NMHSs.
REFERENCES:
B.E. Goodison et al, The WMO Solid Precipitation Measurement Intercomparison” - Final Report,
(WMO/TD - No. 872, IOM 67), 1998
WMO 2008: WMO Guide To Meteorological Instruments and Methods of Observation(WMONo.8)
B. Sevruk, WMO Questionnaire On Recording Precipitation Gauges State-Of-The-Art, Water
Science and Technology, 45(02), 139-145, 2002
Sevruk, B. and Klemm S. (1989). Catalogue of national standard precipitation gauges. WMO,
Instruments and Observing Methods Rep., WMO/TD-No. 313, 50 pp., 1989.
Yang, D et al, Bias correction of long-term (1973-2004) daily precipitation data over the northern
regions, Geophysical Research Letters, Vol 32, 2005
Smith, C., Watson, S., An assessment of the double Alter wind shield for reducing wind bias in
snowfall measurements made with the Geonor T-200B, CMOS Congress 2009, May 31-June
4, Halifax, NS, Canada
Fischer, A., Nitu, R., Improving the quality of in-situ measurement of solid precipitation in Canada,
TECO conference, St Petersburg, Nov, 2008
Nespor, V. and Sevruk, B, Estimation of Wind-Induced Error of Rainfall Gauge Measurements
Using a Numerical Simulation, , Journal of Atmospheric and Oceanographic Technology,
Volume 16, 1998
Zweifel, A, and Sevruk, B, Accuracy Of Solid Precipitation Measurement Using Heated
Recording Gauges In The Alps
10