9th INTERNATIONAL WORKSHOP on PRECIPITATION IN URBAN AREAS
Urban Challenges in Rainfall Analysis
6-9 December, 2012, St. Moritz, Switzerland
WMO INTERCOMPARISON OF INSTRUMENTS AND METHODS FOR
THE MEASUREMENT OF PRECIPITATION AND SNOW ON THE
GROUND
by
(1)
(2)
(3)
(4)
R. Nitu , O. Aulamo , B. Baker , M. Earle , B. Goodison(5), J. Hoover(6), J. Hendrikx(7), P.
Joe(8), J. Kochendorfer(9), T. Laine(2), E. Lanzinger(10), H. Liang(11), L. Lanza(12), S. Landolt (13), R.
Rasmussen(14), Y.A. Roulet(15), C. Smith(16), A. Samanter(17), F. Sabatini(18), E. Vuerich(19), V.
Vuglinsky(20), M. Wolff(21), and D. Yang(22)
(1) Environment
Canada, 4905 Dufferin St, Toronto,ON, Canada (rodica.nitu@ec.gc.ca)
Meteorological Institute, Arctic Research Centre, Sodankyla, Finland (osmo.aulamo@fmi.fi)
(3)
NOAA Air Resources Laboratory, 456 S. Illinois Ave, Oak Ridge, TN, USA (bruce.baker@noaa.gov )
(4) Environment Canada, 4905 Dufferin St, Toronto,ON, Canada (michael.earle@ec.gc.ca)
(5) WMO, EC-PORS, GCOS, 7 bis, avenue de la Paix, CH 1211 Geneva, Switzerland (barry.go@rogers.com)
(6) Environment Canada, 4905 Dufferin St, Toronto, ON, Canada (jeffery.hoover@ec.gc.ca)
(7) Montana State University, Bozeman, MT, 59717-3480, USA (jordy.hendrikx@montana.edu)
(8) Environment Canada, 4905 Dufferin St, Toronto, ON, Canada (paul.joe@ec.gc.ca)
(9) NOAA Air Resources Laboratory, 456 S. Illinois Ave, Oak Ridge, TN, USA (john.kochendorfer@noaa.gov)
(10) Deutscher Wetterdienst, Frahmredder 95, 22393 Hamburg, Germany (eckhard.lanzinger@dwd.de )
(11) China Meteorological Administration, No. 46 Zhongguacun, Nandajie, Beijing 100081, China (lhhaoc@cma.gov.cn)
(12) DICAT, University of Genoa , Genova, Italy (luca@diam.unige.it)
(13) National Center for Atmospheric Research, 3450 Mitchell Lane, Boulder, CO USA (landolt@ucar.edu)
(14) National Center for Atmospheric Research, 3450 Mitchell Lane, Boulder, CO USA (rasmus@ucar.edu)
(15) Météo Suisse, Station Aérologique, Case postale 316, CH-1530 Payerne, Switzerland (yves-alain.roulet@meteoswiss.ch)
(16) Environment Canada, 11 Innovation Boulevard, Saskatoon, SK, Canada (craig.smith@ec.gc.ca)
(17) Environment Canada, 4905 Dufferin St, Toronto,ON, Canada (amal.samanter@ec.gc.ca)
(18) CNR-IBIMET, Institute of Biometeorology, Via Giovanni Caproni, 8, 50145 Florence, Italy (f.sabatini@ibimet.cnr.it)
(19) Italian Met Service – Air Force, Centre of Meteorological Experimentations, Rome, Italy (vuerich@meteoam.it)
(20) State Hydrological institute, 2nd Line 23, St.Petersburg, 199053, Russian Federation (vuglins@vv4218.spb.edu)
(21) Norwegian Meteorological Institute, P.O. Box 43 Blindern, Henrik Mohns plass 1, 0313 Oslo, Norway (mareile.wolff@met.no)
(22) Environment Canada, 11 Innovation Boulevard, Saskatoon, SK, Canada (daqing.yang@ec.gc.ca)
(2) Finnish
ABSTRACT
WMO initiated in 2010 the organization of an intercomparison of instruments and methods of observation of solid
precipitation and snow on the ground, and will be conducted as a joint effort of the WMO Members and instrument
manufacturers. The formal intercomparison is planned to commence in the fall of 2012, being organized on multiple
sites in the Northern and the Southern Hemisphere, and will assess a wide range of instruments and systems currently
used for the measurement of solid precipitation and snow on the ground, with various measuring principles, and
installed in multiple configurations, as well as new technologies. The experiment will take place over two winter
seasons. The organization of the intercomparison on multiple sites will allow the concurrent assessment of the same
sensor models in multiple climates, and over a broad range of operational conditions. This paper presents the
organization of the experiment, an overview of the participating sites and the instruments selected for the
intercomparison.
Keywords: precipitation, snow, measurement, intercomparison
1 INTRODUCTION
The Commission for Instruments and Methods of Observation (CIMO) of the World Meteorological
Organization (WMO) agreed in 2010 to organize, as a matter of urgency, an intercomparison to assess the
impact of automation and to determine the errors in measurement of snowfall, snow depth and solid
precipitation in cold climates, from automatic weather stations. The organization of the Solid Precipitation
Intercomparison Experiment (WMO SPICE) was endorsed at the Sixteenth Congress of WMO and the work
Nitu et al., Intercomparison of precipitation and snow on the ground measurements
1
9th INTERNATIONAL WORKSHOP on PRECIPITATION IN URBAN AREAS
Urban Challenges in Rainfall Analysis
6-9 December, 2012, St. Moritz, Switzerland
commenced in 2011. SPICE is led by an International Organizing Committee (IOC) which brings together
representatives from Canada, China, Germany, Italy, New Zealand, Russia, Switzerland, and the USA.
Canada has assumed the leadership role in the organization and the running of the experiment.
2 SPICE OBJECTIVES
In consultation with a broad range of stakeholders representing the national Meteorological Services, the
WMO Commission for Hydrology, WMO Commission for Agriculture Meteorology, World Climate
Research Program – Working Group on Nowcasting,, the WMO- Executive Council for Polar Observations,
Research, and Services (EC-PORS), the Global Climate Observing System (GCOS), the Remote Sensing
community, the IOC defined the objectives of SPICE, and planning to focus on:
I. Providing guidance on the performance of modern automated systems measuring (i) total precipitation
amount, especially when the precipitation is solid, (ii) snowfall (height of newly fallen snow), and (iii)
snow depth (snow on the ground).
II. Recommending automated field reference system(s) for the unattended measurement of solid
precipitation.
III. Understanding and documenting the differences between a field reference system using an automatic
gauge and different automatic systems, and between automatic and manual measurements of solid
precipitation.
Building on the results and recommendations of previous intercomparisons, SPICE will investigate and
report the measurement and reporting of precipitation amount over various time periods (minutes, hours,
days, season) as a function of precipitation phase, and of snow on the ground (snow depth); as snow depth
measurements are closely tied to snowfall measurements, the intercomparison will address their linkages.
Recommendations will be made to WMO Members, WMO programmes, manufacturers and the scientific
community, on the ability to accurately measure solid precipitation, on the use of automatic instruments, and
the improvements possible. All available remotely sensed precipitation data will be identified in the SPICE
database, however their analysis is beyond the scope of this intercomparison. The SPICE dataset can be later
used for studies contributing to improving the spatial and temporal estimates of precipitation.
3 MOTIVATION
The use of a broad range of instruments and configurations significantly impacts the ability to derive
homogeneous results at large scale, and has serious consequences for the accuracy and consistency of local
and global precipitation time series (Sevruk, 1994). The results of the 2008 CIMO Survey on instruments
and methods for measuring solid precipitation, published in WMO/TD-No. 1544, IOM No. 102 (Nitu and
Wong, 2010), indicated that, globally, manual observations are still the primary method for measuring solid
precipitation, snowfall, and snow on the ground; however, the transition from manual to automatic
instruments has accelerated in many developing countries, like Russia, China, and the Himalayan region.
The CIMO Survey report indicates that a large variety of automatic instruments are being used for measuring
solid precipitation, worldwide, including within the same country. This variety exceeds by far the variety of
manual standard precipitation gauges (Goodison et al., 1998). The instruments vary in terms of their
measuring principle, sensitivity, capacity, orifice area, installation height, whether or not using a wind shield.
The impact of the technology on the variability of measurement is illustrated in Figure 1, which shows
accumulated precipitation amounts from automatic gauges of different types and configurations during rain
and snow events at the Canadian Centre for Atmospheric Research Experiments (CARE) measurement site
on Feb. 11-12, 2009, and Jan. 28-29, 2009, respectively. For the rain event (Figure 1a), the specific
instrument and configuration does not impact significantly the measured precipitation amount, with the
exception of the optical instrument/ system (Vaisala PWD22). For the snow event (Figure 1b), however,
marked differences of the measured amounts of precipitation are observed among the different instrument
types and configurations.
As the wind induced gauge undercatch is the largest source of errors for measurements of solid and mixed
precipitation (Goodison et al., 1998), the 1987-1993 WMO Intercomparison recommended wind adjustments
for the methods assessed, developed using observations available at the time, mainly daily and synoptic
observations. Today’s automatic stations provide precipitation data hourly or even minutely. At these time
scales the dynamics and the climatology of precipitation is different. To address this and contributing to
making available accurate precipitation data sets, SPICE will focus on developing adjustment functions for
Nitu et al., Intercomparison of precipitation and snow on the ground measurements
2
9th INTERNATIONAL WORKSHOP on PRECIPITATION IN URBAN AREAS
Urban Challenges in Rainfall Analysis
6-9 December, 2012, St. Moritz, Switzerland
the current measurement methods, using higher temporal resolution observations of wind and precipitation
type.
Figure 1 – Accumulated precipitation amounts over 24 hours from different automatic gauges and configurations for (a)
a rain event on Feb. 11-12, 2009 (air temperature > 4.6 °C) and (b) a snow event on Jan. 28-29, 2009 (air temperature <
-6.4 °C) at the Canadian Centre for Atmospheric Research Experiments (CARE) site in Egbert, ON. Weighing gauges,
tipping bucket rain gauges, and optical gauges are denoted by green, yellow, and blue bars, respectively. For weighing
gauges, the wind shield configuration is provided in parentheses: DFIR for double-fence intercomparison reference
shield, S for single-Alter shield, T for Tretyakov shield, and U for unshielded.
The 1987-1993 WMO intercomparison recommended as a secondary field reference standard, the Double
Fence International Reference (DFIR), a double fence with the outside diameter of 12 m, surrounding a
manual Tretyakov gauge (Goodison et al., 1998). Since then, for many experiments, the field reference was
configured by replacing the manual gauge with automatic gauges to address the need for higher temporal
resolution reference data. A thorough characterisation of a field reference using automatic gauges is a key
expectation of the scientific community and is a major objective for SPICE.
4 ORGANIZATION OF WMO SPICE
In early 2012 the WMO Members and the manufacturers were invited to participate in the coordinated
SPICE experiment focusing on the automatic instruments currently operational and new and emerging
technologies.
The IOC accepted the proposals for organizing SPICE experiments on 15 field sites hosted by Australia,
Canada, Chile, Finland, Japan, Norway, New Zealand, Russia, Switzerland, Poland, and United States of
America, and a laboratory in Italy. The instruments selected for SPICE are provided by the host sites,
representing their measurement programs, by other WMO Members, and by instrument manufacturers.
Overall, 24 types and models of instruments measuring solid precipitation, snowfall, and snow on the ground
are included in SPICE, representing current configurations of measurement networks.
The organization of the intercomparison and the collection and the sharing of data are governed by the
SPICE Data Protocol, which enables the data exchange, while aiming at protecting the integrity of the SPICE
dataset and of the results published in the Final Report.
4.1
Participating Instruments
4.1.1 Measurement of Precipitation Amount
The instruments accepted for the intercomparison cover all the principles of operation used worldwide for
the measurement of precipitation. Three categories of automatic instruments have been included in SPICE;
tipping bucket rain gauge (TBRG), weighing gauge (WG), and optical sensors.
The TBRG measures the amount of liquid precipitation by recording the number of tips which correspond to
a fixed amount of precipitation, which is the nominal value of each of the two tipping buckets. The TBRGs
vary in sensitivity and size and only heated TBRGs are included in SPICE. Overall, TBRG provide a large
percentage of the precipitation amount measurements in all climate regimes, estimated at about 80% of the
total of observations by automatic instruments (Nitu and Wong, 2010).
Nitu et al., Intercomparison of precipitation and snow on the ground measurements
3
9th INTERNATIONAL WORKSHOP on PRECIPITATION IN URBAN AREAS
Urban Challenges in Rainfall Analysis
6-9 December, 2012, St. Moritz, Switzerland
A weighing gauge (WG) weighs the precipitation collected in a bucket and derives the precipitation amounts
based on the detected mass or load. The WGs included in SPICE cover the measuring principles currently
operational: vibrating wire load, single point electronic load, and strain gauge; with the collecting capacity
from 600 mm to 1500 mm, and the collecting area is 200 cm2 or 400 cm2. Heating of the WGs is a feature
increasingly used to address the ice build-up and snow capping and will be assessed in SPICE.
An optical sensor uses the scattering or obscuration by hydrometeors, to determine the type of precipitation
and to estimate its intensity and accumulation. Given the increasing use of optical sensors for deriving
precipitation accumulation, SPICE will focus on assessing the ability of these instruments to provide
representative reports of accumulated precipitation over various time scales.
4.1.2 Use of Wind Shields
Recognizing the impact of windshields on the catch efficiency of instruments, SPICE will investigate and
provide recommendation on the optimal gauge and shield combination for each type of instrument, for
different collection conditions/climates, by conducting relevant field experiments and theoretical studies of
the airflow/snowflake trajectory around shields. The results of the CIMO 2008 survey indicate that at the
time only 28% of the automatic instruments (WGs and TBRGs) were configured with wind shields, and the
WGs are used in a much larger proportion with shields. The wind shields in use at the time of the survey
were the Alter, Nipher, Tretyakov, or of a special design (e.g. Japan Meteorological Administration).
4.1.3 Measurement of Snow on the Ground and Snowfall
The 2008 CIMO survey results indicate that automatic instruments measuring the depth of snow on the
ground were used in a small proportion. The significant decrease of manual measurements of snow on
ground and snowfall, in many countries, requires a better understanding of the performance of automatic
instruments and their role in replacing manual observations. Most instruments measuring snow depth are
ultrasonic or sonic ranging depth sensors, which measure the elapsed time between emission and return of an
ultrasonic pulse sent vertically down to the ground surface. Other snow depth instruments operate on the
principle of phase variation of visible laser when it bounces off the snow surface. Both technologies are
represented in SPICE.
4.2
Configuration of Working Field Reference Systems
4.2.1 Reference for Measurement of Precipitation Amount
The term DFIR (Double Fence Intercomparison Reference) as defined during the previous WMO Solid
Precipitation Intercomparison (Goodison et al., 1998) refers to the complete system comprising the octagonal
double-fence and the Tretyakov manual collector and shield placed in its centre. In order to differentiate the
DFIR system from a similar configuration using an automatic gauge in the centre of the octagonal doublefence, the IOC decided to refer to the latter as Double Fence Automatic Reference (DFAR).
For the purpose of running SPICE, the IOC defined three levels of working field reference configurations:
i. Working field reference R1: the DFIR designated as secondary field reference at the end of the WMO
Intercomparison of Solid Precipitation 1989-1993 (Goodison et al., 1998).
ii. Working field reference R2: a heated automatic WG with a single Alter windshield within a DFIRfence, together with a capacitive precipitation detector.
iii. Working field reference R3: a pair of identical automatic WGs heated in the same manner, one
unshielded and the second installed with a single Alter shield, and a capacitive precipitation detector.
The configuration of the R3 reference is based on the fact that, globally, the operational instruments
measuring precipitation amounts are typically either unshielded or have a single shield (Nitu and Wong,
2010). The same rationale was applied during the first WMO Solid Precipitation intercomparison. The IOC
selected for inclusion in the working field reference system of SPICE two gauges with wide operational use,
and which have demonstrated consistent performance during the tests preceding the formal experiment.
These are the Geonor T200-B3, with 3 transducers and capacities of 600 mm or 1000 mm and the OTT
Hydromet GmbH Pluvio2, 200 cm2 inlet opening, with 1500 mm capacity, both heated.
4.2.2 Reference for Measurement of Snow on the Ground and Snowfall
The recommended reference for the measurement of Snow on Ground and Snowfall are manual
measurements taken in a predefined configuration (i.e. using snowboards) and supported by an array of
ancillary measurements. Additionally, function of the time scale of the observations, a composite reference
Nitu et al., Intercomparison of precipitation and snow on the ground measurements
4
9th INTERNATIONAL WORKSHOP on PRECIPITATION IN URBAN AREAS
Urban Challenges in Rainfall Analysis
6-9 December, 2012, St. Moritz, Switzerland
derived using instruments available would be explored. Additional work is being conducted to refine this
reference.
4.3 Participating Sites
The 15 participating sites, organized by country and the coordinating agency are provided below, along with
the associated working field references available on site:
i. Australia (Snowy Hydro Ltd.): Guthega Dam weather station, Kosciuszko National Park, New South
Wales, Australia (R3).
ii. Canada (Environment Canada): Centre for Atmospheric Research and Experiments (CARE), Egbert,
Ontario (R1, R2, R3); Bratt’s Lake, Saskatchewan (R2, R3); Caribou Creek, Saskatchewan (R2, R3).
iii. Chile (Centro de Estudios Avanzados en Zonas Áridas): Tapado AWS, Región de Coquimbo (R3).
iv. Finland (Finnish Meteorological Institute): Sodankylä Arctic Research Centre, ARC (R2, R3); this site
is a primary location for the assessment of snowfall and snow on the ground.
v. Japan (National Agriculture and Food Research Organization, NARO; National Institute of Polar
Research, NIPR): NARO site, Joetsu; NIPR site, Rikubetsu, Hokkaido (experiments to start in 2013).
vi. New Zealand (National Institute of Water and Atmospheric Research Ltd.): Mueller Hut Electronic
Weather Station site (R3).
vii. Norway (Norwegian Meteorological Institute): Haukeliseter, Vinjeveien, Telemark, Norway (R2, R3).
viii. Poland (Institute of Meteorology and Water Management, National Research Institute of Poland): Hala
Gasienicowa, with a focus on the measurement of snow on the ground and snowfall.
ix. Russian Federation (Roshydromet): Valdai, State Hydrological Institute, Valdai Branch, and Voljskaya
Hydro Meteorological Observatory (Volga), Gorodec, Nijny Novgorod Reg.
x. Switzerland (MeteoSwiss): Weissfluhjoch (Davos); the site is a partnership between MeteoSwiss and
the Swiss Institute for Snow and Avalanche Research (R2, R3).
xi. United States of America (National Oceanographic and Atmospheric Administration (NOAA)):
NOAA/FAA/NCAR Winter Precipitation Testbed (Marshall) site, Boulder, Colorado (R1, R2, R3).
Additionally, the WMO-CIMO Lead Centre “Benedetto Castelli” on Precipitation Intensity, University of
Genoa, Italy, will contribute to SPICE with the laboratory characterisation of instrument performance.
Figure 2: SPICE participating Sites
5 RUNNING THE EXPERIMENT
The SPICE experiment will start in the fall of 2012, and will run for two winter seasons, in both
hemispheres. The IOC has decided that each site host will share with the Instrument Providers the data from
their instruments, complemented by a set of ancillary measurements.
Nitu et al., Intercomparison of precipitation and snow on the ground measurements
5
9th INTERNATIONAL WORKSHOP on PRECIPITATION IN URBAN AREAS
Urban Challenges in Rainfall Analysis
6-9 December, 2012, St. Moritz, Switzerland
Given the duration of the experiment, the broad engagements, and the need to actively engage the
community in the validation of the experiment results, the IOC has decided that publications prior to the
publishing of the final report are encouraged within the terms of the SPICE Data Protocol.
The project will conclude with the publication by WMO of the Final Report, which will include the results
addressing the project objectives, as derived from the results from each of the participating sites and the
relative assessment of results for similar configurations in various climate conditions.
REFERENCES
Nitu, R., and Wong, K. (2010), CIMO Survey on national summaries of methods and instruments for solid precipitation
measurement at AWS, WMO/TD-No. 1544, IOM No. 102, World Meteorological Organization, 57 p.
Goodison, B., Louie, P. Y. T., and Yang, D. (1998), WMO solid precipitation measurement intercomparison final
report, WMO/TD-No. 872, IOM No. 67, World Meteorological Organization, 212 pp.
Sevruk, B. (1994), Spatial and temporal inhomogeneity of global precipitation data, in: Global Precipitation and
Climate Change, NATO ASI Series, I 26, 219-230, Springer Verlag, Berlin.
Report of Meeting, International Organizing Committee (IOC) for the WMO Solid Precipitation Intercomparison
Experiment (SPICE), First Session, 5-7 October 2011, Geneva, Switzerland.
Report of Meeting, International Organizing Committee (IOC) for the WMO Solid Precipitation Intercomparison
Experiment (SPICE), Second Session, 11-15 June 2012, Boulder, CO, USA
http://www.wmo.int/pages/prog/www/IMOP/reports/2012/IOC-SPICE-2.pdf
Nitu et al., Intercomparison of precipitation and snow on the ground measurements
6