1. WMO Intercomparison of Instruments and methods for the Measurement of Solid Precipitation and Snow on the Ground, WMO-SPICE:
Pre-SPICE Studies
R. Rasmussen(1) ,R. Nitu (2),B. Baker(3),E Lanzinger (4),D Yang(7),C Smith(7),Y.A. Roulet (5),B Goodison((15) ,H Liang(6),F Sabatini(8),J Kochendorfer (3),M Wolff(9) ,J Hendrikx (10), E Vuerich(11),L Lanza(12),P. Joe(2),O Aulamo(13),V Vuglinsky(14) , Michael Earle(2) ,Jeffery Hoover(2), S Landolt(1)
(1) National Center for Atmospheric Research, Boulder, USA, (2) Environmemt Canada, Toronto, Canada, , 3) NOAA Air Resources Laboratory, Oak Ridge, TN, ,4)Deutscher Wetterdienst, Hamburg, Germany,(5)Météo Suisse,PAYERNE, Switzerland, (6)China Meteorological Administration, BEIJING, China. 7)Environment Canada, Saskatoon, Canada . (8) CNR-IBIMET, Institute of Biometeorology, Florence, Italy, (9)Norwegian Meteorological Institute,Oslo, Norway, (10)Montana State University, Montana, U.S.A., (11) Italian Met Service – Air Force, Centre of Meteorological Experimentations, Vigna di Valle (Rome) Italy, 12) DICAT - Department of Civil, Environmental and Territorial Engineering, University of Genoa , GENOVA Italy , 13) Finnish Meteorological Institute, Arctic Research Centre, Sodankyla , (14) State Hydrological institute,St.Petersburg, Russian Federation, (15) World meteorological Organization, EC-PORS, GCOS, Geneva Switzerland

2. How well are we measuring snow?

3. Challenge to Measuring Snowfall
Updraft generated upstream of gauge

4. What is SPICE?
International Intercomparison of: 1) automatic weighing snowgauges 2) snow depth sensors at various sites around the world

5. Due to updraft, snowgauges significantly undercatch
Data from single alter GEONOR gauge at Marshall (box plots)
Model results (lines) from Fluent simulation of flow past single Alter GEONOR with snowflake trajectory modeling
Single Alter GEONOR
Theriault et al. 2011

6. Challenges to Measuring Snowfall
Capping of gauge

7. When and Where? Main experiment: - Winter 2012/2013
Pre-SPICE: Winter 2011/2012 at:
Canada (CARE site near Toronto and Bratt’s Lake, Saskachewan)
USA (Boulder CO site near NCAR)
Finland (Jokioinen, near Helsinki)
Norway (Haukiliseter site, south-central alpine site)
Switzerland (Weissfluhjoch/Davos)
Italy (Lab testing at U. of Genoa)
German (Lab outdoor testing at Deutsher Wetterdienst, Hamburg)
Rodica Nitu will present an overview of the organizational structure of SPICE in the next talk

8. Key goal of Pre-SPICE: Establish Field Reference Systems for SPICE
Need to establish a recognized field reference using automatic gauges to calculate the undercatch as a function of wind speed (transfer function).
Proposed configuration: DFIR shield surrounding a Weighing Gauge (WG)
Develop transfer functions to characterize the proposed reference against the WMO Secondary Reference (Tretyakov gauge inside DFIR);
Options for the configuration of a Filed Reference using WG:
Type of WG: proposed Geonor T-200, OTT Pluvio 2;
Configuration of WG:
heated/non-heated;
wind shield:
DFIR,
Double Alter, Belfort Double Alter.

9. Instruments and Site Focus for Pre-SPICE
A WMO secondary field reference, using a manual Tretyakov gauge installed in a DFIR, was used in Canada and Finland.
A DFIR shield surrounding an automatic precipitation weighing gauge was configured and assessed as a field working reference in Canada (CARE and Bratt’s Lake), Finland (Jokioinen), Norway (Haukeliseter), and USA (Boulder),(DFAR)
In Switzerland, the focus of the pre-SPICE experiment was the assessment of the heating of a weighing gauge, in alpine conditions, in the presence of large amounts of snowfall.
A laboratory calibration of the instruments considered for use as part of the working field reference was performed at the Precipitation Intensity Lead Centre in Genoa (Italy). This focused on the testing of the weighing gauges at temperatures representative of the winter operating conditions.
Week long meeting was held the week of June 11th in Boulder, Colorado to review the results of the Pre-SPICE testing

10. Current reference for snowfall observations are manual
Bush Gauge
Primary Reference (Valdai, Russia)
Only one!

11. Secondary Reference: DFIR surrounding a Tretyakov precipitation gauge
WMO Solid Precipitation Measurement Inter-comparison (WMO CIMO IOM Report No. 67, WMO/TD-No. 872, 1998
Recommended : “The Double Fence Intercomparison Reference (DFIR) should be accepted as a secondary reference for the (manual) measurement of solid precipitation;” (section 6.1.2)
(DFIR): “The octagonal vertical double-fence inscribed into circles 12 m and 4 m in diameter, with the outer fence 3.5 m high and the inner fence 3.0 m high surrounding a Tretyakov precipitation gauge mounted at a height of 3.0 m.”( section 2.2.2)
Double Fence Intercomparison Reference

12. Cross section of DFIR shield

13. Set-up and some results from Weissfluhjoch/Davos
Pre-SPICE Winter
Set-up and some results from Weissfluhjoch/Davos

14. Testing at Davos, Switzerland by MeteoSwiss
Testing at Davos, Switzerland by MeteoSwiss - Impact of heating on OTT Pluvio2 200 cm2 gauge with Tretyakov shield

16. Schematic of ice bridging inside the OTT PLUVIO2

17. Summary of Testing at Switzerland Impact of heating unshielded OTT Pluvio2 200 cm2 gauge
1. For prevailing alpine conditions, measurement is difficult with an unheated gauge: snow can induce mechanical effects on the sensor, with possible formation of ice or snow bridges between the housing and the weighing elements.
2. Heating of the ring is mandatory to ensure good quality measurements in Alpine winter conditions with high snow amount.
3. No under-catch due to the heating.

18. Jokioinen: Pre-SPICE field 60° 48' 50.44" N, 23° 29' 51.40” E,
104 m, from the sea level
Measurements –
Manual measurements only once every day at 0600 UTC
Compared manual measurement in DFIR with Tretyakov shield and can with automated measurement in DFIR with OTT Pluvio2 gauge
Finnish Meteorological Institute
15/04/2017

19. Rain
Both gauges in DFIR
Finnish Meteorological Institute
15/04/2017

20. Snow
Both gauges in DFIR
Finnish Meteorological Institute
15/04/2017

21. Results from testing in Finland:
The daily precipitation amount from the automated gauge in the DFIR compared well with the manual measurement from the Tretyakov can also in a DFIR.
Suggests that can use an automated gauge in a DFIR as a reference for SPICE (DFAR).
Finnish Meteorological Institute
15/04/2017

22. Summary of results from the Canadian contribution to Pre-SPICE
Toward the definition of a SPICE field reference
Rodica Nitu, Paul Joe, Daqing Yang, Craig Smith, Michael Earle, Jeff Hoover, Jim Waller, Faisal Boudala
Pre-SPICE Workshop
Boulder, Colorado
June 11, 2012

23. Measurements from two field sites
Centre for Atmospheric Research Experiments (CARE)
Egbert, Ontario
CARE
Bratt’s Lake (BL)
Bratt’s Lake, Saskatchewan
Canada
Bratt’s Lake
CARE
Bratt’s Lake

24. Instruments at the Canadian CARE site
Tretyakov manual gauge in DFIR
Geonor T-200 heated (DFIR)
Pluvio 2 (Belfort Double Alter)
Geonor T-200 in BDA

25. The Bratt’s Lake Precipitation Intercomparison Facility
Bratt’s Lake, Saskatchewan:
Under test:
Unheated Geonor in a DFIR
Heated Geonor (using the CRN heating algorithm), in a DFIR
Heated Geonor gauge (CRN heating)
Top left shows a summary of the Geonor Catch Efficiency – Wind Speed relationships for the shield configurations shown on the right. The dashed line is the dry snow relationship for the DFIR as compared to the Valdai bush gauge by Yang et al. The 3 shield configurations are the double Alter, single Alter, and large double fence. The bottom photo shows the test facility after infrastructure upgrades (note the trench scars) with the DFIR at the top left, the Geonor in a large double fence bottom left and the RCS top centre. There are 20 pedestals w/ 14 currently empty.
Unheated Geonor gauge

26. Summary of Canadian Pre-SPICE testing
Rim heating has a positive impact on the gauge performance (accuracy, improved catch), both GEONOR T-200 or OTT PLUVIO2.
Both GEONOR T-200 and OTT PLUVIO2 performed well in terms of accuracy
Either GEONOR T-200 or OTT PLUVIO2 acceptable at weighing gauge for Field Reference

27. Marshall (Boulder, CO USA) site with DFIR and Alter shielded snow guages

28. Summary of Pre-SPICE Results from Marshall Field Site, USA
1. Gauge performance in Small DFIR and standard DFIR similar in terms of accuracy.
2. Provided estimate of the performance of various wind shield types.
3. Noted the importance of heating the entire collection tube of the GEONOR T-200 gauge.
4. Noted that the current OTT PLUVIO2 heating algorithm may not be sufficient for heavy, wet snow conditions (dumps occurred in this situation).

29. Test site Haukeliseter, Norway Pre-SPICE Data
Mareile Wolff, Ketil Isaksen, Ragnar Brækkan
Norwegian Meteorological Institute, Oslo, Norway

30. Haukeliseter
59°48.71’N, 7°12.86’E, 991m

31. Summary
Identified the need for improved identification of the onset and offset of weak precipitiation events
- Suggested use of a precipitation detector, yes/no sensor, video camera
Blowing snow needs to be taken into account in open, high alpine sites

32. Laboratory tests of SPICE reference instruments
WMO-CIMO LEAD CENTRE “BENEDETTO CASTELLI” ON PRECIPITATION INTENSITY ITALY
Laboratory tests of SPICE reference instruments
2nd WMO SPICE, Boulder ,CO, USA, June 2012

33. Experiment setup Temperature tests
Gauges in the cold chamber
Set-up for OTT
Set-up for Geonor

34. Conclusions 3/3
The dynamic tests show that both gauges are able to reproduce likely real world events with comparable performance;
The results from laboratory tests open the way for possible improvements of their accuracy and precision (including low intensities, the interest for SPICE)
The Lead Centre is currently conducting additional tests.
Matteo Colli from the U. of Genoa is currently visiting NCAR and testing the GEONOR T-200 gauge with the NCAR snow machine for heater performance for various snowfall rate and temperature conditions (snow capping elimination).

35. Reference Systems for SPICE
LWE Precipitation Amount

36. Field Reference for Precipitation Amount
As agreed during the Geneva Meeting October 2012
R1: DFIR + Tretyakov gauge (manual measurements) + Tretyakov shield designated in the intercomparison as secondary field reference WMO/TD-872/1998);
R2: DFIR + automatic weighing gauge (AWG) + shield the model and the configuration of the AWG and its shield will be determined at the end of the 11/12 pre-SPICE experiment.
R3: An automatic weighing gauge in a (small) windshield with sufficient characterization and history, to have a degree of confidence for the purpose of meeting specific objectives, as agreed between the host country and the IOC. The characterization of R3 must be done in relation to the R1 and R2, and could be done as part of SPICE. This is a pragmatic approach for sites contributing to meeting the SPICE objectives (e.g. complex terrain with heavy wet snow), but where the installation of a DFIR is not feasible.

37. Working Field Reference for Precipitation Amount
As agreed during the Boulder Meeting June 2012
R1: DFIR + Tretyakov gauge (manual measurements) + Tretyakov shield designated in the intercomparison as secondary field reference WMO/TD-872/1998);
R2: DFIR + automatic weighing gauge (AWG) + shield
Shield will be an Alter shield
Either GEONOR T-200 or OTT PLUVIO2 acceptable as the weighing gauge
Both gauges will be heated with the CRN heating algorithm (heat between +2 and -5 C, every 24 hours otherwise)
A capacitive precipitation detector will be used to detect the presence of precipitation

38. Working Field Reference for Precipitation Amount
As agreed during the Boulder Meeting June 2012
R3: An automatic weighing gauge in a (small) windshield paired with an unshielded gauge
- The Working Field Reference System type R3 consists of a pair of identical automatic weighing gauges heated in the same manner, one being unshielded and the second installed with a single Alter Shield. The configuration of the Alter shield is specified by the SPICE IOC. Gauge type: GEONOR T-200 or OTT PLUVIO2
- A capacitive precipitation detector will be used to detect the presence of precipitation. + + A

39. Working Field Reference for Precipitation Amount
As agreed during the Boulder Meeting June 2012
R3: An automatic weighing gauge in a (small) windshield with sufficient characterization and history, to have a degree of confidence for the purpose of meeting specific objectives, as agreed between the host country and the IOC.
- The Working Field Reference System type R3 consists of a pair of identical automatic weighing gauges heated in the same manner, one being unshielded and the second installed with a single Alter Shield. The configuration of the Alter shield is specified by the SPICE IOC.
- A capacitive precipitation detector will be used to detect the presence of precipitation.
General: All participating sites focusing on the assessment of the measurement of precipitation amount shall have, as the minimum requirement, an R3 type reference.
By deriving the catch ratio of the two gauges in the R3 reference, the applicability of transfer functions for different sites and climatology's will be evaluated. An R3 reference must also be installed at all S1 and S2 sites in order to establish traceability of the R3 reference to the R2 and R1 reference systems.
It was agreed that transfer functions for various gauge configurations will only be determined at S1 or S2 sites.

40. How well are we measuring snow?
Hopefully, after SPICE
a lot better!

41. Thank You! rasmus@ucar.edu