Gauge Precipitation and Snow Depth Measurements

Daqing Yang University of Alaska Fairbanks

Barry GoodisonEnvironment Canada

• Gauge network: global coverage with various operational, national/regional networks.

• Gauge network data: long-term and fundamental, defining global snowfall/climate regimes and changes.– 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 / snow depth sensor: snowfall observations at the national/regional networks, providing snow depth info, not SWE.

• Snow pillow/snowboard: snow accumulation changes over time -(in)direct measurement of snowfall.

Status of Observations

Arctic Precipitation Issues

• Operational networks – our knowledge base – Decline of the networks in the northern regions, including

Siberia, Alaska and N. Canada– Few stations in the mountain regions – How to sustain and improve the operational networks

• Data quality and compatibility across national boundaries– Large biases in gauge measurements of solid precipitation– Incompatibility of precipitation data due to difference in

instruments and methods of data processing – Difficulties to determine precipitation changes in the arctic

regions

• Validation of precip data, including satellite and reanalysis products and fused products at high latitudes.

Various networks

Synoptic/climate stations on land above 45°N and the Arctic Ocean drifting stations

Russia

Mongolia

Kazakhstan

Greenland

China

Canada

• Sparseness of the networks.

• Uneven distribution of measurement sites, i.e. biased toward coastal and the low-elevation areas, less stations over mountains and oceans.

AMBLER (1994 - 95) HAPPY VALLEY CAMP (1970 - 77) POINT HOPE (1924 - 82)

AMBLER WEST (1981 - 92) KILLIK (1981) PRUDHOE BAY (1986 - 99)

BARTER ISLAND AP (1947 - 1988) KIVALINA (1973 - 75) SAG RIVER DOT (2000 - present)

CANDLE (1903 - 50) KOBUK (1953 - 79) SELAWIK (1953 - 55)

CAPE THOMPSON (1960 - 63) KUPARUK (1983 - present) SHISHMAREF (1919 - 1973)

COLVILLE VILLAGE (1996 - present) LONELY (1977 - 81) SHUNGNAK CAA (1949 - 1950)

DEADHORSE AIRPORT (1999 - present) NOATAK (1917 - 24) UMIAT ARPT (1945 - 2001)

GALBRAITH (1970 - 80) NOORVIK (1997) WAINWRIGHT (1935 - 1968

NWS Climate Station Network

XX

Xhttp://climate.gi.alaska.edu/Stations/Arctic/index.html

NRCS SNOTEL / Wyoming gauge network

www.wcc.nrcs.usda.gov/snotel/Alaska/alaska.html

NRCS National Water and Climate Center

UAF/WERC Kuparukbasin rain

gauge stations

http://www.uaf.edu/water

Upper headwater, July 14-18, 1999

0

1

2

3

4

5

6

7

8

9

10

195 196 197 198 199 200

Julian day

Hou

rly ra

infa

ll (m

m)

0

10

20

30

40

50

60

70

80

90

100

Cum

ulat

ive

rain

fall

(mm

)l

Heavy rainfall events in summer of 1999

NOAA US CRN

http://www.ncdc.noaa.gov/oa/climate/uscrn/

Various gauges and snow fences

National standard gauges tested in Barrow

Canadian Nipher

Hellmann

Russian Tretyakov

US 8”

Blowing snow impact: blocking snow fence

Barrow, CRN- DFIR, Mar 3/03

Barrow, UAF-DFIR, Mar 3/03

Barrow, CRN- DFIR, Mar 3/03Barrow, CRN- modified DFIR, at 2.5m, Mar 3/03

Barrow, UAF Wyoming snow fence, Mar/03 Barrow, UAF DFIR, Mar 03

Blowing snow blocking

gauge / wind shield

Barrow, UAF-DFIR, Mar 3/03

NOAA ETI recording gauge at 1m, Barrow, Mar 3/03

Tretyakov gauge in the DFIR Barrow, Mar 3/03

Blowing snow impact: gauge overcatch?

a) Dikson (73.30N, 80.24E)

05

101520253035

0 2 4 6 8 10 12 14 16 18 20

Daily mean w ind speed at 2m (m/s)

Daily

sno

wfa

ll pre

cipi

tatio

n (m

m)

a) DIKSON, 73.30N, 80.24E

0

20

40

60

80

100

120

1 2 3 4 5 6 7 8 9 10 11 12

Mon

thly

pre

cipi

tatio

n (m

m)

trace amount

wind loss

measured

• 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

Biases in Gauge Measurements

WMO double fence intercomparison reference (DFIR)

in Barrow, AK

WMO Solid Precipitation Intercomparison

CRN modified DFIR

Goodison, B.E., P.Y.T. Louie, and D. Yang, 1998: WMO solid precipitation measurementintercomparison, final report, WMO/TD-No. 872, WMO, Geneva, 212pp.

0

20

40

60

80

100

120

0 1 2 3 4 5 6 7 8 9

Wind speed at gauge height (m/s)

Rat

io o

f gau

ge c

atch

to th

e D

FIR

(%

)

Canadian Nipher NWS 8" AlterNWS 8" unsh Hellmann unshTretyakov

0

5

10

15

20

25

30

35

1 2 3 4 5 6 7 8 9 10 11 12Months

Prec

ipita

tion

(mm

)

trace

wind-loss

measuredOverall mean for the NP drifting stations, 1957-90 (Yang, 1999)

Overall mean for 61 climate stations in Siberia, 1986-92 (Yang and Ohata, 2001)0

10

20

30

40

50

60

70

80

1 2 3 4 5 6 7 8 9 10 11 12

Month

Prec

ipita

tion

(mm

)

trace amount

wind correction

measured

0

5

10

15

20

25

30

Jan

Feb

Mar

Apr

May Jun

Jul

Aug

Sep

Oct

Nov

Dec

measureable trace

05

1015202530354045

Jan

Feb

Mar

Apr

May Jun

Jul

Aug

Sep

Oct

Nov

Dec

tracewetting losswind lossmeasured

Prec

ip (

mm

)Pr

ecip

day

s

Bias corrections of daily precipitation data, Barrow, 1982-83

(Yang et al., 1998)

a) Pm (mm) b) Pc (mm) c) CF

Mean Gauge-Measured (Pm) and Bias-Corrected (Pc) Precipitation, and Correction Factor (CF) for January

-180

-150

-120

-90

-60

-30

0

30

60

90

120

150

180

45 60 75 90

Pc (mm)

0 - 1010 - 2020 - 3030 - 4040 - 5050 - 6060 - 7070 - 8080 - 9090 - 390

-180

-150

-120

-90

-60

-30

0

30

60

90

120

150

18045 60 75 90

CF

1 - 1.11.1 - 1.21.2 - 1.31.3 - 1.41.4 - 1.51.5 - 1.61.6 - 1.71.7 - 1.81.8 - 1.91.9 - 2.3

-150

-120

-90

-60

-30

0

3060

90

120

150

45 6 0 75 90

Pm (mm)

0 - 1010 - 2020 - 3030 - 4040 - 5050 - 6060 - 7070 - 8080 - 9090 - 330

• 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.

Yang et al., 2005, GRL

y = 1.2103x - 0.1012R2 = 0.9448

-15

-12

-9

-6

-3

0

3

6

9

12

15

-15 -12 -9 -6 -3 0 3 6 9 12 15

Pm trend (mm)

Pc tr

end

(mm

y = 1.0575xR2 = 0.9962

-7

-6

-5

-4

-3

-2

-1

0

1

2

3

4

5

6

7

-7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7

Pm trend (mm)

Pc tr

end

(mm

Jul.

Jan.

Impact of Bias-Corrections on Precip TrendPm & Pc Trend Comparison, Selected Stations with Data > 25 Yrs during 1973-04

Yang et al., 2005, GRL

Impact of Bias-Corrections on

Northern Hydrology:

CLM3 simulations with/without P

corrections, 1973-045~25%

Auto gauges and snow depth sensor

Challenges for auto QA/QC

0

2

4

6

8

10

12

14

16

18

20

1200

1300

1400

1500

1600

1700

1800

1900

2000

2100

2200

2300 0 10

020

030

040

050

060

070

080

090

010

0011

0012

0013

0014

0015

0016

0017

0018

0019

0020

0021

0022

0023

00

Time (15 min intervals)

Acc

umul

ated

Pre

cipi

tatio

n (m

m) a

nd 2

m w

ind

spee

d (m

/s)

-20

-15

-10

-5

0

5

Air

Tem

pera

ture

(deg

.C)Ws2 (m/s)

Unsh. Belfort (mm total)

Alter Belfort (mm total)

Nipher Belfort (mm total)

DFIR Belfort (mm total)

Ta (deg.C)

WMO Study: Timing and catch differences of Belfort Gauges at Kortright, Ontario Feb. 19-20/1988

Automation of precipitation measurementsAutomation of precipitation measurements

Need for adjustmentsNeed for adjustments

0

20

40

60

80

100

120

adjusted Nipher Belfort Belfort Belfort Belfort

DFIR Nipher unshielded unshielded Alter shielded Nipher shielded DFIR shielded

Types of Precipitation Gauges & Shields

Prec

ipita

tion

Am

ount

(% o

f adj

uste

d D

FIR)

all types(R, S, X) snow only

Mean annual accumulated winter precipitation > 3.0 mm, of different gauge types and shielding as a percentage of DFIR (adjusted for catch deficiency) at the Canadian Evaluation Station at Kortright Centre, Ontario from 1987-1991.

WMO result: Geonor vs. DFIR Jokioinen, Finland, 12h data, 12/1988-4/1993

-5

0

5

10

15

20

25

30

-5 0 5 10 15 20 25 30

Geonor + DFIR (mm)

Geo

nor (

mm

)

rainmixedsnow

Catch-wind relation, 12h snowfall, DFIR > 3mm, Jokioinen

y = -7.5744x + 93.485R2 = 0.4693

0

20

40

60

80

100

120

0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0

Wind speed at 3 m (m/s)

Ratio

of G

eono

r to

Geo

nor-DFI

R (%

)

Oct Feb Jun Oct Feb Jun

Dep

th (c

m)

0

10

20

30

40 Test Auto A (SR50)Test Auto BTest Auto CTest Auto DTest Auto ETest Auto FStation AutoStation Manual (ruler)

Snow Depth Spatial Variability and Fixed-Point Measurements

Edmonton International Airportopen landscape

see a high degree of spatial variability even over a short distance (3 to 300m)

of six temporary and a fixed station SR50, and manual ruler measurements, none are statistically similar to each other

Challenges:to provide the best quality measurements to

the research communityfor the research community to recognize these

issues when using the data (e.g. comparisons with spaceborne data)

Arctic Precipitation Issues

• Operational networks – our knowledge base – Decline of the networks in the northern regions, including

Siberia, Alaska and N. Canada– Few stations in the mountain regions – How to sustain and improve the operational networks

• Data quality and compatibility across national boundaries– Large biases in gauge measurements of solid precipitation– Incompatibility of precipitation data due to difference in

instruments and methods of data processing – Difficulties to determine precipitation changes in the arctic

regions

• Validation of precip data, including satellite and reanalysis products and fused products at high latitudes.

Measuring freshly fallen snowfall with Snow Measuring freshly fallen snowfall with Snow BoardsBoards

Original

Weaverboard 2000

for use as an Observer’s aid

10cm snowfall is 10mm precipitation

Russian Meteorological

stations

MSC Networks – precip/Snow Cover