Preserving Continuity of Long-Term Daily Maximum and Minimum Temperature Observations with Automation of Reference Climate Stations using Overlapping Data and Meteorological Conditions

Figures & data

Fig. 1 Locations of the 22 pairs of stations where overlapping records were assessed in this study.

Table 1. Metadata for the 22 pairs of stations. The “*” indicates automated stations with daily temperatures modified to match the observing window at staffed stations; “**” indicates that only the daily maximum temperatures were modified for the observing time; “***” indicates that there is only 1.1 yr of overlapping data for maximum temperature and 0.5 yr of overlapping data for minimum temperature at Moose Jaw. plt, mlt, clt, elt, nlt specify Pacific, Mountain, Central, Eastern, and Newfoundland Local Time. Staffed station instruments are liquid-in-glass thermometers or thermistors with electronic Remote Temperature Indicators (RTI).

Station Name	Province/Territory	Daily Observing Times		Separation Distance (km)	Elevation Difference (m)	Staffed Station Instruments	Overlap (years)
		Manual	Automated
Princeton	BC	7, 16 plt	22 plt*	0	0.0	RTI	10
Vernon	BC	7, 16 plt	22 plt*	0	0.2	Liquid-in-Glass	4.3
Creston	BC	7, 16 plt	22 plt*	2.2	48.8	Liquid-in-Glass	10
Dawson	YT	5, 16 plt	22 plt*	1.9	0.3	Liquid-in-Glass	10
Paulatuk	NT	8, 17 mlt	23 mlt*	0	1.7	RTI	10
Tuktoyaktuk	NT	7, 16 mlt	23 mlt**	0.7	0.0	RTI	10
Lupin	NU	6, 18 mlt	23 mlt*	0.8	2.1	Liquid-in-Glass	10
Holman	NT	8, 17 mlt	23 mlt*	0	5.8	RTI	10
Sachs Harbour	NT	8, 17 mlt	23 mlt*	1.9	1.5	Liquid-in-Glass	10
Edmonton	AB	23 mlt	23 mlt	0	0.4	RTI	10
Lacombe	AB	7, 16 mlt	23 mlt*	3.7	12.7	Liquid-in-Glass	0.9
Beaverlodge	AB	7, 16 mlt	23 mlt	2.1	0.1	RTI	4.6
Moose Jaw	SK	0 clt	0 clt	0	0.3	RTI	***
Regina	SK	0 clt	0 clt	0	0.0	RTI	10
La Ronge	SK	0 clt	0 clt	0	0.0	RTI	10
Sprague	MB	7, 16 clt	0 clt*	1.9	0.4	Liquid-in-Glass	10
Dauphin	MB	0 clt	0 clt	0	0.0	RTI	10
Gillam	MB	0 clt	0 clt	0	0.2	RTI	10
Thunder Bay	ON	1 elt	1 elt	0	0.4	RTI	0.9
Kapuskasing	ON	7, 16 elt	1 elt*	0	0.1	RTI	1.4
Ottawa	ON	7, 16 elt	1 elt*	0	0.0	RTI	10
Gander	NL	2 nlt	2 nlt	0	0.2	RTI	7.5

Fig. 2 Differences between (a) manual and original automated daily maximum temperatures (Tx); (b) manual and original automated daily minimum temperatures (Tn); (c) manual and modified automated daily maximum temperatures; and (d) manual and modified automated daily minimum temperatures at Dawson from 1995 to 2005. The modified automated daily data are the data modified to diminish the effect of different observing times.

Fig. 3 Examples of biases for (a) daily maximum temperature at Regina; (b) daily maximum temperature at La Ronge; (c) daily minimum temperature at La Ronge; and (d) daily maximum temperature at Tuktoyaktuk. The left panels show the average two-year seasonal-by-wind biases for specific wind categories and the average seasonal biases. The right panels indicate the count of events for the seasonal-by-wind bias and seasonal bias.

Fig. 4 Box plots showing the 5th, 25th, 50th, 75th, and 95th percentiles of the differences (°C in absolute value) between unadjusted and adjusted observations from automated and staffed stations of (a) daily maximum and (b) daily minimum temperatures at Sprague; the abbreviations along the horizontal axis are explained in Section 4.b.

Table 2. Mean Absolute Difference (MAD) between the unadjusted and adjusted maximum and minimum temperatures from automated and staffed stations. The daily temperatures from automated stations 1–13 were modified for the observing time, and the daily temperatures from automated stations 14–23 were not modified for the observing time. Columns 5 to 8 have no values if the overlapping period is shorter than two and/or five years, respectively; columns 9 and 10 have no values if there is no statistically significant shift (at the 0.05 level) detected at the joining date by statistical methods (see Section 6.a). The “*” indicates that only daily maximum temperatures were modified for the observing time for Tuktoyaktuk. The “**” indicates that no wind categorization was applied to either Edmonton or Dauphin maximum or minimum temperatures or to the Gillam minimum temperature: the seasonal bias is substituted for the seasonal-by-wind bias. The results in bold (column 2) give the adjustment for the observing time; the results in italics (columns 3 and 4) give the series for the full period; the results in bold italic (columns 5-8) give the results for the trimmed series.

No.	Location	Prov./Territ.	Maximum Temperature (°C)
			1 Unadj	2 ModObsTfu	3 SWB2fu	4 SWB5fu	5 SWB2tr	6 SWB5tr	7 SB2tr	8 SB5tr	9 MRadj	10 QMadj
1	Princeton	BC	0.7	0.5	0.5	0.5	0.5	0.6	0.5	0.6	0.7	0.8
2	Vernon	BC	0.9	0.8	0.8	0.8	0.9		0.9
3	Creston	BC	0.8	0.8	0.7	0.7	0.7	0.7	0.7	0.7
4	Dawson	YT	1.2	0.8	0.8	0.7	0.8	0.8	0.8	0.8
5	Paulatuk	NT	1.2	0.6	0.7	0.7	0.8	0.7	0.8	0.8
6	Tuktoyaktuk*	NT	1.1	0.5	0.5	0.5	0.5	0.5	0.5	0.6
7	Lupin	NU	1.2	0.8	0.6	0.6	0.6	0.7	0.6	0.7
8	Holman	NT	1.1	0.7	0.8	0.7	0.9	0.9	0.9	0.9
9	Sachs Harbour	NT	1.0	0.7	1.2	0.6	1.3	0.6	1.1	0.6
10	Lacombe	AB	0.8	0.6	0.6	0.6
11	Sprague	MB	1.0	0.8	0.6	0.6	0.6	0.6	0.6	0.5	0.9	0.9
12	Kapuskasing	ON	1.1	0.8	0.4	0.4					0.9	0.9
13	Ottawa	ON	0.5	0.4	0.4	0.3	0.4	0.3	0.4	0.3	0.5	0.5
14	Tuktoyaktuk*	NT
15	Edmonton**	AB	0.4		0.3	0.2	0.3	0.2	0.3	0.3	0.3	0.3
16	Beaverlodge	AB	0.5		0.5	0.5	0.5		0.5
17	Moose Jaw	SK	0.4		0.4	0.4
18	Regina	SK	0.2		0.2	0.2	0.2	0.2	0.2	0.2
19	La Ronge	SK	0.4		0.2	0.2	0.2	0.3	0.2	0.3	0.4	0.4
20	Dauphin**	MB	0.1		0.1	0.1	0.1	0.2	0.1	0.2	0.4	0.4
21	Gillam**	MB	0.4		0.3	0.3	0.3	0.3	0.3	0.3
22	Thunder Bay	ON	0.2		0.2	0.2
23	Gander	NL	0.4		0.4	0.4	0.3	0.3	0.3	0.3
			Minimum Temperature (°C)
No.	Location	Prov./Territ.	1 Unadj	2 ModObsTfu	3 SWB2fu	4 SWB5fu	5 SWB2tr	6 SWB5tr	7 SB2tr	8 SB5tr	9 MRadj	10 QMadj
1	Princeton	BC	0.7	0.4	0.3	0.3	0.3	0.3	0.3	0.3	0.7	0.7
2	Vernon	BC	2.6	2.6	1.2	1.1	1.2		1.2		2.0	2.2
3	Creston	BC	1.1	1.1	1.0	1.0	1.0	1.0	1.0	1.0	1.0	1.0
4	Dawson	YT	1.0	0.7	0.5	0.5	0.5	0.5	0.5	0.5
5	Paulatuk	NT	1.2	0.7	0.6	0.6	0.7	0.7	0.6	0.7
6	Tuktoyaktuk*	NT
7	Lupin	NU	1.3	0.9	0.7	0.7	0.7	0.8	0.7	0.8
8	Holman	NT	1.0	0.7	0.8	0.7	0.8	0.9	0.8	0.9
9	Sachs Harbour	NT	1.2	0.8	0.9	0.8	0.9	0.8	0.9	0.8
10	Lacombe	AB	1.2	0.8	0.8	0.8					1.4	1.4
11	Sprague	MB	1.5	0.9	0.9	0.9	1.0	0.8	1.0	0.8	1.4	1.5
12	Kapuskasing	ON	1.6	0.9	0.4	0.4					1.9	1.9
13	Ottawa	ON	0.9	0.3	0.3	0.3	0.3	0.4	0.3	0.4	1.0	1.0
14	Tuktoyaktuk*	NT	0.5		0.5	0.5	0.5	0.6	0.5	0.6
15	Edmonton**	AB	0.2		0.2	0.2	0.2	0.2	0.2	0.2
16	Beaverlodge	AB	0.4		0.4	0.4	0.4		0.4		0.5	0.5
17	Moose Jaw	SK	0.6		0.6	0.6					1.0	0.8
18	Regina	SK	0.2		0.2	0.2	0.2	0.2	0.2	0.2	0.7	0.8
19	La Ronge	SK	0.3		0.3	0.2	0.3	0.3	0.3	0.3
20	Dauphin**	MB	0.2		0.2	0.2	0.2	0.2	0.2	0.2	0.5	0.5
21	Gillam**	MB	0.4		0.4	0.4	0.4	0.5	0.4	0.4
22	Thunder Bay	ON	0.3		0.2	0.2
23	Gander	NL	0.6		0.6	0.5	0.3	0.3	0.3	0.2

Table 3. MADs between conventional and automated stations for days adjusted by largest wind biases at select stations with well-defined wind dependence.

Location	Largest Spring Bias Based on 2-year Overlap (°C)	Spring Wind Category for Maximum Temperatures	Count (days)	MAD (°C)
				ModOTfu	SB2fu	SWB2fu
Creston	0.9	strong	83	0.94	0.69	0.54
Paulatuk	−1.5	very light	137	1.51	1.19	1.16
Tuktoyaktuk	−1.6	very light	103	1.45	0.91	0.71
Holman	−1.0	very light	194	1.36	1.52	1.73

Fig. 5 Differences in daily maximum temperature between the conventional station and automated station at Tuktoyaktuk: (a) between original daily values (blue) and after modification for the observing time at the automated station (red) and (b) after modification for the observing time at the automated station (red); as well as after it was adjusted for seasonality (purple) and both seasonality and wind dependence (green).