Landscape air temperature thresholds (TA) and percent misclassified precipitation (error) for 12 years of meteorological observations from 40 stations across the Scandinavian Peninsula were derived and compared using both manual and geographic information system (GIS) landscape classification methods. Dew-point, wet-bulb, and wet bulb 0.5 were also tested. Both classification methods used the following west to east landscape categories: windward (WW) ocean, coast, fjord, hill, and mountain in Norway; and leeward (LW) mountain, hill, rolling terrain, and coast in Sweden. GIS landscape classification has the advantages of automating the classification process and increasing objectivity. The GIS classification was based on station location (LW or WW) relative to the Scandinavian mountain range, and the % water or range of elevation change within 15 km. The GIS and manual method had the same TA for 20 stations, and similar total reduction in error (2.29 to 2.17% respectively) when compared to country TA. Therefore, automated GIS landscape classification can be used to decrease error from common country or global scale TA. Wet-bulb temperature thresholds for GIS landscapes resulted in a greater reduction in error (8.26%) compared to air (2.29%), and dew-point (−16.67%) thresholds. However, finding stations reporting relative humidity or wet-bulb temperature may limit its widespread use.

INTRODUCTION

Precipitation falling as rain or snow can affect many different aspects of life such as transportation safety on roads, strength or thickness of sea ice (Lundberg & Feiccabrino 2009), runoff in rivers, animal migration patterns (Stenseth et al. 2002), or water storage in reservoirs for electricity production, drinking, or recreation. Surface based models for these purposes use a precipitation phase determination scheme (PPDS) to assign precipitation to a liquid (rain) or solid (snow) phase. Many PPDS apply a surface air temperature threshold (TA) (e.g. USACE 1956), surface dew point temperature threshold (TD) (e.g. Marks et al. 2013), or a surface wet bulb temperature threshold (TW) (e.g. Matsuo et al. 1981).

A PPDS scheme often uses either: (1) a single temperature threshold (TRS) where all precipitation events occurring in temperatures equal to or colder than a TRS are classified as snow, while all events occurring in warmer temperatures are classified as rain; or (2) a dual temperature threshold scheme using a rain (TR) and snow (TS) temperature threshold with all precipitation occurring in temperatures equal to or colder than TS classified as snow, all events occurring in temperatures equal to and warmer than TR classified as rain, and for precipitation in temperatures between TS and TR a decreasing snow fraction (SF) from TS = 100% to TR = 0% is used to assign the precipitation phase. The advantage of a dual threshold scheme is that it can account for sleet occurring in temperatures near a TRS, however in some areas sleet is rarely reported. For PPDS purposes, the TRS or TS and TR are set to the temperature value/s resulting in the least amount of misclassified precipitation. These threshold values are often applied over large areas irrespective of terrain, ocean, or seasonal influences on different landscapes. These and other methods, along with meteorological reasoning for their use, are reviewed in Feiccabrino et al. (2015).

Surface based PPDS methods do not account for microphysical processes, e.g. evaporation, sublimation, freezing and melting (Thériault & Stewart 2010), which are very calculation intensive, and important for assigning a precipitation phase in atmospheric models. The calculations of microphysical processes require detailed lower atmospheric measurements not often brought into surface (hydrological) models (Harder & Pomeroy 2013). Instead, surface models rely on the assumption of constant atmospheric conditions, such as a set decrease in air temperature with height (e.g. the CHRM model using −7.5 °C/km lapse rate) (Fang et al. 2013).

Assuming constant atmospheric conditions over a large area with varying landscapes is not ideal when, for example, a snowpack (open water) conductively cools (warms) near surface air. The atmospheric depth to which land or ocean temperatures have an effect is generally restricted to a couple hundred meters, and this depth can increase or decrease depending on the amount of time an air-mass is over the cooling (heating) source, wind speeds, or other factors. Terrain effects, for example orographically enhanced precipitation, will increase the depth of the atmosphere required to melt falling snow (the melting layer) toward the ground due to increased heat energy required to melt the additional snow. This results in a lowering of the 0 °C isotherm and snow elevations compared to upwind locations (Minder et al. 2011). Therefore, increasing the precipitation rate would cause changes from the surface model assumption of constant atmospheric conditions by lowering cloud layers, snow elevations, and the 0 °C isotherm. The degree to which orographic enhancement occurs is dependent on the geometry of the terrain, and melting layer properties (Minder et al. 2011), for example, relative humidity (RH). On the leeside of mountains, air descends drier, and warms at or near the dry adiabatic lapse rate (9.8 °C/km) causing another change from normal atmospheric conditions. These terrain effects are known to occur on the individual to hill-slope scale.

Dai (2008), in a global study, found a warm bias of +0.7 °C for TA over ocean regions (1.9 °C) compared to land regions (1.2 °C). The warm bias over oceans is most likely caused by open water transferring heat to the near surface atmosphere. The added surface warmth increases atmospheric instability (Ólafsson & Haraldsdóttir 2003), which by definition increases the environmental lapse rate, or actual decrease in air temperature with height, compared to the model assumed constant lapse rate. This decreases the depth of an atmospheric warm layer from the expected depth, thus increasing the chance of snow not fully melting before reaching the ground at a given surface air temperature. Stewart (1992) observed that precipitation phase transition often occurred near the coast, supporting the idea of ocean heat affecting lower atmospheric air temperatures which affects the precipitation phase in near coastal locations.

Topography and warm ocean temperatures are therefore known to affect atmospheric conditions above individual locations which should result in different TA values. However, it is not known if terrain effects can be projected to larger geographical landscapes such as the leeward (LW) side of a mountain range rather than being restricted to the individual hill-slope scale. If station TA values within a geographical landscape are found to be similar, and differ from values in other landscapes, then geographical regions (countries) could be broken into sub-regions based on landscape categorization to reduce misclassified precipitation error.

This study has two main goals: (1) present a landscape classification technique using geographic information system (GIS) to (a) automate geographic landscape classification, (b) decrease the subjectivity from manual classification found in the earlier conference paper by Feiccabrino (2015), and (c) determine if changes in TA values for landscape (sub-regional) can be used to decrease error without downscaling to the individual hill-slope scale; and (2) using the same weather stations as the earlier paper, decrease the temperature resolution from 0.5 to 0.1 °C for a comparison of both (a) the misclassified precipitation percentage resulting from PPDS schemes using TA, TD, and TW and (b) the misclassified precipitation as a result of manual and automated landscape classification.

STUDY AREA

This study is conducted using 40 weather stations; four Norwegian Sea, 16 Norwegian, and 20 Swedish (Table 1); on the Scandinavian Peninsula (Figure 1). The Scandinavian Peninsula is bisected by the Scandinavian Mountain range running SSW to NNE on the Swedish–Norwegian border. Predominant atmospheric winds are from the west and much of the weather impacting the area passes over the Atlantic Ocean and Norwegian Sea before reaching the peninsula. The orientation of the mountain range has two main affects: (1) it causes orographic enhancement on the windward (WW) Norwegian side and down-sloping drying effects on the LW Swedish side of the peninsula; and (2) it can deflect storms NW along the mountains rather than passing west over the mountains.
Table 1

Norwegian Sea, Norwegian, and Swedish weather stations names, assigned station numbers, country and adjusted Dai (2008) classification with TA and percentage of misclassified precipitation as a result of the classification TA, and the total sample size (with percentage of all precipitation observations between October and May having an air temperature of − 3 to 5 °C)

Station name Station # Country class Country TA % Misc precip Dai adj. class Dai adj. TA % Misc precip Precip sample size 
Norwegian sea ocean platforms        2,368 (34%) 
 Ormen Lange Sea 1.8 13.04 Sea 1.8 13.04 93 (28%) 
 Draugen Sea 1.8 25.18 Sea 1.8 25.18 414 (31%) 
 Heidrun Sea 1.8 15.43 Sea 1.8 15.43 1,177 (35%) 
 Norne Sea 1.8 14.93 Sea 1.8 14.93 684 (35%) 
Norwegian stations (WW)        14,073 
 Buholmråsa Fyr Norway 1.1 17.68 Sea coast 1.7 16.92 264 (25%) 
 Nordøyan Fyr Norway 1.1 16.52 Sea coast 1.7 15.75 909 (38%) 
 Sklinna Fyr Norway 1.1 14.17 Sea 1.8 14.95 516 (45%) 
 Myken Norway 1.1 13.97 Sea 1.8 13.19 1,150 (38%) 
 Hjelvik-Myrbø Norway 1.1 18.60 Fjord 1.0 18.27 302 (40%) 
 Tingvoll-Hanem 10 Norway 1.1 9.13 Fjord 1.0 9.03 970 (45%) 
 Trondheim-Voll 11 Norway 1.1 13.33 Fjord 1.0 12.96 271 (48%) 
 Finnøy Hamarøy 12 Norway 1.1 8.60 Fjord 1.0 8.74 693 (43%) 
 Selbu-Stubbe 13 Norway 1.1 12.94 Inland 1.3 13.44 801 (45%) 
 Verdal-Reppe 14 Norway 1.1 12.69 Inland 1.3 12.50 1,593 (42%) 
 Høylandet-Drageidet 15 Norway 1.1 17.26 Inland 1.3 18.37 815 (51%) 
 Bardufoss 16 Norway 1.1 11.90 Inland 1.3 12.21 1,929 (46%) 
 Råros Lufthavn 17 Norway 1.1 15.84 Inland 1.3 15.03 623 (39%) 
 Nordli – Holand 18 Norway 1.1 10.79 Inland 1.3 10.45 1,771 (48%) 
 Fiplingvatn 19 Norway 1.1 15.49 Inland 1.3 15.15 1,153 (56%) 
 Saltdal 20 Norway 1.1 11.67 Inland 1.3 8.94 331 (38%) 
Swedish stations (LW)        38,050 
 Korsvattnet 21 Sweden 1.3 7.28 Inland 1.3 7.28 3,561 (36%) 
 Gäddede 22 Sweden 1.3 15.25 Inland 1.3 15.25 581 (35%) 
 Gielas 23 Sweden 1.3 8.25 Inland 1.3 8.25 2,326 (32%) 
 Mierkenis 24 Sweden 1.3 9.57 Inland 1.3 9.57 2,756 (29%) 
 Hemavan 25 Sweden 1.3 10.88 Inland 1.3 10.88 762 (36%) 
 Föllinge 26 Sweden 1.3 9.72 Inland 1.3 9.72 1,784 (31%) 
 Hallhåxåsen 27 Sweden 1.3 8.00 Inland 1.3 8.00 1,891 (30%) 
 Hoting 28 Sweden 1.3 7.06 Inland 1.3 7.06 1,959 (32%) 
 Gubbhögen 29 Sweden 1.3 8.80 Inland 1.3 8.80 1,992 (31%) 
 Vilhelmina 30 Sweden 1.3 10.47 Inland 1.3 10.47 1,780 (28%) 
 Torpshammar 31 Sweden 1.3 10.85 Inland 1.3 10.85 1,796 (35%) 
 Krångede 32 Sweden 1.3 10.97 Inland 1.3 10.97 2,115 (33%) 
 Hemling 33 Sweden 1.3 9.04 Inland 1.3 9.04 2,118 (35%) 
 Fredrika 34 Sweden 1.3 6.78 Inland 1.3 6.78 2,055 (31%) 
 Åsele 35 Sweden 1.3 8.44 Inland 1.3 8.44 1,924 (31%) 
 Kuggören 36 Sweden 1.3 11.15 Coast 1.1 11.37 1,860 (44%) 
 Brämön 37 Sweden 1.3 10.90 Coast 1.1 11.51 1,945 (44%) 
 Lungö 38 Sweden 1.3 12.85 Coast 1.1 11.66 1,950 (43%) 
 Järnäsklubb 39 Sweden 1.3 11.46 Coast 1.1 11.22 2,063 (42%) 
 Holmön 40 Sweden 1.3 12.08 Coast 1.1 9.54 832 (42%) 
Station name Station # Country class Country TA % Misc precip Dai adj. class Dai adj. TA % Misc precip Precip sample size 
Norwegian sea ocean platforms        2,368 (34%) 
 Ormen Lange Sea 1.8 13.04 Sea 1.8 13.04 93 (28%) 
 Draugen Sea 1.8 25.18 Sea 1.8 25.18 414 (31%) 
 Heidrun Sea 1.8 15.43 Sea 1.8 15.43 1,177 (35%) 
 Norne Sea 1.8 14.93 Sea 1.8 14.93 684 (35%) 
Norwegian stations (WW)        14,073 
 Buholmråsa Fyr Norway 1.1 17.68 Sea coast 1.7 16.92 264 (25%) 
 Nordøyan Fyr Norway 1.1 16.52 Sea coast 1.7 15.75 909 (38%) 
 Sklinna Fyr Norway 1.1 14.17 Sea 1.8 14.95 516 (45%) 
 Myken Norway 1.1 13.97 Sea 1.8 13.19 1,150 (38%) 
 Hjelvik-Myrbø Norway 1.1 18.60 Fjord 1.0 18.27 302 (40%) 
 Tingvoll-Hanem 10 Norway 1.1 9.13 Fjord 1.0 9.03 970 (45%) 
 Trondheim-Voll 11 Norway 1.1 13.33 Fjord 1.0 12.96 271 (48%) 
 Finnøy Hamarøy 12 Norway 1.1 8.60 Fjord 1.0 8.74 693 (43%) 
 Selbu-Stubbe 13 Norway 1.1 12.94 Inland 1.3 13.44 801 (45%) 
 Verdal-Reppe 14 Norway 1.1 12.69 Inland 1.3 12.50 1,593 (42%) 
 Høylandet-Drageidet 15 Norway 1.1 17.26 Inland 1.3 18.37 815 (51%) 
 Bardufoss 16 Norway 1.1 11.90 Inland 1.3 12.21 1,929 (46%) 
 Råros Lufthavn 17 Norway 1.1 15.84 Inland 1.3 15.03 623 (39%) 
 Nordli – Holand 18 Norway 1.1 10.79 Inland 1.3 10.45 1,771 (48%) 
 Fiplingvatn 19 Norway 1.1 15.49 Inland 1.3 15.15 1,153 (56%) 
 Saltdal 20 Norway 1.1 11.67 Inland 1.3 8.94 331 (38%) 
Swedish stations (LW)        38,050 
 Korsvattnet 21 Sweden 1.3 7.28 Inland 1.3 7.28 3,561 (36%) 
 Gäddede 22 Sweden 1.3 15.25 Inland 1.3 15.25 581 (35%) 
 Gielas 23 Sweden 1.3 8.25 Inland 1.3 8.25 2,326 (32%) 
 Mierkenis 24 Sweden 1.3 9.57 Inland 1.3 9.57 2,756 (29%) 
 Hemavan 25 Sweden 1.3 10.88 Inland 1.3 10.88 762 (36%) 
 Föllinge 26 Sweden 1.3 9.72 Inland 1.3 9.72 1,784 (31%) 
 Hallhåxåsen 27 Sweden 1.3 8.00 Inland 1.3 8.00 1,891 (30%) 
 Hoting 28 Sweden 1.3 7.06 Inland 1.3 7.06 1,959 (32%) 
 Gubbhögen 29 Sweden 1.3 8.80 Inland 1.3 8.80 1,992 (31%) 
 Vilhelmina 30 Sweden 1.3 10.47 Inland 1.3 10.47 1,780 (28%) 
 Torpshammar 31 Sweden 1.3 10.85 Inland 1.3 10.85 1,796 (35%) 
 Krångede 32 Sweden 1.3 10.97 Inland 1.3 10.97 2,115 (33%) 
 Hemling 33 Sweden 1.3 9.04 Inland 1.3 9.04 2,118 (35%) 
 Fredrika 34 Sweden 1.3 6.78 Inland 1.3 6.78 2,055 (31%) 
 Åsele 35 Sweden 1.3 8.44 Inland 1.3 8.44 1,924 (31%) 
 Kuggören 36 Sweden 1.3 11.15 Coast 1.1 11.37 1,860 (44%) 
 Brämön 37 Sweden 1.3 10.90 Coast 1.1 11.51 1,945 (44%) 
 Lungö 38 Sweden 1.3 12.85 Coast 1.1 11.66 1,950 (43%) 
 Järnäsklubb 39 Sweden 1.3 11.46 Coast 1.1 11.22 2,063 (42%) 
 Holmön 40 Sweden 1.3 12.08 Coast 1.1 9.54 832 (42%) 
Figure 1

Study area with station locations and prevailing atmospheric winds.

Figure 1

Study area with station locations and prevailing atmospheric winds.

The weather data are free and publicly available from the Swedish Meteorological and Hydrological Institute (SMHI 2016) and the Norwegian Meteorological Institute (NMI 2016) websites. A description of the data and further station information are found in Feiccabrino (2015). In Feiccabrino (2015), the weather stations were manually assigned to the geographic landscapes; WW ocean, WW coast, WW fjords, WW hills, and WW mountain (valleys) in Norway, and LW mountain (valleys), LW hills, LW rolling, and LW Coast in Sweden. The 2015 study found that using different TA values for geographic landscapes is a viable option for decreasing misclassified precipitation error in regional studies. It also found TW to be 0.5 °C in all landscapes (at a 0.5 °C scale), which resulted in less misclassified precipitation than TD and TA for eight of nine landscapes. However, dew point, wet-bulb, and RH measurements are much rarer than surface air temperature at hydrological and meteorological sites (e.g. Kane & Stuefer 2015). Therefore it is important to find techniques to decrease misclassified precipitation using surface air temperature, a commonly reported value.

The Scandinavian weather stations have 25–56% winter (October–May) precipitation occurring in surface air temperatures of −3 to 5 °C (Figure 2 and Table 1), a temperature range over which snow and/or rain can occur. Within this air temperature range, there is a high percentage of misclassified precipitation (7–25% in Table 1) for TA values applied over a regional (country) scale without regard for topography or other local effects.
Figure 2

Charts from left to right depict; % of winter precipitation between −3 and 5 °C, % misclassified precipitation for observations between −3 and 5 °C using country (Norwegian Sea, Norway, and Sweden) TA values, and % reduction in error when using a modified Dai (2008) landscape classification method for assigned TA.

Figure 2

Charts from left to right depict; % of winter precipitation between −3 and 5 °C, % misclassified precipitation for observations between −3 and 5 °C using country (Norwegian Sea, Norway, and Sweden) TA values, and % reduction in error when using a modified Dai (2008) landscape classification method for assigned TA.

An initial test applying slight modifications to Dai's (2008) method with different landscape TA for WW ocean, WW coast, WW fjords, inland, and LW coast stations was found to decrease error from country values (Table 1 and Figure 2). However, better landscape definitions could be used to further decrease misclassified precipitation, especially for inland areas.

METHODS

Geographic coordinates and associated tabular data for each meteorological station were imported into ArcGIS (version 10.2.1) and initially displayed using the WGS 1984 geographic coordinate system. In order to perform spatial analysis, the data were reprojected in a UTM projected coordinate system (ETRS 1989 – TM32). Elevation/hillshade and slope base maps were created using a European digital elevation model (GTOP30) with a 30-arc second (ca. 1 km) resolution. This layer was produced by the USGS EROS Data Center (USGS 1996) and was obtained through ArcGIS online. ArcGIS was used to classify the stations based on landscape features and also to produce a series of maps that highlight the spatial distribution of the results.

The classification of meteorological stations using ArcGIS involved three sets of landscape-based parameters. The first parameter in the classification was used to approximate the water (ocean) or land influence on a weather station. Ice-free water in the winter would normally be a surface based heat source that affects temperature thresholds. Buffer zones of 15 km were created surrounding each station and then intersected with the land outline for Norway and Sweden. The area of land contained within each of the intersected buffer zones was quantified using the calculate geometry function. The following categories were established based on these results: <10% land = ocean; 10–60% land = coast; 60–90% land = fjord (inlet); >90% land = land (Table 2). These numbers were chosen for their ability to separate the near coastal stations into distinct landscapes with different relative influences of land and ocean. Some smaller islands off the main shore were categorized as ocean stations which could be separated out if another landscape category was added, or the 10% land requirement for coast was decreased.

Table 2

GIS landscape classification method using a 15 km buffer

Step 1% land Step 2 elevation range Step 3 WW vs LW 
<10% land Ocean/water Ocean (WW and LW) 
10% < land <60% Coast (WW and LW) 
60% < land <90% Fjord (inlet) (WW and LW) 
90% < land 0–499 m Rolling (WW and LW) 
500–999 m Hill (WW and LW) 
1,000–1,500 m Mountain (WW and LW) 
Step 1% land Step 2 elevation range Step 3 WW vs LW 
<10% land Ocean/water Ocean (WW and LW) 
10% < land <60% Coast (WW and LW) 
60% < land <90% Fjord (inlet) (WW and LW) 
90% < land 0–499 m Rolling (WW and LW) 
500–999 m Hill (WW and LW) 
1,000–1,500 m Mountain (WW and LW) 

The land sites were further categorized based on the range of elevation that fell within the 15 km buffer zones. This relief value should indicate the amount of orographic enhancement (Minder et al. 2011) or down-sloping (LW) influence on lower atmospheric conditions over an area, therefore providing a reasonable reflection of terrain affects at a higher spatial resolution than the regional scale without having to consider wind datum or more detailed topographic data. The process of calculating the relief value involved extracting the digital elevation model (DEM) that fell within the buffer zones and then converting the extracted DEM into a vector (polygon) feature class. The polygon data were then summarized by station number under the fields of maximum and minimum values. A new table was created during the summarization and the range (max–min elevations) was then calculated for each site buffer using the field calculator. Finally, the new table with range was joined to the original buffer layer. Now each site could be categorized by the range of elevation within each buffer zone. The following classifications were made based on these data: range of 1,500–1,000 m = mountain; range of 1,000–500 m = hill; range of 500–0 m = rolling (Table 2). The values for these classifications were chosen to be both objective with a range of 500 m each, and representative of the amount of relief in the area. The values can be expanded or retracted depending on the topography of a study area, or have extra categories added.

The third and final parameter used in the classification was WW versus LW based on the prevailing direction of westerly flow and the location of the Scandinavian Mountains.

For the TA, TD, and TW analysis, dew point temperatures were calculated using Equation (1), a common formula: 
formula
1
Since atmospheric pressure was not reported at the Swedish stations, TW was calculated using an empirical formula (Stull 2011) that does not require air pressure: 
formula
2
For all analysis, the % misclassified precipitation (% error) is: 
formula
3
where PS is a snow precipitation event, PR is a rain precipitation event, and PTotal is the total of all rain and snow precipitation events. TRS is the rain snow temperature threshold value that results in the least misclassified precipitation. Precipitation events rather than mass are used in this analysis due to the Norwegian datasets not containing consistently reported precipitation mass values.
Reduction in error is calculated as: 
formula
4
where the % error from country TA (Table 1) is used as the reference value.

Datasets for precipitation events with surface air temperatures of −3 to 5 °C were built for each country, landscape, and individual weather station to determine TA, TD, and TW values.

RESULTS AND DISCUSSION

The GIS classification method produced a product similar to that of Feiccabrino's (2015) manual classification (Table 3 and Figure 3). The GIS method resulted in an average reduction in error of 2.29% compared to 2.17% from the manual method. The GIS landscape classification performed better than the manual classification for the Swedish stations (2.42 to 1.24% respectively) but resulted in less error reduction for the Norwegian stations (2.14 to 3.33% respectively). When considering the 18 hill and mountain stations in which there was a difference in TA between the methods, the GIS and manual landscape classifications resulted in similar reductions in error (2.48 to 2.31% respectively). The largest improvement came from reclassification of the manual LW mountain stations having 0.39% average reduction in error compared to 4.70% when reclassified as LW mountain and LW hill stations in the GIS method.
Table 3

North Sea, Norwegian, and Swedish weather station assigned station numbers, manual and GIS landscape classification with TA, percentage of misclassified precipitation for TA, % (less) reduction in error compared to country classification % error (Table 1), with similar analysis for TD and TW for the GIS landscape classification

Station # (Figure 1Manual landscape Landscape TA °C Landscape % error % Less error GIS landscape Landscape TA °C Landscape % error % Less error Landscape TD °C Landscape % error % Less error Landscape TW °C Landscape % error % Less error TW 0.5 °C % less error Station TA °C 
Norwegian stations WW 
 1 WW OCEAN 1.8 13.04 0.00 WW OCEAN 1.8 13.04 0.00 −0.4 18.48 −41.72 0.5 11.96 8.28 8.28 1.5 
 2 WW OCEAN 1.8 25.18 0.00 WW OCEAN 1.8 25.18 0.00 −0.4 19.85 21.17 0.5 21.94 12.87 12.87 2.2 
 3 WW OCEAN 1.8 15.43 0.00 WW OCEAN 1.8 15.43 0.00 −0.4 19.60 −27.03 0.5 16.51 −7.00 −7.00 1.8 
 4 WW OCEAN 1.8 14.93 0.00 WW OCEAN 1.8 14.93 0.00 −0.4 17.42 −16.68 0.5 13.61 8.84 8.84 1.8 
 5 WW COAST 1.8 17.30 2.15 WW COAST 1.7 16.92 4.30 −0.5 25.67 −45.19 0.3 17.49 1.07 1.07 1.5 
 6 WW COAST 1.8 15.64 5.33 WW COAST 1.7 15.75 4.66 −0.5 20.46 −23.85 0.3 15.72 4.84 4.84 1.9 
 7 WW COAST 1.8 14.95 −5.50 WW OCEAN 1.8 14.95 −5.50 −0.4 11.84 16.44 0.5 11.46 19.12 19.12 1.0 
 8 WW COAST 1.8 13.19 5.58 WW OCEAN 1.8 13.19 5.58 −0.4 16.28 −16.54 0.5 13.49 3.44 3.44 1.5 
 9  WW FJORD 1.0 18.27 1.77  WW FJORD 1.0 18.27 1.77 0.3 22.92 −23.23 0.4 17.61 5.32 −0.91 1.6 
 10 WW FJORD 1.0 9.03 1.10 WW FJORD 1.0 9.03 1.10 0.3 13.00 −42.39 0.4 8.67 5.04 −1.20 1.2 
 11 WW FJORD 1.0 12.96 2.78 WW FJORD 1.0 12.96 2.78 0.3 18.89 −41.71 0.4 15.00 −12.53 −12.53 0.7 
 12 WW FJORD 1.0 8.74 −1.63 WW FJORD 1.0 8.74 −1.63 0.3 12.43 −44.53 0.4 10.26 −19.30 −13.37 1.2 
 13 WW HILL 1.1 12.94 0.00 WW HILL 1.3 13.44 −3.86 0.0 21.31 −64.68 0.4 15.31 −18.32 −24.11 1.1 
 14 WW HILL 1.1 12.69 0.00 WW HILL 1.3 12.50 1.50 0.0 13.63 −7.41 0.4 10.96 13.63 15.60 1.5 
 15 WW HILL 1.1 17.26 0.00 WW HILL 1.3 18.37 −6.43 0.0 15.30 11.36 0.4 14.00 18.89 9.91 0.6 
 16 WW HILL 1.1 11.90 0.00 WW MTN 1.2 12.06 −1.34 −0.3 14.21 −19.41 0.4 10.93 8.15 5.71 1.1 
 17 WW MTN 1.4 14.71 7.13 WW HILL 1.3 15.03 5.11 0.0 13.10 17.30 0.4 12.38 21.84 20.33 1.7 
 18 WW MTN 1.4 10.45 3.15 WW HILL 1.3 10.45 3.15 0.0 11.86 −9.92 0.4 9.60 11.03 9.18 1.3 
 19 WW MTN 1.4 15.06 2.78 WW MTN 1.2 15.15 2.19 −0.3 19.30 −24.60 0.4 16.35 −5.55 0.39 1.4 
 20 WW MTN 1.4 8.33 28.62 WW MTN 1.2 9.24 20.82 −0.3 15.00 −28.53 0.4 7.58 35.05 20.82 1.4 
Swedish stations LW 
 21 LW MTN 1.5 7.20 1.10 LW HILL 1.6 7.25 0.41 0.2 8.47 −16.35 0.5 7.04 3.30 3.30 1.4 
 22 LW MTN 1.5 16.98 −11.34 LW HILL 1.6 18.20 −19.34 0.2 16.64 −9.11 0.5 14.21 6.82 6.82 0.9 
 23 LW MTN 1.5 7.04 14.67 LW HILL 1.6 6.70 18.79 0.2 10.05 −21.82 0.5 7.11 13.82 13.82 1.6 
 24 LW MTN 1.5 8.41 12.12 LW HILL 1.6 8.23 14.00 0.2 10.86 −13.48 0.5 6.48 32.29 32.29 1.7 
 25 LW MTN 1.5 12.47 −14.61 LW MTN 1.0 9.83 9.65 No RH reported       1.0 
 26 LW HILL 1.2 9.38 3.50 LW ROLLING 1.3 9.72 0.00 0.3 9.27 4.63 0.5 7.53 22.53 22.53 1.2 
 27 LW HILL 1.2 8.48 −6.00 LW ROLLING 1.3 8.00 0.00 0.3 10.93 −36.63 0.5 8.65 −8.12 −8.12 1.4 
 28 LW HILL 1.2 7.37 −4.39 LW ROLLING 1.3 7.06 0.00 0.3 10.13 −43.48 0.5 6.85 2.97 2.97 1.3 
 29 LW HILL 1.2 8.65 1.70 LW ROLLING 1.3 8.80 0.00 0.3 9.96 −13.18 0.5 7.45 15.34 15.34 1.1 
 30 LW HILL 1.2 10.14 3.15 LW ROLLING 1.3 10.47 0.00 0.3 9.98 4.68 0.5 8.62 17.67 17.67 1.0 
 31 LW ROLLING 1.3 10.85 0.00 LW ROLLING 1.3 10.85 0.00 0.3 12.89 −18.80 0.5 10.13 6.64 6.64 1.3 
 32 LW ROLLING 1.3 10.97 0.00 LW ROLLING 1.3 10.97 0.00 0.3 12.91 −17.68 0.5 9.97 9.12 9.12 1.5 
 33 LW ROLLING 1.3 9.04 0.00 LW ROLLING 1.3 9.04 0.00 0.3 8.81 2.54 0.5 7.62 15.71 15.71 1.1 
 34 LW ROLLING 1.3 6.78 0.00 LW ROLLING 1.3 6.78 0.00 0.3 7.12 −5.01 0.5 5.21 23.16 23.16 1.0 
 35 LW ROLLING 1.3 8.44 0.00 LW ROLLING 1.3 8.44 0.00 0.3 8.92 −5.69 0.5 7.09 16.00 16.00 1.1 
 36 LW COAST 1.1 11.37 −1.97 LW COAST 1.1 11.37 −1.97 0.3 13.38 −20.00 0.6 11.52 −3.32 −2.06 1.4 
 37 LW COAST 1.1 11.51 −5.60 LW COAST 1.1 11.51 −5.60 0.3 12.48 −14.50 0.6 10.83 0.64 −5.50 1.5 
 38 LW COAST 1.1 11.66 9.26 LW COAST 1.1 11.66 9.26 0.3 13.00 −1.17 0.6 12.18 5.21 4.44 0.8 
 39 LW COAST 1.1 11.22 2.09 LW COAST 1.1 11.22 2.09 0.3 11.62 −1.40 0.6 10.94 4.54 8.73 1.1 
 40 LW COAST 1.1 9.54 21.03 LW COAST 1.1 9.54 21.03 0.3 13.31 −10.10 0.6 10.52 12.91 12.91 1.0 
Station # (Figure 1Manual landscape Landscape TA °C Landscape % error % Less error GIS landscape Landscape TA °C Landscape % error % Less error Landscape TD °C Landscape % error % Less error Landscape TW °C Landscape % error % Less error TW 0.5 °C % less error Station TA °C 
Norwegian stations WW 
 1 WW OCEAN 1.8 13.04 0.00 WW OCEAN 1.8 13.04 0.00 −0.4 18.48 −41.72 0.5 11.96 8.28 8.28 1.5 
 2 WW OCEAN 1.8 25.18 0.00 WW OCEAN 1.8 25.18 0.00 −0.4 19.85 21.17 0.5 21.94 12.87 12.87 2.2 
 3 WW OCEAN 1.8 15.43 0.00 WW OCEAN 1.8 15.43 0.00 −0.4 19.60 −27.03 0.5 16.51 −7.00 −7.00 1.8 
 4 WW OCEAN 1.8 14.93 0.00 WW OCEAN 1.8 14.93 0.00 −0.4 17.42 −16.68 0.5 13.61 8.84 8.84 1.8 
 5 WW COAST 1.8 17.30 2.15 WW COAST 1.7 16.92 4.30 −0.5 25.67 −45.19 0.3 17.49 1.07 1.07 1.5 
 6 WW COAST 1.8 15.64 5.33 WW COAST 1.7 15.75 4.66 −0.5 20.46 −23.85 0.3 15.72 4.84 4.84 1.9 
 7 WW COAST 1.8 14.95 −5.50 WW OCEAN 1.8 14.95 −5.50 −0.4 11.84 16.44 0.5 11.46 19.12 19.12 1.0 
 8 WW COAST 1.8 13.19 5.58 WW OCEAN 1.8 13.19 5.58 −0.4 16.28 −16.54 0.5 13.49 3.44 3.44 1.5 
 9  WW FJORD 1.0 18.27 1.77  WW FJORD 1.0 18.27 1.77 0.3 22.92 −23.23 0.4 17.61 5.32 −0.91 1.6 
 10 WW FJORD 1.0 9.03 1.10 WW FJORD 1.0 9.03 1.10 0.3 13.00 −42.39 0.4 8.67 5.04 −1.20 1.2 
 11 WW FJORD 1.0 12.96 2.78 WW FJORD 1.0 12.96 2.78 0.3 18.89 −41.71 0.4 15.00 −12.53 −12.53 0.7 
 12 WW FJORD 1.0 8.74 −1.63 WW FJORD 1.0 8.74 −1.63 0.3 12.43 −44.53 0.4 10.26 −19.30 −13.37 1.2 
 13 WW HILL 1.1 12.94 0.00 WW HILL 1.3 13.44 −3.86 0.0 21.31 −64.68 0.4 15.31 −18.32 −24.11 1.1 
 14 WW HILL 1.1 12.69 0.00 WW HILL 1.3 12.50 1.50 0.0 13.63 −7.41 0.4 10.96 13.63 15.60 1.5 
 15 WW HILL 1.1 17.26 0.00 WW HILL 1.3 18.37 −6.43 0.0 15.30 11.36 0.4 14.00 18.89 9.91 0.6 
 16 WW HILL 1.1 11.90 0.00 WW MTN 1.2 12.06 −1.34 −0.3 14.21 −19.41 0.4 10.93 8.15 5.71 1.1 
 17 WW MTN 1.4 14.71 7.13 WW HILL 1.3 15.03 5.11 0.0 13.10 17.30 0.4 12.38 21.84 20.33 1.7 
 18 WW MTN 1.4 10.45 3.15 WW HILL 1.3 10.45 3.15 0.0 11.86 −9.92 0.4 9.60 11.03 9.18 1.3 
 19 WW MTN 1.4 15.06 2.78 WW MTN 1.2 15.15 2.19 −0.3 19.30 −24.60 0.4 16.35 −5.55 0.39 1.4 
 20 WW MTN 1.4 8.33 28.62 WW MTN 1.2 9.24 20.82 −0.3 15.00 −28.53 0.4 7.58 35.05 20.82 1.4 
Swedish stations LW 
 21 LW MTN 1.5 7.20 1.10 LW HILL 1.6 7.25 0.41 0.2 8.47 −16.35 0.5 7.04 3.30 3.30 1.4 
 22 LW MTN 1.5 16.98 −11.34 LW HILL 1.6 18.20 −19.34 0.2 16.64 −9.11 0.5 14.21 6.82 6.82 0.9 
 23 LW MTN 1.5 7.04 14.67 LW HILL 1.6 6.70 18.79 0.2 10.05 −21.82 0.5 7.11 13.82 13.82 1.6 
 24 LW MTN 1.5 8.41 12.12 LW HILL 1.6 8.23 14.00 0.2 10.86 −13.48 0.5 6.48 32.29 32.29 1.7 
 25 LW MTN 1.5 12.47 −14.61 LW MTN 1.0 9.83 9.65 No RH reported       1.0 
 26 LW HILL 1.2 9.38 3.50 LW ROLLING 1.3 9.72 0.00 0.3 9.27 4.63 0.5 7.53 22.53 22.53 1.2 
 27 LW HILL 1.2 8.48 −6.00 LW ROLLING 1.3 8.00 0.00 0.3 10.93 −36.63 0.5 8.65 −8.12 −8.12 1.4 
 28 LW HILL 1.2 7.37 −4.39 LW ROLLING 1.3 7.06 0.00 0.3 10.13 −43.48 0.5 6.85 2.97 2.97 1.3 
 29 LW HILL 1.2 8.65 1.70 LW ROLLING 1.3 8.80 0.00 0.3 9.96 −13.18 0.5 7.45 15.34 15.34 1.1 
 30 LW HILL 1.2 10.14 3.15 LW ROLLING 1.3 10.47 0.00 0.3 9.98 4.68 0.5 8.62 17.67 17.67 1.0 
 31 LW ROLLING 1.3 10.85 0.00 LW ROLLING 1.3 10.85 0.00 0.3 12.89 −18.80 0.5 10.13 6.64 6.64 1.3 
 32 LW ROLLING 1.3 10.97 0.00 LW ROLLING 1.3 10.97 0.00 0.3 12.91 −17.68 0.5 9.97 9.12 9.12 1.5 
 33 LW ROLLING 1.3 9.04 0.00 LW ROLLING 1.3 9.04 0.00 0.3 8.81 2.54 0.5 7.62 15.71 15.71 1.1 
 34 LW ROLLING 1.3 6.78 0.00 LW ROLLING 1.3 6.78 0.00 0.3 7.12 −5.01 0.5 5.21 23.16 23.16 1.0 
 35 LW ROLLING 1.3 8.44 0.00 LW ROLLING 1.3 8.44 0.00 0.3 8.92 −5.69 0.5 7.09 16.00 16.00 1.1 
 36 LW COAST 1.1 11.37 −1.97 LW COAST 1.1 11.37 −1.97 0.3 13.38 −20.00 0.6 11.52 −3.32 −2.06 1.4 
 37 LW COAST 1.1 11.51 −5.60 LW COAST 1.1 11.51 −5.60 0.3 12.48 −14.50 0.6 10.83 0.64 −5.50 1.5 
 38 LW COAST 1.1 11.66 9.26 LW COAST 1.1 11.66 9.26 0.3 13.00 −1.17 0.6 12.18 5.21 4.44 0.8 
 39 LW COAST 1.1 11.22 2.09 LW COAST 1.1 11.22 2.09 0.3 11.62 −1.40 0.6 10.94 4.54 8.73 1.1 
 40 LW COAST 1.1 9.54 21.03 LW COAST 1.1 9.54 21.03 0.3 13.31 −10.10 0.6 10.52 12.91 12.91 1.0 

Light gray shading indicates stations with a change in TA between manual and GIS classification.

Figure 3

Top images depict the classification categories on slope/relief maps. The bottom images depict and reduction in error (actual values given in Table 3 under % less error) on a topographic map.

Figure 3

Top images depict the classification categories on slope/relief maps. The bottom images depict and reduction in error (actual values given in Table 3 under % less error) on a topographic map.

The results in Figure 3 and Table 3 indicate that the proposed GIS method is able to replicate and/or improve the reduction in misclassified precipitation from manual landscape classification. This landscape classification was done to embrace terrain (Minder et al. 2011) and ocean effects (Dai 2008) in the lower atmosphere which are ignored when using a global or regional TA value, while not downscaling to the individual/hill-slope scale.

The GIS landscape classification was used to assess the use of air, wet bulb and dew-point temperatures in predicting thresholds. Comparing TA, TW, and TD (Figure 4), TD has more misclassified precipitation than TA and TW (Table 3) with most stations having increases in error compared to regional/country TA (Table 1 and Figure 4). Most study stations are near towns and villages in valleys with sub-saturated atmospheric conditions, so it is expected that the microphysical processes acting on hydrometers would be different than that of the saturated upslope mountain environment used by Marks et al. (2013) which found TD to be a better TRS than TA and TW. It is difficult to find weather data on mountains or other sparsely populated areas, much less finding stations reporting more than temperature and precipitation (Kane & Stuefer 2015), that is why improving TA is an important goal. Here, mostly valley stations were used, so this may have to be considered when choosing a TRS for mountains.
Figure 4

Misclassified precipitation for GIS landscape TA, TW, and TD values left to right respectively for each station across the top images with their respective reduction in error (actual values given in Table 3 under % less error) in the lower images.

Figure 4

Misclassified precipitation for GIS landscape TA, TW, and TD values left to right respectively for each station across the top images with their respective reduction in error (actual values given in Table 3 under % less error) in the lower images.

TA and TW (Figures 4 and 5) have similar misclassified precipitation results with the exception of LW rolling, and WW ocean landscapes where TA had the same landscape and regional values while TW resulted in decreased misclassified precipitation at many of those stations. TW had the least amount of error averaging 10.76% misclassified precipitation compared to 13.60% for TD and 11.44% for TA. TW also has the most consistent landscape TRS values ranging from 0.3 to 0.6 °C (Table 3). There is little error difference between wet-bulb temperature 0.5 °C and landscape TW (Table 3) (Figures 5 and 6). However, individual station TW (13.81%) has a much greater reduction in error than landscape TW (8.26%) (Figure 6). This indicates more improvement is possible. However, individual station thresholds are not practical when looking for a TRS method for a region with varying landscapes.
Figure 5

Average % misclassified precipitation for TA, TD, TW, and TW = 0.5 °C for stations within each landscape, and average misclassified precipitation using landscape values for the groups of Norwegian (mainland), Swedish, and all stations.

Figure 5

Average % misclassified precipitation for TA, TD, TW, and TW = 0.5 °C for stations within each landscape, and average misclassified precipitation using landscape values for the groups of Norwegian (mainland), Swedish, and all stations.

Figure 6

Comparison of errors (left to right) using TW = 0.5 °C, GIS landscape TW, and station specific TW.

Figure 6

Comparison of errors (left to right) using TW = 0.5 °C, GIS landscape TW, and station specific TW.

The landscape TA values agreed well with Dai (2008). The warmest values were over the ocean and (WW) Norwegian coast (1.8 and 1.7 °C respectively). However, one surprise was the LW hill stations which have a TA of 1.6 °C. This could be due to down-sloping effects leading to a deeper dry air layer between the cloud and ground. In a dry layer the air could be assumed to warm at between 9.8 and 7.5 °C/km (Fang et al. 2013). Depending on the atmospheric conditions, snow could fall through a couple hundred meters of dry air without completely melting (Venne et al. 1997). The coolest TA values were in the lowlands near the shores indicating that the near surface heating of the ocean does not affect surface or lower atmospheric temperatures far inland (Table 3). Interestingly, TA values for inland and mountains are warmer (1.2 and 1.3 °C with LW hills at 1.6 °C) than WW fjord and LW coastal stations (1.0 and 1.1 °C respectively). TRS value fluctuations for individual stations within hill and mountain landscapes suggest much of the differences occur on the hill-slope scale as local terrain and winds differ greatly compared to lowland flat terrain. In the winter, oceans cause relatively greater lower atmospheric instability than land due to conductive heating at the ground atmosphere interface leading to warmer TRS values over open water (Ólafsson & Haraldsdóttir 2003). Applying the same logic, one would expect snow covered ground to increase lower atmospheric stability due to cooling at the ground atmosphere interface leading to cooler TRS values over snow covered ground. However, Feiccabrino (2015) did not find any notable variation in TA values for either land with and without snow cover, or ocean stations with varying sea surface temperatures.

CONCLUSIONS

The GIS landscape classification method was both able to add objectivity to a previous, more subjective, manual classification with similar decreases in misclassified precipitation, indicating that it is a viable method and should be tested in a surface hydrological model.

The GIS landscape TA results indicate that decreasing the geographical scale from regional/country to sub-regional/landscape reduces misclassified precipitation by 2.29% of the possible 7.60% when using individual station thresholds. The landscapes show ocean stations to have the warmest TA values while low relief land stations have the coolest. Generally warmer TA values were found in mountain and hill landscapes than for the inland rolling hill or LW coast stations.

Over land, high variability in individual station TA values were found in mountain and hill landscapes. This indicates that while sub-regional TA values based on a 15 km GIS landscape identification scheme improved precipitation phase determination, there may be further benefit from a smaller scale GIS analysis modified to identify upslope, downslope, or valley areas within high relief terrain.

Ocean and land effects on TA were able to advect with prevailing atmospheric winds over coastal stations, most notably for the WW coast and WW ocean landscapes with TA = 1.7 and 1.8 °C respectively and LW rolling, LW coast landscapes with TA = 1.3 and 1.1 °C respectively. The ocean effect did not push further than 15 km inland as fjord/inlet stations had a TA value of 1.0 °C which is more representative of low relief land stations.

GIS landscape TW (8.26%) reduced misclassified precipitation from regional TA by more than three times as much as landscape TA values (2.29%). Wet-bulb thresholds are superior to TA for point locations, however many stations do not report the required information to calculate wet-bulb. Therefore, the use of TW is highly recommended if the information exists, or a surface wet-bulb layer is verified to project well through a model layer.

REFERENCES

REFERENCES
Fang
X.
Pomeroy
J. W.
Ellis
C. R.
MacDonald
M. K.
DeBeer
C. M.
Brown
T.
2013
Multi-variable evaluation of hydrological model predictions for a headwater basin in the Canadian Rocky Mountains
.
Hydrol. Earth Syst. Sci.
17
(
4
),
1635
1659
.
Feiccabrino
J.
2015
A cross sectional study of rain/snow threshold changes from the North Sea across the Scandinavian Mountains to the Bay of Bothnia
. In:
Proceedings of the 20th International Northern Research Basins Symposium and Workshop (20th NRB)
,
Kuusamo, Finland
,
16–21 August
, pp.
4
13
. .
Feiccabrino
J.
Graff
W.
Lundberg
A.
Sandström
N.
Gustafsson
D.
2015
Meteorological knowledge useful for the improvement of snow rain separation in surface based models
.
Hydrology
2
(
4
),
266
288
.
Lundberg
A.
Feiccabrino
J.
2009
Sea ice growth, modeling of precipitation phase
. In:
Proceedings of the 20th International Conference on Port and Ocean Engineering under Arctic Conditions (POAC 09)
,
Luleå, Sweden
,
9–12 June
.
Available from: www.poac.com/PapersOnline.html (accessed 23 September 2016), file name POAC09-10
.
Matsuo
T.
Sato
Y.
Sasyo
Y.
1981
Relationship between types of precipitation on the ground and surface meteorological elements
.
J. Meteorol. Soc. Jpn.
59
(
4
),
462
475
.
Minder
J. R.
Durran
D. R.
Roe
G. H.
2011
Mesoscale controls on the mountainside snow line
.
J. Atmos. Sci.
68
(
9
),
2107
2127
.
Norwegian Meteorological Institute (NMI)
2016
Meteorological Data
.
Available from: http://eklima.met.no (accessed 30 January 2016)
.
Ólafsson
H.
Haraldsdóttir
S. H.
2003
Diurnal, seasonal, and geographical variability of air temperature limits of snow and rain
. In:
Proceedings of the International Conference on Alpine Meteorology (ICAM 2003
),
Brig, Switzerland
,
19–23 May
, pp.
473
476
.
Available from: www.map.meteoswiss.ch/map-doc/icam2003/741.pdf (accessed 23 September 2016)
.
Stenseth
N. C.
Mysterud
A.
Ottersen
G.
Hurrell
J. W.
Chan
K. S.
Lima
M.
2002
Ecological effects of climate change
.
Science
5585
(
297
),
1292
1296
.
Stewart
R. E.
1992
Precipitation types in the transition region of winter storms
.
Bull. Am. Meteorol. Soc.
73
(
3
),
287
296
.
Swedish Meteorological and Hydrological Institute (SMHI)
2016
Meteorological Data
.
Available from: http://opendata-download-metobs.smhi.se/explore/?parameter=8 (accessed 30 January 2016)
.
United States Army Corps of Engineers (USACE)
1956
Snow Hydrology
.
Summary report of the snow investigations
.
North Pacific Division
,
Portland, OR
, p.
437
.
United States Geological Survey (USGS)
1996
Global 30 Arc-Second Elevation (GTOP30)
. .
Venne
M. G.
Jasperson
W. H.
Venne
D. E.
1997
Difficult Weather: A Review of Thunderstorms, Fog, and Stratus, and Winter Precipitation Forecasting
;
AIRFM Command Tech. Report #A246633
,
Augsburg Coll Minneapolis MN Center for Atmospheric and Space Sciences
,
Minneapolis, MN
,
USA
.