Role of blowing snow in snow processes in Qilian Mountainous region

HongYi Li , Jian Wang, XiaoHua Hao

Cold and Arid Region Environmental and Engineering Research Institute, Chinese Academy of Sciences, Lanzhou, Gansu 730000,China

1 Introduction

Blowing snow plays an important role in snow processes in mountainous region. Its transportation and sublimation accounts for noticeable part in the snow hydrologic cycle. Snow accumulation and spring snowmelt discharge are largely influenced by the redistribution of snow cover because of blowing snow (Kaneet al., 1991;Leonard and Maksym, 2011). Some studies have demonstrated that blowing snow sublimation accounted for 10%–50% of precipitation in arctic prairie region in winter (Pomeroyet al., 1998). In some tundra regions, 18%of annual snowfall was transported to sink areas while only 86.5% accumulated on the ground at catchment scale, because of blowing snow (Pomeroy and Gray,1997). Remote sensing data recently has been used to analyze the blowing snow processes. Some researchers have used MODIS products to monitor the blowing snow properties over Antarctica (Palmet al., 2011).

Different blowing snow models have been developed for a long time. An earlier model named PBSM (Prairie Blowing Snow Model) was developed and used in Arctic prairie region (Pomeroyet al., 1993). Recently more similar jobs focused on exploring the role of blowing snow in snow hydrological cycle in different conditions(Grootet al., 2011). Bowlinget al.(2004) coupled a parameterized scheme of blowing snow into distributed hydrological model, in which topographic factors are important inputs. Blowing snow processes have been inserted into many latest distributed cold region hydrologic models such as Snow Model (Liston and Elder,2006), ALPINE-3D (Lehninget al., 2006) and CRHM(Pomeroyet al., 2007). All these researches about blowing snow focused on two aspects, the first was the redis-tribution of blowing snow, and the second was the blowing snow sublimation which is much larger than surface sublimation.

There were few researches about blowing snow in China. Wang and Zhang (1999) have suggested a regionalization of snow drift in China. More related researches focused on the transportation of blowing snow in some special region (Liu and Lin, 2005; Donget al., 2010; Liet al., 2011). However, blowing snow sublimation and the role of blowing snow in the hydrological cycle have been rarely studied in cold regions in China.

In Qilian mountainous region, snow is abundant in spring and autumn. The spring discharge from the upriver of this region is the most important water resource for the three major basins: Shulehe River Basin, Heihe River Basin and Shiyanghe River Basin. Some observations have found the blowing snow was an important phenomena in this region, with high wind speed (Liet al.,2009). There were almost no studies on the role of blowing snow in snow processes in this region. We studied the energy and mass processes of blowing snow by field investigation and model simulation at detail.

2 Observation

We set our measurements at Binggou watershed in Qilian mountainous region. Observations have found there was abundant precipitation in this region influenced majorly by continental climate (Yanget al., 1992; Liet al., 2010; Chenet al., 2011). Blowing snow frequently occurred with high wind speed. The drifted snow was redistributed at large range on the interaction between wind and topographies. Snowpack were drifted over from the top of mountains under the high wind speed, so as to extensive redistributed snow accumulated in the valley and sink.

There are two Auto Weather Stations (AWS). The Yakou station (4,146 m; 38°00′N, 100°14′E) located in the upstream of the watershed, and the Binggou station(3,449 m; 38°04′N, 100°13′E) located in the outlet of the watershed (Liet al., 2009). Wind speed, air temperature,relative humidity, precipitation and radiation were measured in the two AWSs. In addition, snow depth was measured by ultrasonic instrument in Yakou station,without this measurement in Binggou station. The interval of observation time is 10 minutes. The air temperature and precipitation in 2008 snow season were illustrated in figure 1 and figure 2.

We launched our field investigation on Mar., 2008.Two snow surface sublimation pans were set near the two AWSs. The pan was made by white plastics so as to avoid the solar radiation heating and heat transfer between different snow layers. The size of pan is 25cm×40cm with height of 20 cm. We weighed the snow mass in the pan to calculate the snow loss and snow density in 7:00 and 19:00 daily. The measuring range of the weighing balance is 5,000 g, and the accuracy is 0.1 g.We set fixed rules to monitor snow depth change and the results were compared with ultrasonic snow measurements. Besides the measurements in AWS, the precipitation was recorded manually for a calibration for automatic measured results from AWS.

Figure 1 Measured precipitation at Yakou station during 2008 snow season

Figure 2 Measured air temperature at Yakou station during 2008 snow season

3 Simulation methods for snow transport and blowing snow sublimation

The occurrence of blowing snow is unstable over different spatial and temporal regions. Some studies have indicated that blowing snow phenomena were intermittent even in concentrated blowing snow events (Schmidt,1986). Because of the high speed of drifted snow particles in the air, these particles are sublimated rapidly than in a static status in the ground surface. Obviously the method for computing snow surface sublimation is not fit for this situation. Pomeroyet al.(1993) suggested an efficient model (PBSM) for dealing with blowing snow.Many parameterized schemes in hydrologic models were from the algorithms of PBSM (Liston and Elder, 2006).

Generally, blowing snow was divided into saltation layer and suspension layer. The mass change of blowing snow can be formulated by:

where,Bsaltis the transport rate of saltation layer,Bsuspis the transport rate of suspension layer,BSis the sublimation rate of blowing snow.

3.1 The occurrence of blowing snow

Blowing snow events occur when the motive force from wind exceeds the shear force in snow particles. The occurrence of blowing snow depends on the synthesis of wind speed, snow surface condition, air temperature and so on. Statistic results indicated that there is relevant correlation between mean critical wind speed and air temperature (Li and Pomeroy, 1997a). In a given air temperatureTa(°C), the mean critical wind speed driving blowing snow can be formulated empirically (Li and Pomeroy,1997b):

The above formulation only give an average wind speed, more detailed snow status are needed to determine accurately the occurrence of blowing snow. For solving this problem, it is assumed that the occurrence of blowing snow is a random process. The probability of occurrence is similar to a cumulative normal distribution (Esseryet al., 1999):

where,u10is the wind speed at 10 m.

For dry snow and fresh snow, there are:

whereAis snow age.

For wet snow and old snow, the values are empirically given as:ū=21 m/s,δ=7 m/s.

We empirically used the air temperature as the criteria of distinguishing dry snow and wet snow. When air temperature is greater than 0 °C, the snowpack is assumed to melt and regarded as wet snow.

3.2 Snow transport and sublimation

The transport rate of blowing snow is the sum of fluxes in saltation layer and suspension layer:

Instead of the complex physical based formulation,Esseryet al.(1999) simplified the determination formulation for transport rate:

Blowing snow sublimation (Qbs) was calculated by the following equation (Esseryet al., 1999):

wherebis an empirical constant,σ2is the unsaturation of humidity,F(T)only depends on temperature.

3.3 Snow surface sublimation

Snow surface sublimation was calculated as the following equation (Cline, 1997; Hoodet al., 1999):

where ΦMis the stability function for momentum, ΦEis the stability function for water vapor,ραis air density,Pis the atmosphere pressure,Kis the Karman constant (0.4),z2is the measurement height of 2 m,z0is the aerodynamic roughness,zdis the zero plane displacement. There we set the value as 0 cm for snow surface,e2is the vapor pressure at height ofz2,essis the vapor pressure at snow surface,v2is the wind speed at height ofz2.

Snow sublimation includes the surface sublimation and drifted snow sublimation at vertical one dimension.With continues field observation, the time of occurrence of blowing snow could be determined. However, it is difficult to carry out the continuous observation. It is assumed the occurrence of blowing snow is in random, so the sublimation of snow can be determined by the following equation:

wherePsis the probability of blowing snow occurrence mentioned above.

The empirical simulation methods used in the study are from the parameterized schemes in other cold region.These schemes have not been verified using detailed local data, since we have insufficient instruments in the region. There may be some uncertainty in the simulation.However, these schemes informed a quantitative reference method for evaluating the role of blowing snow in the snow processes.

4 Influence of blowing snow on snow accumulation at plot scale

4.1 Field investigation results

We analyzed the snow depths from field investigation at Yakou station in Mar., 2008. The results were contrasted with the measured snow depth change. Blowing snow was rare in Binggou station, so Binggou station was not involved in this analysis. The results indicated that the snow accumulation did not meet the snowfall events strictly. We computed the positive delta of snow depth, and assumed this positive change is the increasing snow height. The snow density of increased snow layer was assumed as 0.2 g/cm3. Snow Water Equivalents(SWE) change because of snowfall and drifting events were distinguished according to the field records in Mar.,2008. The results were contrasted with the measured snow depth increases from AWS (Figure 3).

Figure 3 Measured snowfall, blowing snow and snow water equivalent change at Yakou station in Mar., 2008

The contrast results indicated that the weighing measured increases of snowfall and blowing snow meet the measured SWE change from AWS basically. There were six snowfall events as in-situ observed in Mar.,2008. Obviously, the snow mass increases were not corresponding to each snowfall events. The reason is the influence of drifting on the snow accumulation. For example, in Mar. 6, the weighing measured SWE increase was 8.24 mm while the results from AWS was 2.14 mm.The other huge discrepancy was in Mar. 20. We observed an obvious snowfall event in Mar. 17, however the weighing measured result was -0.31 mm. The drifting effect was the most important reason accounting for the disagreement, although snow sublimation may contribute a little mass decrease. The measurement results, when only blowing snow occurred without snowfall, meet the measured SWE change from AWS.

The above analysis demonstrated that snow depth was influenced largely by blowing snow at Yakou station.It is necessary to distinguish the blowing snow events from snowfall events, and pay more attention to the uncertainty of snowfall measurements because of blowing snow, for a better understanding and modeling of snow processes.

4.2 Simulated and observed occurrences of blowing snow

Simulation results indicated that the critical wind speed and probability of blowing snow occurrence changed seasonally (Figure 4, Figure 5). Critical wind speed was lower before the end of Feb., while probability of occurrence was higher too. This period was an accumulation season, there were more snowfall events and the air temperature was less than 0 °C. Snow surface was dry and without heavy densification, so the blowing snow occurrence was frequently. With air temperature increased, the critical wind speed increased and the occurrence probability of blowing snow decreased.The major reasons were the increase of air temperature,snow surface melt and refrozen of snow particles. For a validation, we contrasted the observed results in Mar. 5,Mar. 12, Mar. 21 and Mar. 31. We observed evident blowing snow phenomena in these days, and the simulated probabilities of occurrence of blowing snow were 0.72, 0.83, 0.81 and 0.94 separately. It matched the field observation actually.

Figure 4 Mean critical wind speed for blowing snow occurrences at Yakou station

Figure 5 Occurrence probability of blowing snow at Yakou station

4.3 Snow change influenced by blowing snow

The blowing snow transports were calculated by the equation(3)and equation(4). The illustration (Figure 6)indicated the major snow transportations occurred from the earlier Feb. to the later Apr., 2008. The most frequently transportations were in February and March. The increases of wind speed and air temperature promoted the occurrence of blowing snow; however the occurrence was restricted by the snowmelt. So blowing snow occurred relatively frequently in winter. With the air temperature increased below 0 °C, the occurrence probability increased before snowmelt. In this period, snow transport increased under higher wind speed. When snowmelt season coming, the probability of occurrence decreased,snow transport decreased accordingly.

Figure 7 shows the continuing snow depth change from AWS measurements at Yakou station. It illustrates that the main snowfall events occurred in Mar., Apr.and May. There was little precipitation before Mar.,while the snow depth changed with large fluctuations,as the dash circle indicated in figure 7. Obviously there were some blowing snow events, and it is not a typical snow melt or sublimation process. The dry snowpack in winter was easily to be drifted and resulted as the fluctuated snow depth change. The similar phenomena could be found in March. The simulation results matched the observation.

Figure 6 Simulated daily blowing snow transport at Yakou station

Figure 7 Snow depth change over whole snow season at Yakou station (the period with frequent blowing snow was marked using the dash circle)

5 Influence of blowing snow on snow sublimation

There we made two steps to explore how blowing snow influence the snow sublimation. First, snow surface sublimation was simulated and contrasted with the weighing measurements in Yakou station and Binggou station. The role of blowing snow was ignored in the first step, so as to analyze the bias of simulation in different wind conditions. Next, we computed the snow surface sublimation and blowing snow sublimation, and quantitatively analyzed how much blowing snow sublimation accounts for the total sublimation.

We made weighing measurements majorly in Mar.,2008. So we chose this period to simulate the snow surface sublimation. The measurements have proved there were only few blowing snow events in Binggou station and frequent blowing snow events occurred in Yakou station. Using equation(7), we simulated the snow surface sublimation and compared the results with the observations. The contrast indicated that there was good linear agreement between measurements and simulation in Binggou station (Figure 8). However, there was large disagreement in Yakou station. It demonstrated that there existed large error when only snow surface sublimation was considered in simulating sublimation under a high wind speed conditions. Of course, snow surface sublimation results can meet the measurements if without blowing snow occurrence. The wind speed in Yakou station was greater more than Binggou station, and the snow particles were drifted to air. When snow particles suspended in air, it was sublimated rapidly, and some transported snow were deposited there. It is the reason that simulation results of snow surface sublimation cannot meet the measured snow mass change at Yakou station.

We computed the snow sublimation in the snow season from Jan. to Apr., 2008, using equation(8). Figure 9 shows the results which indicated the blowing snow sublimation was the greatest in February. In April, condensation exceeded the snow surface sublimation and blowing snow sublimation was greater. Blowing snow sublimations accounted for 27.2%, 34.5%, 35.1% and 145.7% of total sublimation of each month, respectively. The total blowing snow sublimation over the whole snow season accounted for 41.5% of total snow sublimation. The results quantitatively illustrated the importance of blowing snow in snow sublimation.

Figure 8 Measured and simulated snow surface sublimation at different AWS: (a) Binggou, (b) Yakou

Figure 9 Monthly snow surface sublimation and blowing snow sublimation at Yakou station

6 Conclusions

We analyzed the role of blowing snow in snow processes at plot scale in Qilian mountainous region. Only results from two stations were analyzed because of the limitation of observations.

The results indicate that blowing snow was frequently at a high altitude (Yakou station, 4,146 m). Snowpack was redistributed largely around the station under high wind speed conditions. Without considering blowing snow process in hydrologic model, the simulation results existed large error from real situation. The measured snow depth and precipitation should be analyzed so as to avoid the influence of blowing snow.

The simulation and field investigation results indicate that blowing snow was influenced largely by wind speed,air temperature, snow surface status and other environment factors, with obviously seasonal features. Blowing snow was more frequent in snow accumulation season when air temperature was less than 0 °C. The transport amounts of blowing snow were greatest on the beginning of snowmelt. With air temperature increased and more snowmelt occurred, blowing snow phenomena decreased.

Blowing snow sublimation accounted for a large portion of snow mass loss. Big error existed if only snow surface sublimation was considered in modeling snow sublimation, because the drifted snow particles were sublimated rapidly in air. The evaluated blowing snow sublimation accounted for 41.5% of the total snow sublimation at Yakou station in 2008 snow season.

This work has been funded by the National Natural Science Foundation of China (Grant Nos. 91325203,41001240, 41071227).

Bowling LC, Pomeroy JW, Lettenmaier DP, 2004. Parameterization of blowing-snow sublimation in a macroscale hydrology model. Journal of Hydrometeorology, 5(5): 745–762. DOI: 10.1175/1525-7541(2004)005<0745:POBSIA>2.0.CO;2.

Chen Q, Chen TY, Zhang YX, 2011. Analyses of energy field characteristics and precipitation in Qilian Mountains. Journal of Glaciology and Geocryology, 33(5): 1046–1054.

Cline DW, 1997. Snow surface energy exchanges and snowmelt at a continental, midlatitude Alpine site. Water Resources Research,33(4): 689–701. DOI: 10.1029/97WR00026.

Dong Z, Li H, Zuo HJ, 2010. Effect of vegetation height and coverage on snow drift on the typical grassland in Xilin Gol, Inner Mongolia.Journal of Glaciology and Geocryology, 32(6): 1106–1110.

Essery R, Li L, Pomeroy J, 1999. A distributed model of blowing snow over complex terrain. Hydrological Processes, 13(14–15):2423–2438. DOI: 10.1002/(SICI)1099-1085(199910)13:14/15<242 3::AID-HYP853>3.0.CO;2-U.

Groot Zwaaftink CD, L?we H, Mott R,et al., 2011. Drifting snow sublimation: A high-resolution 3-D model with temperature and moisture feedbacks. Journal of Geophysical Research, 116(D16):D16107. DOI: 10.1029/2011JD015754.

Hood E, Williams M, Cline D, 1999. Sublimation from a seasonal snowpack at a continental, mid-latitude alpine site. Hydrological Processes, 13(12–13): 1781–1797. DOI: 10.1.1.69.6406.

Kane DL, Hinzman LD, Benson CS, 1991. Snow hydrology of a headwater arctic basin 1. physical measurements and process studies.Water Resources Research, 27(6): 1099–1109. DOI:10.1029/91WR00262.

Lehning M, V?lksch I, Gustafsson D,et al., 2006. ALPINE3D: a detailed model of mountain surface processes and its application to snow hydrology. Hydrological Processes, 20(10): 2111–2128. DOI:10.1002/hyp.6204.

Leonard KC, Maksym T, 2011. The importance of wind-blown snow redistribution to snow accumulation on Bellingshausen Sea ice.Annals of Glaciology, 52(57): 271–278.

Li C, Zhu HW, Gao Y, 2011. Studied on erosive mechanisms of snowdrift in cold regions highway. Journal of Glaciology and Geocryology, 33(4): 784–790.

Li HY, Wang J, Bai YJ,et al., 2009. The snow hydrological processes during a representative snow cover period in Binggou watershed in the upper reaches of Heihe River. Journal of Glaciology and Geocryology, 31(2): 293–300.

Li L, Pomeroy JW, 1997a. Estimates of threshold wind speeds for snow transport using meteorological data. Journal of Applied Meteorology,36(3): 205–213. DOI: 10.1175/1520-0450(1997)036<0205%3A EOTWSF>2.0.CO%3B2.

Li L, Pomeroy JW, 1997b. Probability of occurrence of blowing snow.Journal of Geophysical Research: Atmospheres, 102(D18):21955–21964. DOI: 10.1029/97JD01522.

Li X, Li XW, Li ZY,et al., 2009. Watershed allied telemetry experimental research. Journal of Geophysical Research, 114: D22103.DOI: 10.1029/ 2008JD011590.

Li YY, Zhang Q, Xu X,et al., 2010.Relationship between precipitation and terrain over the Qilian Mountains and their ambient areas.Journal of Glaciology and Geocryology, 32(1): 52–61.

Liston GE, Elder K, 2006. A distributed snow-evolution modeling system (Snow Model). Journal of Hydrometeorology, 7(6): 1259–1276.DOI: 10.1175/JHM548.1.

Liu HH, Lin Y, 2005. Change trend and temporal-spatial distribution of snowdrift in China. Arid Zone Research, 22(1): 125–129.

Palm SP, Yang Y, Spinhirne JD,2011.Satellite remote sensing of blowing snow properties over Antarctica. Journal of Geophysical Research, 116(16): D16123. DOI: 10.1029/2011JD015828.

Pomeroy JW, Gray DM, Landine PG, 1993. The prairie blowing snow model—characteristics, validation, operation. Journal of Hydrology,144(1–4): 165–192. DOI: 10.1016/0022-1694(93)90171-5.

Pomeroy JW, Marsh P, Gray DM, 1997. Application of a distributed blowing snow model to the arctic. Hydrological Processes, 11(11):1451–1464. DOI: 10.1002/(SICI)1099-1085(199709)11:11<145 1::AID-HYP449>3.0.CO;2-Q.

Pomeroy JW, Gray DM, Shook KR,et al., 1998. An evaluation of snow accumulation and ablation processes for land surface modelling,Hydrological Processes, 12(15): 2339–2367. DOI:10.1002/(SICI)1099-1085(199812)12:15<2339::AID-HYP800>3.0.CO;2-L.

Pomeroy JW, Gray DM, Brown T,et al., 2007. The cold regions hydrological model: a platform for basing process representation and model structure on physical evidence. Hydrological Processes,21(19): 2650–2667. DOI: 10.1002/hyp.6787.

Schmidt RA, 1986. Transport rate of drifting snow and the mean wind-speed profile. Boundary-layer Meteorology, 34(3): 213–241.

Wang ZL, Zhang ZZ, 1999. Regionalization of snow drift in China.Journal of Mountain Research, 17(4): 312–317.

Yang ZN, Yang ZH, Zhang XC, 1992. Runoff and its generation model of cold region in Binggou basin of Qilian mountain. Memoirs of Lanzhou Institute of Glaciology and Geocryology, Chinese Academy of Sciences, 7: 91–100.