Other explanations can be the cold content of the snowpack and the heat flux from the snowpack into the glacier ice or vice versa, but these processes can only partly explain the energy surplus. Besides the process-based explanations, measurement errors could explain a part of the energy surplus. We observed condensation in the downward-looking radiation sensor that potentially results in underestimation of the outgoing shortwave radiation.
Moreover, the derived turbulent fluxes are also uncertain and that could influence the energy balance closure as well. Sublimation rates peak in early afternoon Figure 3 , which coincides with findings of Reba et al. Positive net radiation in the daytime results in an increase in the turbulence in the surface boundary layer Wagnon et al. However, sublimation is strongly reduced on days where atmospheric humidity is high.
High humidity prohibits sublimation as the atmosphere is saturated and near-surface water vapor pressure gradients are weak. On days with low atmospheric humidity, wind speeds tend to be higher. Higher wind speeds result in a well-mixed layer above the snow surface and sustained vapor pressure gradients that support sublimation. The primary driver for sublimation is the shortwave radiation and soon after the glacier is sunlit the sublimation increases, conditioned by an initial vapor pressure deficit.
Once the wind speeds increase the sublimation is further enhanced and both sublimation and wind peak around The observed average daily sublimation rate 1. Reba et al. Sublimation rates of 0. In the Sierra Nevada, Spain, maximum sublimation rates of 0. Sexstone et al. However, Cullen et al. Also, high sublimation rates have been observed in the Andes at high altitude. Wagnon et al. Litt et al. The favorable climate conditions at high altitude, i. The bulk-aerodynamic method underestimates the latent heat flux in this study Figure 7 , whereas Fitzpatrick et al.
However, katabatic flow occurs mainly during night on Yala Glacier, which is excluded from the analysis as sublimation is negligible. The residuals for the bulk-aerodynamic method show no relation with meteorological variables Figure 8 , but only show a relation with the time of day.
In the early morning the sublimation is overestimated, whereas in the afternoon it is underestimated. The overestimation in the morning could be explained by stable atmospheric conditions which occur in the early morning.
The discrepancies between observed and modeled peak sublimation by the Penman-Monteith equation are explained by high-altitude conditions. This equation is driven by two terms, i. The net radiation typically peaks earlier than the sublimation rate, indicating that the Penman-Monteith equation is stronger driven by the net radiation than the vapor pressure deficit.
This a direct result of the air density which is approximately half of the air density at this altitude compared to sea level. The air density is a factor multiplied with the vapor pressure deficit, reducing the weight of this term for calculation of sublimation. The net radiation is negative during the late afternoon, which results in deposition instead of sublimation and, therefore, we omitted these values in Figure 7.
The low performance may also be partly explained by the uncertainties regarding the observed net radiation section Observed Surface Energy Balance, Meteorology and Sublimation as it is strongly driven by this variable. The aerodynamic resistance r a in the Penman-Monteith equation was used for calibration. Values of r a for sublimation of snow strongly vary in literature and relations between wind speed and r a have been used to estimate r a over a snow surface Nakai et al.
However, all these relations gave no satisfactory results. Nakai et al. A similar approach was tested, but no relation was found between r a and wind speed. The Kuchment and Gelfan empirical relation strongly underestimates the sublimation, which indicates that this empirical relation is not transferrable between regions.
The use of an empirical relation is often region-specific or even glacier-specific due to different climate and topographical conditions in other geographic regions and glaciers. However, linear regressions and multiple linear regressions show that sublimation at AWS Yala Glacier can be predicted with reasonable accuracy by wind speed and vapor pressure deficit Table 2.
Interestingly, off-glacier meteorological data has almost equal predicting capabilities as on-glacier data. Spatially distributed sublimation is strongly related to variations of wind speed in space Figure 9. Close to the ridge, wind speed is typically high Figure 9 , resulting in high daily sublimation totals. This illustrates a high spatial variability of sublimation on Yala Glacier.
The humid day shows lower sublimation totals than the non-humid day as high humidity leads to smaller near-surface vapor pressure gradients, resulting in lower sublimation rates. The surface temperature is lower on the humid day compared to the non-humid day Figure 9. On high-humidity days the observed net radiation is lower than on low-humidity days, resulting in less warming of the snow surface. On high-humidity days cloud cover is often present, which reduces the incoming shortwave radiation and therefore reduces the net shortwave radiation.
Although the net longwave radiation is larger on humid days, the shortwave radiation dominates the net radiation, leading to less warming of the surface and consequently colder snow surfaces on high-humidity days. This occurs regularly on Yala Glacier on the humid days and reduces near-surface vapor pressure gradients. The sublimation totals may differ considerably when extrapolated to the whole winter season, and the quality of the sublimation estimates is largely dependent on the quality of the WRF fields see Supplementary Material.
For example, wind speeds are typically overestimated over crests using atmospheric modeling at very high resolution e. This could lead to overestimation of our sublimation totals close to the ridge. The used scaling method, in which an average scaling factor is calculated between the WRF fields and the in situ observations, does not take into account the complex and potential non-linearity of the system, which may increase the uncertainty. The monthly cumulative sublimation shows large temporal variation Figure 5.
The monthly sublimation is highest in October and December when the relative humidity is lowest. Dry air enhances sublimation as it results in a steep near-surface vapor pressure gradient. Contrastingly, in January the monthly sublimation is relatively low when the relative humidity is also low.
This is likely explained by the coldest surface temperatures in January compared to the other months. Cold surface temperatures lead to lower saturation vapor pressure at the surface, reducing the near-surface vapor pressure gradient and therefore sublimation. The sublimation at the location of AWS Yala Glacier equals the simulated sublimation averaged over the entire glacier.
The simulated sublimation fields show high spatial variability, where sublimation totals are approximately a factor 1.
This illustrates that the fractions of snowfall returned to the atmosphere may have high spatial variability as well. It is likely that the fraction is higher at more wind-exposed locations, such as the ridge. However, the cumulative winter snowfall has uncertainties that are related to i undercatch of snowfall by the pluviometer, ii the actual snow-rain point, and iii spatial variability in precipitation.
Collier and Immerzeel showed with WRF simulations that, at the location of the pluviometer used in this study, the snowfall is 1. This would indicate even higher importance of sublimation to the water balance.
Even though cumulative winter snowfall is uncertain, our results show that sublimation and evaporation is a significant component of the water balance. Therefore, it is crucial to include this component in future hydrological and mass balance studies. Studies should be performed to estimate the importance of high-altitude sublimation at both catchment and regional scales. The bulk-aerodynamic method can for example be implemented in existing hydrological models and applied on a larger scale, either forced by WRF simulations, a meteorological monitoring network, or a combination of both.
This study quantifies only surface snow sublimation while blowing snow sublimation may also play an important role. A wide variation of blowing snow sublimation rates have been reported in literature. This variety is a result of different climate regions and blowing snow model setup Groot Zwaaftink et al.
For example, it has been reported that the sublimation of suspended particles is several factors higher than surface sublimation, as there is more ventilation and supply of dry air Strasser et al. However, most models do not include temperature and humidity feedbacks and therefore lack the self-limiting process of blowing snow sublimation Groot Zwaaftink et al. Simulating blowing snow sublimation is beyond the scope of this study and might have resulted in an underestimation of the sublimation in this study.
Therefore, future research should focus on quantifying the occurrence of blowing snow events and corresponding sublimation rates in the Himalaya. An eddy covariance experiment was conducted to measure snow sublimation on Yala Glacier at an altitude of 5, m a. The eddy covariance measurements show that the cumulative sublimation is 32 mm for a day period.
The average sublimation rate of 1. The performance of parameterizations of different complexity i. The bulk-aerodynamic method outperformed the other parameterizations and was used to simulate sublimation at the location of the eddy covariance system from 15 October to 20 April Furthermore, the spatial variability of sublimation was simulated with the bulk-aerodynamic method for a humid and non-humid day.
Required meteorological field were obtained from WRF simulations and field observations. The sublimation at the location of the eddy covariance system equals the simulated sublimation averaged over the entire glacier and is therefore representative for the seasonal sublimation on Yala Glacier. The spatial patterns of sublimation are strongly linked to the modeled wind speed patterns.
The sublimation totals on the non-humid day are a factor 1. This illustrates that the fraction of snowfall returned to the atmosphere due to sublimation may be much higher close to the ridge that is more wind-exposed.
This study quantifies surface sublimation only and future research should focus on including the sublimation of blowing snow as this may increase the sublimation estimate. We conclude that sublimation is an important component of the water balance and glacier mass balance; future hydrological and mass balance studies in the Himalaya can no longer ignore this component. ES wrote the initial version of the manuscript. PB performed the WRF simulations. All authors participated in fieldwork during autumn or spring or both.
This project was supported by funding from the European Research Council ERC under the European Union's Horizon research and innovation program grant agreement no. The views and interpretations in this publication are those of the authors and are not necessarily attributable to ICIMOD. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The reviewer VV declared a shared affiliation, with no collaboration, with one of the authors, JS, to the handling editor at time of review.
We thank the four reviewers for their constructive comments that helped improving the manuscript. Baral, P. Preliminary results of mass-balance observations of yala glacier and analysis of temperature and precipitation gradients in Langtang valley, Nepal. Bernhardt, M. The influence of lateral snow redistribution processes on snow melt and sublimation in alpine regions.
Bowling, L. Parameterization of blowing-snow sublimation in a macroscale hydrology model. Box, J. Sublimation on the Greenland ice sheet from automated station observations. Brock, B. Measurement and parameterisation of aerodynamic roughness length variations at Haut Glacier d'Arolla, Switzerland. Brun, E. Simulation of Northern Eurasian local snow depth, mass, and density using a detailed snowpack model and meteorological reanalyses.
Collier, E. High-resolution modeling of atmospheric dynamics in the Nepalese Himalaya. Cullen, N. Energy-balance model validation on the top of Kilimanjaro, Tanzania, using eddy covariance data. Fitzpatrick, N. Surface energy balance closure and turbulent flux parameterization on a mid-latitude mountain glacier, purcell mountains, Canada.
Earth Sci. Fujita, K. Spatially heterogeneous wastage of Himalayan glaciers. Gascoin, S. Wind effects on snow cover in Pascua-Lama, dry andes of chile, adv. Water Resour. Gordon, M. On snow depth predictions with the Canadian land surface scheme including a parametrization of blowing snow sublimation. Ocean 44, — Groot Zwaaftink, C. Drifting snow sublimation: A high-resolution 3-D model with temperature and moisture feedbacks.
Seasonal simulation of drifting snow sublimation in Alpine terrain. Herrero, J. Evaposublimation from the snow in the Mediterranean mountains of Sierra Nevada Spain.
Cryosphere 10, — Immerzeel, W. The importance of observed gradients of air temperature and precipitation for modeling runoff from a glacierized watershed in the Nepalese Himalaya.
Knowles, J. Energy and surface moisture seasonally limit evaporation and sublimation from snow-free alpine tundra. Kuchment, L. The determination of the snowmelt rate and the meltwater outflow from a snowpack for modelling river runoff generation. Litt, M. MacDonald, M. Parameterizing redistribution and sublimation of blowing snow for hydrological models: tests in a mountainous subarctic catchment. On the importance of sublimation to an alpine snow mass balance in the Canadian rocky mountains.
Earth Syst. So, the amount of snow that sublimates back into the air depends on the typical winter weather for a given location. The Weather Guys. Skip to content. Home About Listen Live! Search for:. How Much does Snow Evaporate? This is one sign that sublimation is underway. The snow is turning from a solid directly to a gas, bypassing the liquid watery stage. Even if it never melts. Each molecule jiggles at a particular speed and in a huge mass of them, some are always moving fast enough to escape the water or ice and join their gaseous buddies in the atmosphere.
Dry air accelerates this phase change. Solar infrared is also why the interior of your car heats up when parked in sunlight. Now find out why water is so amazing and unique!
From the beautiful stars and planets to magical auroras and eclipses, he covers everything under the Sun and Moon! Skip to main content. You are here This Week's Amazing Sky. How Snow Disappears Without Melting. Shedding Light on Sublime Sublimation.
0コメント