Snow and Snowmelt Runoff Accounts for 50-80% of the annual runoff in the Western US – critical for water resources Drives periodic flooding in the Western US Need to understand and quantify changes due to Climate change Vegetation and land use change It may be obvious to this audience, but just to set the stage, in considering snow and runoff some of the potential effects of climate change are: (read from slide). These speak to the need to examine trends and understand the interactions involved.

Learning Objectives To be able to quantify the amount of snow in terms of depth, density and snow water equivalent To be able to describe how snow is measured To be able to calculate the energy required to melt a snowpack and how quickly it may melt

Snow Pack Characteristics

Snow Pack Characteristics What is a Snow Pack? Porous Medium ice + air (+ liquid water) Generally composed of layers of different types of snow more or less homogeneous within one layer Ice is in form of crystals and grains that are usually bonded together forms a texture with some degree of strength

Snow Pack Characteristics Snow Water Equivalent (SWE) The height of water if a snow cover is completely melted, on a corresponding horizontal surface area. Snow Depth x (Snow Density/Water Density)

Density of Snow Cover Snow Depth for One Inch Water Snow Type Density (kg/m3) Wild Snow 10 to 30 98” to 33” Ordinary new snow immediately after falling in still air 50 to 65 20” to 15” Settling Snow 70 to 90 14” to 11” Average wind-toughened snow 280 3.5” Hard wind slab 350 2.8” New firn snow 400 to 550 2.5” to 1.8” Advanced firn snow 550 to 650 1.8” to 1.5” Thawing firn snow 600 to 700 1.6” to 1.4”

Quantifying Snow Water Equivalent Field Snow Surveys (courses, pits, tubes) SNOTEL system Stereo photo interpretation Gamma radiation LIDAR Gravity Microwave Remote sensing Model + data assimilation This reviews some of the methods for quantifying snow water equivalent. Ground based methods include field snow surveys and the SNOTEL system based on snow pillows. These have the disadvantage of being point measurements that are hard to extend over a large watershed. Remote sensing offers the only real opportunity to obtain spatial snow water equivalent and here are some of the methods. (Discuss advantages and limitations). I think that to make real progress in quantifying SWE we really need to integrate models with measurements, an assimilation approach. NOHRSC has made quite a bit of progress in this area, but more is needed.

SNOTEL System for Measuring Snow and Supporting Spring Runoff and Water Supply Forecasts

Snow Water Equivalent at Tony Grove Lake

Multiple Years of SWE 2011 2012

SNOTEL Stations Near Logan

Streamflow in the Logan River From http://nwis.waterdata.usgs.gov/

Predicting Streamflow Data from http://www.wcc.nrcs.usda.gov/snow/

March 25 Quick 2011 Peak Runoff Prediction 1622 +- 540 cfs (95% confidence interval) 2011-06-25 1710 cfs X=51.8 in Y=1622 cfs Var of Y 219657 cfs2 MSE = (1-R2)Var RMSE = 270 (1 std dev error) 95% CI = 2 RMSE =+-540 cfs 2450 cfs in 1907 2000 cfs in 1916 1980 cfs in 1899 1980 cfs in 1984 1970 cfs in 1986

April 18 Quick Peak Runoff Prediction X=62.3 in Y=1798 cfs Var of Y 219657 cfs2 MSE = (1-R2)Var RMSE = 257 (1 std dev error) 95% CI = 2 RMSE =+-515 cfs 2011-06-25 1710 cfs 1984 1986 2006 1997 1982 2450 cfs in 1907 2000 cfs in 1916 1980 cfs in 1899 1980 cfs in 1984 1970 cfs in 1986 1740 cfs in 1997 1530 cfs in 2006 1330 cfs in 1982

Challenges in the prediction of snowmelt and streamflow Climate Change Advance in snowmelt timing Smaller snow pack area Uncertainty in regional variability Land Use Change Transitions to Agriculture or Urban Conifer/Deciduous Bark Beetles Both Diminished statistical validity of past observations for current conditions

Quantifying snowmelt runoff requires knowledge of the quantity of water held in snow packs energy exchanges the magnitude and rate of water lost to the atmosphere by sublimation interception and the role of vegetation the timing, rate, and magnitude of snow melt the fate of melt water If we are to quantify snowmelt runoff we need to know the following four things: (read from slide). These involve combining observation and modeling, so a theme that I will promote is learning through integration of observation and modeling.

Fluxes dependent on Canopy temperature Utah Energy Balance Snowmelt Model To predict accumulation and melt of snow, and surface water inputs in open and forested environments Above Canopy Inputs Qsi ea Ta Precip Wind Qhc(Ta,Tc) Qec(ea,Tc) Qlec(Tc) Fluxes dependent on Canopy temperature Qsib Qsid Qli , Qp, Canopy Variable Water Equivalent Wc Leaf Area Index LAI Direct radiance transmittance τb Diffuse radiance transmittance τd Fluxes dependent on surface temperature Qhs(Ta,Ts) Qes(ea,Ts) Qles(Ts) Qlec (Tc) τbQsib τdQsid τd Qli Qps Energy Content Us Water Equivalent Ws Q Snow Thermally active ground layer De Soil Qg Qm David Tarboton: http://hydrology.usu.edu/dtarb

MFR EQG SWE at CSSL (1986)

Spatial Variability

Conclusions Snow, as a driver of streamflow presents challenges for hydrologists needing to make predictions and inform watershed management Spatial heterogeneity Process physics Observation sparsity It is important to get it right!

Logan River at 6th West in Logan, June 2011

Show and Tell Images from http://www.anri.barc.usda.gov/emusnow/default.htm