10. SAMPLING PROTOCOLS
10.1. SNOW DEPTH
10.1.1.
Notes on Sampling Snow Depth
10.1.2.
Potential Problems and Solutions in Sampling Snow Depth
10.2.
SNOW SURFACE WETNESS
10.3.
SNOW SURFACE ROUGHNESS
10.4. SNOW
PIT EXCAVATION
10.4.1.
Snow pit Orientation and Excavation
10.4.2.
Notes on Snow Pit Location
10.5. SNOW
PIT DEPTH
10.6.
SNOW DENSITY PROFILES
10.6.1.
Basic Protocol for Sampling Snow Density in the Snow Pits
10.6.2.
Notes on Sampling Snow Density
10.7.
SNOW WETNESS PROFILES
10.8.
SNOW TEMPERATURE PROFILES
10.9.
SNOW/SOIL INTERFACE TEMPERATURE
10.10.1.
Notes for Snow Grain Measurement
10.11.
SOIL BULK DENSITY
10.12.
SOIL TEMPERATURE PROFILE
10.13.
SOIL CORES AND SOIL ICE
10. SAMPLING PROTOCOLS
The following steps outline the general procedure for sampling snow depth:
Insert the probe vertically into the snowpack. Depth should be measure on a vertical access through the center of the Earth, not orthogonal to the local slope. It is more difficult to carefully insert the probe vertically if there is more than 5-m of probe attached. The probes are easy to use in shallow or low-density snowpacks. A number of factors make the simple measurement more problematic.
Most of the depths that will be encountered at most of the sampling locations will be less than 5-m. The easiest method for sampling and measuring distances between samples is achieved with a 5-m pre-assembled probe length. This length will allow sampling of most depths, and will allow easy measurement of the distance between the original stake position and subsequent point to be sampled. Since the points are all at distances that are increments of 5-m, it is simple to use the 5-m probe to mark off horizontal distances. When sampling in heavy timber, use a 5-m probe to measure horizontal distances and a 1 or 2-m probe to sample depths. It is sometimes difficult to get the 5-m probe upright in forested areas due to canopy and branches.
If the horizontal measurement puts the measurement location in a tree well do not adjust it to obtain a sample in the deeper snow. Record in the comment section for that point that it was a tree well. If it is suspected that the probe hit a fallen tree, stump, boulder or other surface anomaly, do not resample elsewhere. Record the suspicion in the comment section. Basically, no subjective decisions should be made to alter the sampling scheme that will bias the sample. If the sample location falls exactly on a tree, record zero depth and note that it was a tree location.
Check the compass bearings regularly. In heavy timber it is easy to turn slightly at each measurement. If the compass is checked often for bearing, then more accurate locations will be obtained. The locations of the measurements, as well as the snow depths, are critical to the research objectives.
Snow depth will be sampled with collapsible snow probes.
The probes have 0.01-m increments and snow will be measured to the nearest
0.01-m. Probes are constructed in 1.00-m long sections and can be joined
together to sample snow depth up to about 14 m. If more than 8 sections
are joined before inserting in the snowpack, bending will occur in the
probe sections when attempting to lift the sections vertically from the
horizontal to begin the depth sample. At depths greater than 8 m, simply
add additional sections one at a time to the probes already inserted in
the snowpack. These additional sections can then be removed one or two
sections at a time as the probe is extracted from the snowpack. Again,
removing a probe longer than 8-m and attempting to lay it back down can
result in bending and damage of the individual probe sections. (See cautionary
note below about losing sections in the snowpack.) These probes are very
strong in the vertical direction, but have little strength and considerable
flexibility when stressed from the side. They will simply bend under their
own weight if care in handling is not exercised.
10.1.2. Potential Problems and Solutions in Sampling Snow Depth
Ice lenses. Ice lenses may be mistaken for the ground surface. When in doubt, lift the probe a few more times and ram it down to break through a suspected ice lens. This technique will not damage the tip of the probe, even if it is hitting a rock rather than soil or ice. With experience, one can usually discern between rock, ice and snow based on the feel and sound of the probe.
Sampling too deep. In areas with soft ground, mud or duff it is possible to probe considerable depths below the snow/soil interface. Field workers will develop a feel for the difference between snow and soil or mud with time. If the probe is going too deep, the probe will come back out of the snowpack out with mud on the tip. Simply reinsert the probe next to the last measurement and be careful to find the interface. It should be close to the last depth, less the suspected amount that it penetrated the soil. Some of the sampling areas are in moist or boggy areas that will most certainly present this problem.
Sticking probes. In deep snowpacks the probes may become stuck. This usually occurs in dense snowpacks (densities greater than 400 kg m-3) or very deep snowpacks where the friction on the probe becomes great. Additionally, moisture on the probe from surface melt may refreeze at depths in the cold portion of the snowpack as the probe is inserted. Short, swift downward probing motion followed immediately by retraction of the probe 20 - 40-cm, may minimize both of the problems. This type of rapid hammering motion seems to work well in difficult snowpacks. If a rest is needed, do so with the probe lifted slightly off the bottom of the current depth and chances of icing and sticking will be minimized. A "T" handle can be fitted on the top of the probe using the same set screws that hold additional probe sections together. The T handle is useful for driving the probe down, as well as for getting a probe unstuck. Usually a rotation of the T handle will dislodge a troublesome probe. The T handle should not be used to drive the probe in when more than a meter of probe is sticking out above the snow surface. This can easily result in a bent or broken probe section above the snow since there is little lateral support for the probe shaft.
Bending probes. In order to keep from bending the aluminum shafts as they are inserted into the snowpack, the probe should be held close to the snow surface. A height of 1-m above the snow surface is usually adequate. Holding the probe higher may result in bending or breakage of the shaft section above the snow surface. In low-density, shallow snowpacks, this is usually not an issue and probes can be easily inserted holding the probe in any fashion. If a section is bent, replace it with a different one and return the bent section to the field supervisor at the end of the day. Do NOT attempt to straighten it in the field.
Losing sections. Care must be taken to not lose section of the probe down hole. There are two key reasons for this concern: 1, the equipment is needed for the rest of the day, and 2, the probes are very expensive. If a section is lost down the hole, retrieval should be attempted immediately. Keep in mind that digging time increases exponentially with increase in depth. It will take hours to reach a probe section 3 or 4-m deep. We cannot afford to tie up excessive time recovering equipment during the experiment. If a probe section is lost and it is determined that it cannot be recovered in a reasonable amount of time, record the location with as much accuracy as possible (point number, coordinates, description, etc.) and continue sampling. The best scenario, of course, is to take precautions that will minimize the chance of loss in the first place. Three practices will help greatly toward this end. First, on a regular basis (every five measurements) check all of the screws in the probe to make sure that they are all backed out flush with the probe shaft. They will slowly work their way back into the probe, which can allow the section to become uncoupled. Second, always use both screws when assembling probes, whether simply adding one section or assembling a longer probe. Last, if many sections are in use (greater than five) and disassembly is desired before moving to the next sampling point, remove the entire probe before disconnecting section. It is easy to drop the bottom portion of the probe as it is disassembled it if it is still vertical and in the hole. It may slide back to the base of the snowpack in this case.
General care. These probes are strong in the vertical dimension, but weak laterally as stated above. Care should be taken in transporting them. If skiing with the probes assembled, as will doubtless happen between sampling points, do NOT use the probes as ski poles or any king of support for balance or to keep from falling. Hole the probe near one end, drag the probe behind and be careful not to bind it between trees or brush that will damage the sections. If there is a slightly bent section and it cannot be replaced with an unbent one, use the bent section at the top of the probe. This will minimize drift of the probe at depths and reduce the chance of breakage deep in the snowpack.
Note that the sections are secured to one another by screwing
the screws OUT rather than in (
Figure 61
). It is the shoulder of the screw
set in the hole in the shaft wall that holds the sections together. If
the sections are attached by screwing the setscrews in, they will hold
for a while, but will loosen with use and sections of the probe will be
lost in the hole. This practice will also mar the inside of the shaft and
make assembly and disassembly difficult. Be careful not to back the setscrews
out too far. The screws are stainless steel and the shafts are aluminum.
This means the screw material is stronger that the shaft material and the
threads will eventually cut the shaft wall and enlarge the holes if the
screws are turned out too far. The screws need only be screwed out until
the screw head is flush with the outside diameter of the probe.
Snow surface wetness will be sampled by collecting a thin
layer off the snow surface with a gloved hand. Protocol and classification
are detailed below under Snow Wetness Profiles.
Snow surface roughness is measured at each snow pit site after the snow profile measurements have been taken. Record whether the surface is: smooth; wavy; concave furrows; convex furrows; or random furrows. If the surface is not smooth, lay the snow depth probe across the snow surface above the snow pit wall. If the surface is wavy or furrowed, use the folding ruler to measure the distance from the probe to the surface at the deepest points between contacts. Measure the distance between contacts over the 1-m length. Write down a rough mean for both values. Note that some surfaces will have regular dips and crowns and some will be highly variable. It is impractical to measure the surface roughness in detail due to time and technique constraints. Record basic information that describes the general observations.
Place the black background board horizontally and upright
in the snow pack. Place your field book next to the board, so that the
site identification is clearly visible. Take a single digital photograph
of the board and notebook.
10.4. SNOW PIT EXCAVATION
10.4.1. Snow pit Orientation and Excavation
Snow pit locations were chosen by a random selection of
grid cell locations in each 1-km cell. The same relative locations were
used in all twelve 1-km cells. Two snow pits will be excavated each winter
at each snow pit site. It is important that the first sampling exercise
does not influence the snowpack properties for the second sampling period.
The primary concern is that traffic or disturbance of the snow during the
first survey will alter the site such that anomalous conditions will be
sampled in the subsequent survey. To minimize the problem, we will dig
the first snow pit 2-m directly down slope of the pit-marking stake. The
second pit will be dug 2-m directly up slope of the stake. All snow will
be thrown down slope of the pits. It flat terrain where there is no discernable
slope, the first pit will be located 2-m directly south of the stake and
the second pit will be located 2-m directly north of the stake.
Place the snow depth probe against the pit wall in the
center of the smooth area that will be sampled. Be sure that the probe
tip is not pushed into the ground. Recorded the total depth as measured
on the probe at the snow/air interface. Leave the snow depth probe in place
against the pit wall for depth reference of other snow pit measurements
(e.g. snow density, temperature, etc.).
10.6.1. Basic Protocol for Sampling Snow Density in the Snow Pits
10.6.2.
Notes on Sampling Snow Density
If the digital scale malfunctions, it will be necessary to use the backup spring scale. Place a ski pole, handle first, in the snow pit wall. Attach the spring scale to the tip of the ski pole and the plastic bag to the scale. Tare the scale with the empty bag using the adjusting knob on top of the scale. Extract density measurements as outline above, and dump the sample carefully into the bag and weigh the sample. Be sure to dump the entire sample out before the next measurement is taken. Check the scale tare after every five measurements and adjust as necessary.
If the surface of the snow is irregular, try to find a uniform area to take the surface density measurements. If such a place does not exist, insert the cutter at a level just below the irregularities so that a complete sample is obtained.
Ice lenses may make getting a good sample problematic. If the lens is too thick, sample the snow above and below it. In the field notes record the actual depths sampled and a thickness of the lens not sampled. Most ice lenses can be sample if the cutter is held firmly with both hands while inserting. Do not sample a lens that is too thick as the cutter may be bent.
Snow density samples should be taken in two complete profiles. Start with a sample at the surface and work to the bottom of the profile, taking both samples at a given depth, and then proceeding to the next increment below. The 1000-cc density samplers take a 10-cm high so a complete profile can be taken from the surface to the ground in 10-cm increments. The samples should be taken and recorded in the field book in 10-cm increments following the first sample. For example, if the total depth in the snow pit were 217-cm, then the first two samples would be taken from 217-207-cm; the second two samples would be taken form 207-197-cm, etc. The last two samples would be taken from 17-7-cm. If possible, also take a sample from 10-0-cm, although this will often be impractical due to ground surface irregularities and vegetation.
In very dense snow it may be necessary to sample the snowpack using a rubber mallet to insert the cutter. This should be done with great care. The cutter should be hit only on the two corners below the handle. Excessive force should not be used with he rubber mallet and other objects should not be used to strike the cutter or damage will occur. Once the cutter is in, it may be necessary to drive the lid in with the rubber mallet as well. In this case the cutter must be held in with a wrist, hip, knee or foot depending on the sample level. Both hands are needed to hold the lid and hammer it, but pressure must be provided against the cutter or the lid will for it back out of the pit wall as the lid goes in. This may take a little practice in dense snow.
Be sure the cutter bottom is snow-free before placing it on the scale. A few snow grains will allow the cutter and sample to slide off of the scale.
Be very careful to keep the snow off of the scale. The scales are electronic and moisture does affect them. If they get too wet (this means very little moisture internally), they simply quit working. There is ample opportunity for moisture to get into the scale through the holes for the top plate.
Be sure to remove the top plate and place it in the scale
case for transport before the rest of the scale. Leaving the top plate
on will result in excessive pressure on the load cells when the case is
closed and damage will occur.
Snow wetness measurements will be made throughout the entire pit wall profile. The assumption that all layers with temperatures colder than 0oC are dry is not valid due to thermometer error and uncertainty. It is generally safe to assume that layers with temperatures less than -2oC are dry. All other layers should be tested. Note also that it is possible, and not uncommon, for a cold layer to overlay a melting layer where preferential flow paths have introduced melt water to lower layers while leaving overlying layers cold and dry. The snowpack is sampled by taking a sample from the pit wall with a gloved hand at the location of concern. The following classification and methodology will be used to determine the wetness profiles:
Dry: Snow grains have little ability to adhere to one another when compressed. It is difficult to make a snowball with this snow. Temperature is usually below 0oC, but dry snow may exist at this temperature, particularly in light of thermometer accuracy. Water content by volume is 0%; data code is D.
Moist: Snow tends to stick together when compressed, but liquid water is not visible even with a hand lens. Temperatures are typically at 0oC. Water content by volume is <3%; data code is M.
Wet: This snow adheres well with moderate pressure and is the perfect snow for making snowballs. Water cannot be squeezed out with moderate pressure, but can be seen in the contacts between grains with a 10X lens. The temperature is 0oC and this represents snow in the pendular regime. Water content by volume is 3-8%; data code is W.
Very Wet: Water can be squeezed out with moderate pressure, but the snow matrix still contains a considerable amount of air. The temperature is 0oC and this represents snow in the funicular regime. Water content by volume is 8-15%; data code is V.
Slush: The snow is saturated with water and contains
only isolated air bubbles. Cohesion is minimal and actually increases as
water is pressed out. Water drips freely from sample. The temperature is
0oC. Water content by volume is >15%; data code is S.
10.8. SNOW TEMPERATURE PROFILES
Snow temperature should be measured every 10-cm over the entire snow depth profile ( Figure 71, Figure 72 ). The thermometer should be cooled in the snow before measurements are made for at least 5 minutes. Once cooled, move the thermometer to a new location for the first measurement. An accurate surface temperature measurement is difficult to obtain. Solar radiation is the primary problem, but snow contact is also a problem. If the site is shaded, simply place the thermometer on the surface near the pit and take a measurement after the thermometer equilibrates (2-3 minutes is usually adequate). If the site is receiving direct beam radiation, shade a portion of the surface with a shovel or some other object ( Figure 73 ). Place the thermometer in the shaded area and record the temperature after it equilibrates. The temperature measurements should be made on the same 10-cm intervals as the density measurements over the entire profile.
Use only one thermometer so that relative differences
can be measured, rather than differences between thermometers. The second
thermometer is provided as a backup and should only be used if the other
one is damaged. The temperature measurements should be made simultaneously
with the density measurements. The thermometers should be left in place
for 2-3 minutes before each measurement is taken. If they are not taken
simultaneously with the other measurements, then excessive time will be
needed to complete the entire snow pit. It usually woks best to insert
the thermometer in the undisturbed snow immediately adjacent to the depth
probe. This practice allows accurate depth location and is not in the way
of the density samples. Be careful with the thermometers and place them
in the case provided for transport between pits. Slight bending of the
stem will result in incorrect measurements. Shock or bending will also
change the calibration. If it is suspected that the thermometer is damaged,
use the second thermometer provided. The case ends are marked "A" and "B"
so track can be kept of which one is being used. Be certain not to leave
them in the snow pit bottom after the last reading is taken. If any irregularities
in the measurements are observed, be sure to note them in the field book.
10.9. SNOW/SOIL INTERFACE TEMPERATURE
Temperature should also be measured at the base of the
snowpack. Insert the thermometer at the snow/soil interface on the same
pit wall where other temperature measurements were taken. Be careful not
to force the thermometer into frozen soil or rocks. Note the condition
of the ground in the field book (i.e. hard, frozen, unfrozen, soft, muddy,
etc.).
10.10. SNOW GRAIN SIZE AND TYPE
Snow grain size will be measure for each homogeneous layer
in the snow profile. Use a paintbrush, dry wall brush, putty knife, hand
or other instrument to determine where major grain boundaries exist in
the pit profile. Mark each boundary. Collect a sample from the pit wall
by gently scraping the point of interest with the crystal card. Use the
loupe-style hand lens with graduated reticule to determine the size of
three grains along their long axis. Choose the grains as randomly as possible,
e.g. do not look for the largest grains or smallest grains, simply choose
three and measure them. Record basic crystal type as either "round" or
"faceted". If both forms are present with no clear boundary or separation,
then record the layer as "mixed". Record the location and thickness of
all ice lenses and significant crusts.
10.10.1. Notes for Snow Grain Measurement
Be sure to keep the crystal card cold. The card should be place in the snow in the pit wall when not in use and should be moved back and forth on a cold layer of snow between measurements when the air temperature is warm. This will reduce the amount of melt and crystal change that occurs during observation with the sample on the card.
In some layers there will be a gradual change in crystal type and size over the distance of observation. Arbitrary boundaries may be made, but should be chosen based on as much information as can be gathered, e.g. hardness, texture, paintbrush results, etc. When there is not a clear boundary within a layer, take a size sample (three measurements) at the bottom of the layer and again at the top of the layer and record all of the observations with their approximate location.
Note any irregularities or other observations in the field
book. For instance, if there is a horizon with a large number of pine needles
in it, or vertical ice columns, those are noteworthy.
Soil bulk density will be measured gravimetrically from
soil pits at five intervals over the 40-cm surface layer. This sampling
will take place during snow-free periods in the fall of 2001 only.
10.12. SOIL TEMPERATURE PROFILE
Thermistor strings will be installed during the soil profile
observations. Strings will be attached to stakes that will allow access
to the wires after burial by the seasonal snowpack. The thermistor temperatures
will be recorded during the snow pit observations.
10.13. SOIL CORES AND SOIL ICE
Soil cores will be extracted from the snow pit bottom
after the snowpack measurements are completed. Three 20-cm cores will be
extracted. Two cores will be cut into four 5-cm intervals each, which will
be bagged individually, labeled, and weighed on site. These samples will
then be sent to a laboratory for subsequent soil moisture analyses. The
third soil core will be split and disaggregated on site for determination
of the presence of ice. Where possible, a second set of cores will be sampled
from the 20 to 40-cm depth.