Metadata for Sea Ice Permeabilities as part of the SBI project " Physical-
biological 
control of primary production in Beaufort and Chukchi Sea ice: Its contribution 
to 
shelf-basin interactions in the western Arctic ", Spring Process Cruises, USCGC 
Healy, May/June 2002 and 2004 (data filename 
SBI2002_2004SeaIcePermeabilities.xls):

Dataset Author: Hajo Eicken

Grant Number: NSF 0125464

The data set comprises sea ice permeability measurements on 5cm ice core 
sections. The 
ice cores were collected in May/June 2002 and 2004 during the Shelf-Basin-
Interactions 
(SBI) spring process cruises into the eastern Chukchi and western Beaufort Sea. 

Ice cores were drilled with an 8.5-cm diameter fiberglass barrel ice corer.  
They were 
then cut into 5-cm sections and centrifuged with a Beckmann CPR Rotor G3.7 at 
between 
1200 and 1400 rpm (2500 to 3400 ms-1) for 4 to 5 minutes at -2¼C (Weissenberger 
et al., 
1992). The sections were wrapped airtight in aluminum foil, sealed in a freezer-
quality 
Ziploc¨ bag and stored at Ð40 ¼C until further processing. 

The centrifuged core sections were analyzed in the cold room using a permeameter 
built 
according to specifications by Freitag (1999). 

The design of the permeameter is based on Darcy's Law that links the 
permeability k in 
m2 to the pressure differential _p in Pa across the height of an ice sample _L 
in m:
 
Here, _ is the viscosity of the fluid in Pa s and  is the specific discharge in 
m s-1 
calculated from the mass of the fluid in gram collected on the balance. In the 
experiments, n-decane was used as the fluid to be pumped through the ice because 
of its 
immiscibility with water, its low freezing point of Ð29.7ûC, and a viscosity 
that 
corresponds to that of brine (10-3 Pa s). The ice samples were restrained in a 
tighly fitting 
neoprene sleeve which was secured to two specifically fabricated plastic end 
caps with 
hose clamps. The end caps had openings for the pressure sensor and the 
temperature 
sensor. An OHAUS Navigator balance was placed underneath the drain tube below 
the 
bottom cap and measured the collecting n-decane. The pressure sensor hoses were 
attached to a pressure sensor evaluation board (All Sensors Corp.) which in turn 
was 
connected to a PC for data collection. The temperature probes connected to a 
thermometer which also sent data to the PC. 

The permeameter was used to calculate permeabilities in a 1-D direction along 
the 
principal axis of the core by applying a known pressure gradient to an ice 
sample and 
measuring the amount of liquid that permeates through the sample in a given 
time. The 
height of the decane reservoir was adjusted to build up the required pressure 
head. A 
height of approximately 0.5 m above the ice sample achieved the pressure 
necessary for 
the higher permeabilities (10-12 m2) while a height of up to 1.50 m was 
necessary for the 
lowest permeabilities (10-14 m2). Two pressure sensors measured the respective 
pressures 
and sent the pressure differential value to the computer. The thermometer 
dispatched 
temperature measurements to the computer for n-decane viscosity calculations. 
Lastely, a 
weight measurement from the balance was sent to the PC every 750 milliseconds. 
All 
collected data was recorded into a text file.

Prior to the experiment centrifuged ice core sections were prepared to ensure 
proper fit 
into the permeameter sleeves. Samples were removed from storage bags and the 
aluminum wrapping and adhered onto a metal plate by melting the bottom _ mm of 
the 
sample using a heating plate. The plate was then clamped into a lathe for an 
even 
reduction of the core diameter to 8.3 cm for a snug fit into the neoprene sleeve 
of the 
permeameter. Subsequently, a band saw was used to cut the metal plate off of the 
ice core 
and to square the opposite end of the core. The diameters were measured with a 
caliper 
and the values recorded. The height of the samples was also measured and 
recorded. The 
heights varied between 2.9 and 4.6 cm. The ideal length of ice samples has been 
determined as between 3 and 6 cm to minimize the effect of isolated pore spaces 
(Freitag, 
1999).

Experiments consisted of at least three consecutive runs per sample (where 
possible) and 
were conducted at a temperature of -10 ¡C in the cold room. The highest 
permeability 
measurable with the permeameter was 10-7 m2. This value depends on the smallest 
diameter in the entire permeameter set-up which is an inside valve diameter of 
5.5¥10-3 m. 
The lowest permeability measurable was 10-16 m2.  

Data files (MS Excel) contain 2 lines of header information, followed by the 
data blocks. 
Columns are arranged as follows: date, sample number (if sample was processed 
more 
than once), depth of the core section (distance is measured from the ice-water 
interface) 
in centimeter, permeability measurements from run 1 through run 7 in m-2, the 
permeability average over all runs, the standard deviation of the runs, and the 
dimensions 
of the ice core sections: diameter and height in m.

Cited references: 
Freitag, J. (1999). "The hydraulic properties of Arctic sea ice - Implications 
for the small-
scale particle transport (in German)." Ber. Polarforsch. 325: 150pp.
	
Weissenberger, J., G. Dieckmann, et al. (1992). "Sea ice: a cast technique to 
examine and 
analyze brine pockets and channel structure." Limnol. Oceanogr. 37: 179-183.


Point of contact:
Hajo Eicken, Geophysical Institute, University of Alaska Fairbanks, Fairbanks, 
AK 
99775-7320, phone: 907-474-7280, e-mail; hajo.eicken@gi.alaska.edu