file:  Strategy.txt
version:  27Nov98
contents:  CSU sampling strategies 

    Measurements of ice nucleating aerosol particles were obtained 
during May 1998 NCAR C-130 research flights as part of NASA's FIRE3 
Arctic Cloud Experiment.  This document outlines the sampling strategies 
and procedures for these measurements. 

DEFINITIONS
    CFD = Continuous Flow Diffusion chamber
    SSw = CFD sample liquid water vapor supersaturation (%)
    TEM = Transmission Electron Microscope
    IN  = Ice Nucleating aerosol particle = IFN (Ice Forming Nuclei)

CFD Sampling Conditions
    The CFD samples at one set of temperature and supersaturation (T,SS)
conditions, selected by the operator.  The CSU sampling strategy was to 
choose (T,SS) conditions to be consistent with local clouds and cloud-
forming regions.  Over a period of time, samples were taken both above 
and below water saturation, with the maximum supersaturation typically 
about +5%, and the minimum typically -5%.  Occasionally, samples were 
made at very high SSw (>10%) for a few minutes.  The general strategy 
was: 

    Maintain constant (T,SS) conditions:
    a) During vertical profiles
    b) During impactor samples

    Vary conditions to get (T,SS) spectrum:
    a) When flying at approximately constant altitude in a relatively 
       homogeneous region, for example during ferry flights between Fairbanks 
       and the SHEBA ice camp.  Try to span a 10øC range of temperature and 
       +/-5% in supersaturation.
    b) During some of the IN membrane filter samples.

CFD Impactor Samples
    Samples of ice crystals which nucleated and grew in the CFD were 
impacted onto electron microscope grids.  The strategy was to collect as 
many as practicable with the CFD steady at one (T,SS) setting.  The 
number of crystals was monitored during sampling (based on optical 
counts), with the goal of collecting several hundred particles on each 
grid.  Typical exposure time was 30 minutes. 

PIXE TEM Grid Impactor Samples
    The PIXE impactor is an inertial impactor for size-sorted samples of 
total aerosol larger than ~60nm.  The CSU TEM grids are copper mesh with 
a formvar film and a light carbon coating.  PIXE samples were taken 
directly from the aerosol duct, not through the CFD flow distribution 
panel.  Flow was monitored with a rotameter downstream of the PIXE.  
Samples were obtained simultaneous with CFD impaction samples. Typical 
sample durations will be 10-20 minutes.  Only use stages 1, L2 and L1 
were used, with a TEM grid at L1 (0.06um) or on stages L2 and L1.  
Samples were changed in flight by storing exposed and unexposed disks in 
covered petri dishes.  TEM grids were loaded and unloaded at the hotel 
room.  Blank samples were also taken using the same handling procedures 
as exposed grids but no airflow. 

IN Membrane Filters
    Hydrophobic Nuclepore filters with 0.2um pore size were used.  
Exposed filters are stored and processed at CSU after the project.  
Filter samples were taken directly from the aerosol duct, not through 
the CFD flow distribution panel.  Pairs of filters were collected 
simultaneously with CFD measurements, with one filter at 10 LPM, and the 
other at 2 LPM.  Flow was measured using rotameters with valves located 
just downstream of the filters.  Sample duration was 0.5 to 1 hour.  It 
was not necessary to maintain CFD (T,SS) during filter samples since 
(T,SS) spectra can be derived from the filters during processing.  
Filters were in sealed filter holders except when being loaded or 
unloaded at the hotel room.  Exposed filters were transfered to covered 
plastic petri dishes, labelled to identify the sample.  Blank samples 
were taken, using the same handling procedures as exposed filters but no 
airflow.