Ice type ASPeCt code:

The following notions about translations of the ridged ice type and fraction ASPeCt codes apply to all Arctic and to Antarctic data starting with the WWOS cruise in 2006 and are further amended in 2016 / 2017 as indicated.
100:
Fraction of deformed ice and average ridge height both set to 0 if cells are empty; Otherwise, if the fraction cell contains a 0 then fraction of deformed ice is set to 0.05 and average ridge height is set to 0.1 m
Starting with Arctic 2016 data a 100 is taken as 0.0 ridged ice fraction if ridge height is zero or not given and if the ridged ice fraction is not given (empty cell)
Starting with Antarctic 2017 for a 100 the ridged ice fraction is 0.05 and the ridge height is 20% of the SIT for SIT < 0.2m, otherwise 0.1 m

200: (rafted pancake ice)
Fraction of deformed ice: 0.5
Average ridge height: 0.05 m --> changed to one half of the SIT starting with PIPERS cruise in 2017 for ANTARCTIC

300: (compressed pancake)
Fraction of deformed ice: 0.5; if the ice type is NOT pancake, however, then the fraction is set to 0.25 as it is then compressed, frozen together pancakes which possibly (this is an assumption) have a larger floe size. In these cased 0.5 is only kept if the floe size type is 500 or less in the ASPeCt code.
Average ridge height: 0.02 m --> changed to one half of the SIT starting with PIPERS cruise in  2017 for ANTARCTIC

400: (rafted nilas) - unless there is a number given for the fraction
Fraction of deformed ice: 0.1
Average ridge height: 0.01 m --> change to one half of SIT starting with PIPERS cruise in 2017 for Antarctic

All the above said (for 100 through 400) is obsolete if the ice type is 10, 11 or 12; then ridged ice fraction and ridge height are both set to 0.0

500 to 800 (without any notion about deformation):
Fraction of deformed ice: 0.05
Average ridge height: 0.1 m
Starting with Arctic 2016 data, for a 500 with a ridge fraction of 0 the ridge fraction is set to 0.05, if the ridge fraction is not given at all it is set to -9.9; in both cases ridge height is set to -9.9
Note that sometimes a 500 is given with 0 for ridge fraction but a non-zero ridge height. This underlines that setting a 0 to 0.05 in these cases makes sense.

Cases with partial sea ice concentrations set to 0 but with, e.g. information about ridge fraction and thickness, are considered as < 5% sea ice concentration and set to 3%. This happened for ASSIST data of, e.g. LANCE and Polarstern cruises. If setting these values to 3 % (instead of 0%) but the total SIC is 100% requires that one other SIC value needs to be reduced by 3% - typically I use the ice type with the maximum fractional coverage for this, i.e. if total SIC is 100%, primary SIC is 10%, secondary SIC is 90% and tertiary SIC was 0% then the latter two values are changed to 87% and 3%, respectively. If, what happens as well, there is 0% total sea ice concentration and 0% primary SIC but ice type et cet. given then total and primary SIC are set to 3%.
There are cases where two, e.g. secondary and tertiary SIC are given as 0% but ice type and thickness are given; in these case both partial SIC values are set to 2% and the respective remaining partial SIC is set to the original value minus 4%. If total SIC was 0% in this special case then it is set to 4%.

For the two Healy cruises in the Arctic in 2002 and 2004, melt-pond concentration and snow type on the primary ice are from the "general" melt pond coverage given in the comments, for fraction of deformed ice, sea-ice thickness and snow depth of the primary ice there is no discrimination between ice types in these parameters. For secondary ice, sea-ice thickness results from information about the ice type where we used for dark nilas 3 cm, light nilas 8 cm, grey ice 12 cm, grey white ice 20 cm and for the FYI types 60, 70 and 80 50 cm, 95 cm and 135 cm as fixed sea-ice thickness values; for MYI sea-ice thickness is set to 150 cm. Note that for the secondary ice type we chose the next thinner type of primary ice type in case it was FYI. If primary ice type was MYI than secondary ice type was the thickest FYI category.

In some cruises ridge spacing is given instead of an areal fraction. Our "translation" was then the following - depending on the floesize (given in ASPeCt code)
600 floesize
ridge spacing --> fraction of deformed ice
500 --> 0.05
300 0.1
200 0.15
150 0.2
100 0.25
80 0.3
50 0,35
40 0.4
30 0.45
25 0.5
20 0.55

500 floesize
200 --> 0.05
150 0.1
100 0.15
80 0.2
50 0.25
40 0.3
30 0.35
25 0.4
20 0.45

400 floesize
100 --> 0.05
80 0.1
50 0.15
40 0.2
30 0.25
25 0.3
20 0.35

Starting with 2016 Data in the Arctic the following change applies:
stage-of-melt is not reported; instead use a combination of 
PMPP = melt pond pattern (first digit = 1 for linked, 2 for discrete), 
PMPT = melt pond surface type (second digit = 1 for frozen, 2 for open, 3 for bottom up) and PMDI = dried ice yes or no (= 1 or 0) and 
PMRI = rotten ice yes or no (=1 or 0). 
If any of these four is not reported it gets a 9. A record might be, e.g., 1190 or 2399; if the value is -9 then none of the stage-of-melt variables was reported. 

If wind speeds are given in knots they are transferred into m/s by multiplying with 0.51444

Longitude and latitude inconsistencies are corrected manually

Inconsistencies in air temperatures and sea water temperatures are also corrected manually.
Example: if sea water temperature is -1degC -1degC +1degC -1deg -1degC then the +1degC is taken as an outlier and corrected to -1degC

Cases where all observations are set to 0 in an otherwise seemingly consolidated ice cover are set to missing values provided that total SIC of at least one of the direct adjacent observations is larger / equal than 80% - provided that the ice type is not new ice or young ice. If the ice type is young ice or new ice, then total SIC of 0% can occur next to an observation of 100% total SIC.
We note that this setting of cases with 0% total SIC to missing value potentially reduces the actually present fraction of open water but the gain of reducing SIC noise in otherwise consolidated ice is anticipated to be larger.