Ice Floe Segmentation of MODIS imagery

In the segmentation output folder you will find:

for each day of analysis:

YYYY-MM-DD-JD_terra_final.tif: the final georeferenced masked floes

YYYY-MM-DD-JD_terra.csv: the properties tables for the floes. see https://scikit-image.org/docs/stable/api/skimage.measure.html#skimage.measure.regionprops for a full description of each parameter. The columns are:

label: corresponding number in the .tif file

area (in number of pixels)

centroid-0: centroid of the floe (row)

centroid-1: centroid of the floe (column)

axis_major_length

axis_minor length

orientation: angle between 0th axis (rows) and major axis

perimeter: in pixel units

intensity mean: average red channel value of floe

for each year: mask_values_YYYY.txt : text files with columns:

doy: day of year

ice area (pixels)

total area for analysis (not cloud or land mask)

sea ice concentration

reproj_land.tif: the land mask

df_all_withloc.csv: includes all floe properties from all years- same columns as individual days plus:

year

doy: julian day of year

perim_km (perimeter in km)

area_km (area in km^2)

circ: C = 4*pi *area / (perim^2)

lon_ps: polar stereographic x values epsg:3413

lat_ps: polar stereographic y values epsg:3413

lon: epsg:4326

lat: epsg:4326

outisde of the segmnetation output folder:

df_stats_all2.csv: a combination of all the mask value files with additional columns:

floe area: total area of floes within the image

floe count: total number of floes in the image