In the following graphs, I present the simulation data generated with SimulO.

Spatial overview

Light attenuation profiles

Same as previous but with data normalized.

In this graph, I check how K values change with increasing distance to the center of the melt pond.

Effect of melt pond size on K estimations

In this section, we evaluate the effect of varying melt pond size on predicted light (both Ed and Lu).

Average light profiles

First, I calculate averaged light profiles based 5 different classes of distance from the center of the melt pond. The colored lines in the following graph will be used as reference profiles from which the light intensity at 0.5 meters will be propagated in the water column.

This graph is just another view of the reference profiles.

K estimations

To calculate K, I binned SimulO data into 15 bins of 1 meter starting at 5 meters from the center of the melt pond. Then I used these 30 light profiles (15 Ed and 15 Lu) to propagate Ed from the 5 reference profiles.

Light propagation

Finally, I propagate reference profiles from light intensity estimated at 0.5 meters using previously calculated KEd and KLu.

Errors on predictions

Errors of the predictions have been calculating by the relative difference between the integral of the reference and predicted profiles ((reference - predicted) / reference).