Correlating Satellite Cloud Cover with Sky Cameras

With the spirit of reproducible research, this repository contains all the codes required to produce the results in the manuscript: S. Manandhar, S. Dev, Y. H. Lee and Y. S. Meng, Correlating Satellite Cloud Cover with Sky Cameras, Proc. Progress In Electromagnetics Research Symposium (PIERS), 2017.

Please cite the above paper if you intend to use whole/part of the code. This code is only for academic and research purposes.

Code Organization

The codes are written in python. Thanks to Joseph Lemaitre for providing the scripts to process MODIS multi bands images.

The required dataset are present in this repository. The cloud mask data from MODIS satellite image are present in the folder cloud_mask, the combined cloud mask and cloud coverage data are present in the folder cmask_coverage_result, scripts to download MODIS data and various products can be found in the folder data_preparation, and the daywise cloud coverage data from sky cameras are present in the folder coverage_data.

Dataset preparation

• modisGrabber.py Downloads the MODIS MOD- and MYD- level 5 products.
• cloudmask.py Computes the cloud mass product from the downloaded MODIS data files. It uses the function PrecipitableWaterProduct.py while extracting the cloud mask values.

The calculated MODIS cloud mask data are present in a 3X3 matrix. It is stored in .mat format, along with the corresponding date and time. In our experiments, we use the average value of the various cloud mask values in the 3X3 matrix.

Core functionality

• nearest.py Finds the nearest observation, based on a given time stamp. It also provides the timestamp difference between the queried- and found- timestamp.
• normalize_array.py Normalizes a numpy array into the range [0,1].
• readCoverage.py Helper function to read and analyze all individual coverage files.
• display_cloudmask_on_map.py Displays the cloud mask over a map region.

Reproducibility

In addition to all the related codes, we have shared the analysis figure in the folder ./figs.

Please run the notebook main.ipynb to generate the result mentioned in the paper.