Cloud Optimized GeoTiffs

While Felt can read many raster formats, storing your data as a Cloud Optimized Geotiff (COGs) allows efficient access of subsets of data for fast viewing.

This feature is only available to customers on the Enterprise plan. To upgrade, contact sales.

Configuration

Required?	Configuration option	Recommended values
Mandatory	Overview resampling	bilinear for floating point data nearest for integer or categorical data
Recommended	Compression scheme	YCbCr with JPG compression for 8-bit RGB data DEFLATE, LZW, or ZSTD for floating point data
Recommended	Transparency	Set an explicit Nodata mask band
Mandatory for streaming, recommended for all uploads	Block size	256x256 or 512x512
Recommended*	Summary statistics	Min, max, mean and standard deviation. Stored in the header, per the GeoTIFF specification

Uploading Cloud Optimized GeoTIFFs without summary statistics to Felt can be slow and inefficient, since Felt will have to read large amounts of data in order to compute those values.

Converting raster files to Cloud Optimized GeoTIFFs

1. Install the GDAL command line tools. If you’re using homebrew on MacOS, you can do this with brew install gdal
2. Use the gdal_translate command with the COG output format and the -stats flag. This will create a COG with summary stats and overviews that will be most efficiently streamed into Felt.

Example for floating-point data

$ gdal_translate -stats -of COG -resampling bilinear -co COMPRESS=LWZ {input}.tif {output}.tif

Example for integer data

$ gdal_translate -stats -of COG -resampling nearest -co COMPRESS=JPEG {input}.tif {output}.tif