This month, we look at the distribution of very high resolution (VHR) optical imagery collected by image suppliers during the last year, in order to pick out interesting patterns and insights. The picture above shows a heat map of Very High Resolution (VHR) image acquisitions in the year 2016. The “warmer” (red/yellow) the spot the higher the number of acquisitions.
Very High Resolution imagery has a spatial resolution of between 31cm and 1m, and the available catalogues of these images are scanned and harvested multiple times per day by Geocento’s web application EarthImages. The images can then be discovered and ordered through the platform from relevant image suppliers including DigitalGlobe, Airbus, S I Imaging, Deimos Imaging and ImageSAT.
We thought it would be interesting to look at the distribution of these images around the globe and so have produced a set of maps which we show below.
What is the distribution of VHR imagery related to?
In general, VHR image acquisitions reflect the following:
• The distribution of population, involving more image collection along coasts and over cities.
• The locations of man-made and natural disasters, including in 2016, flooding events (e.g. Louisiana), earthquakes (e.g. Sumatra), wildfires (e.g. California) and pipeline spills (e.g. Colonial, USA).
• Zones of political instability and security concern. Ukraine stands out in this regard. Interestingly, according to the Google Earth blog, recent imagery from this and some other conflict areas does not make it into Google Earth.
• Areas of commercial interest. These are distributed very broadly and reflect a wide variety of applications from the user interested in a one-off image of their local facility, to a multi-national business with operational needs for imagery in various locations around the world.
• National budgets and assets for image collection. National and other agencies provide imagery for their own needs and differences in image acquisitions may reflect, to some degree, differences in budgets. The border between the USA and Canada can be seen as a step change in the number of acquisitions.
• The availability of good imaging conditions including daylight and cloud (see interesting insights into seasonal image collection). Optical imagery cannot be used during the polar night or during conditions of persistent cloud.
How is the distribution of imagery changing?
Well, if we show a map of differences between VHR acquisitions in 2015 vs 2016, with blue being locations with increased image acquisition in 2016 and brown/red being locations with decreased image acquisition in 2016, we see that hotspots can change significantly from one year to the next. Presumably, changes reflect transient events such as disasters, commercial activities and national programmes, while broad patterns that reflect population and imaging conditions remain constant from year to year.
Below, we “home in” on three regions to show some detail, with the left maps being the distribution of VHR image acquisition in 2016 and the right hand map being the change from 2015 to 2016, with brown/red indicating a decrease and blue an increase.
In later blogs, we will look at the distribution of radar imagery to contrast it with optical, and we will look into where people are searching in an aggregated sense, from among the few thousand image searches per month that are undertaken on Geocento’s EarthImages platform.
Any comments and thoughts on this would be very welcome. Do get in touch!