Earth Observation satellites are the main source of global data about the Earth. They are able to provide data for a wide range of applications such as topography, oceanography, Land cover, bio-diversity, pollution, disaster monitoring, atmospheric chemistry and clarity and of course climate. Accuracy and data quality are thus crucial if this information is to be fully utilised and most importantly inform policy makers on key decisions.
It is essential that satellite performance can be validated, and in some cases calibrated, using ground targets calibrated directly by in-situ measurements and local surveys. At present, such calibration/validation campaigns are carried out using teams of scientists mapping out the spectral characteristics of a target area. They need to map out relatively large areas so as to sample the pixel size of the satellite sensor and this is often done literally by walking a spectrometer across the terrain. In some cases, this can be supplemented by the flight of an aircraft, but this tends to be very expensive.
A novel approach to this problem is being developed at NPL. This technique utilises a remote controlled helicopter as an efficient and low cost means of mapping the terrain. The helicopter supports a palm sized reference spectrometer designed with good stray light rejection to make spectral surveys. The prototype instrument is limited to the key spectral range of 400 nm-900 nm, although this will ultimately be extended to the NIR. This project has also been used to engage local school children in the key issue of climate change, where they were involved in building the helicopter as part of a technology club.
The model helicopter can fly over relatively large areas and capture data more rapidly than other ground-based techniques. When fitted with a spectrometer or group of spectrometers, the helicopter can make relatively large-scale measurements at a number of detection angles. It also has a controllable flying range of more than 600 m diameter, which allows a spatial coverage of nearly 2.9 × 105 m2.
The technique has the potential to provide accurate and low cost reference surface measurements for large and difficult terrains over a short period of time, where the data can be used to vicariously calibrate and validate satellite derived information.