A study by Norris and colleagues was published in Nature last week, revealing how the distribution of clouds has been changing since the 1980s, on a planetary scale. The study detailed two main changes in cloud patterns, encouraged by the observed climatic changes which have been occurred over the last 50 years.
Norris stated that the main changes to cloud distribution included alterations to both their height (due to carbon dioxide changes), and to their location (due to a poleward movement of storm tracks and subtropical zones). The research shows that the spatial nature of cloud distribution was varied globally. It was evident that cloud albedo (essentially their ‘whiteness’ or ‘reflectivity’) and amount increased over the NW Indian Ocean and the NW and SW Pacific Ocean, whilst the opposite was observed for many mid-latitude areas (especially the North Atlantic) but also areas such as South Africa and California. This can be observed in the following image which was produced as part of their research.
Image Credit Nature
Whether clouds increased or decreased, this has a dramatic impact on space-borne optical imagery. In the areas where cloud totals decreased, it has become easier to obtain a cloud-free image. This is good news for the majority of applications which use satellite imagery, as cloud cover is often a limiting factor governing which imagery can actually be used.
However, for the areas where cloud totals have increased, obtaining an optical cloud-free, and therefore ‘useful’, image is more difficult. There may be prolonged periods of time where persistent cloud cover restricts the acquisition of imagery. This is particularly problematic in inaccessible regions, where space-borne observations are the only way to see what is happening in an area.
So what are the options for dealing with increased cloud? Well one option is the use of Synthetic Aperture Radar (SAR) data, which has the ability to ‘see through’ clouds and image the area underneath. The acquired SAR imagery is not the same as from other ‘optical’ satellites, yet can be used in a multitude of applications from flood mapping to deforestation detection. With the recent launch of Sentinel-1 and numerous other Synthetic Aperture Radar (SAR) satellites, this is a good option for overcoming the increase in clouds which may be observed in your area of interest. Additionally, there is a surge of recent interest in ‘Image Fusion’, essentially using SAR data to fill in the gaps where clouds are present in optical imagery. Reiche and colleagues present one such method to do this successfully and state that it is a fundamental process for overcoming persistent cloud cover. The methods are however very novel, and therefore not the easiest to employ! Another option could be the use of drones. They can fly beneath the clouds and will therefore not be affected, though their limited range and inability to fly during adverse weather conditions may restrict their exploitation in many large-scale studies.
In summary, the global change in cloud distribution is an interesting one for the EO community. It can be seen as beneficial in some areas, yet detrimental in others. It may change the way that things have been done in the past and we may see an increase in the use of novel techniques like SAR and optical image fusion.