| Glossary | Technologies | Earth observation

Earth observation

Definition and delimitation

Human-induced environmental and climate change is transforming our planet at a breathtaking pace. Rapid settlement expansion, deforestation, the spread of agro-ecosystems, the extraction of raw materials, but also changes in land and water temperatures and the melting of ice sheets and glaciers are just some of the socially relevant dynamics that present us with challenges and attract a great deal of media attention.

Against this backdrop, earth observation – usually satellite-based – allows these processes to be recorded, quantified and visualised in an attractive way. The term earth observation is often used synonymously with the term remote sensing. It describes the collection of information about the earth’s physical, chemical and biological systems. This is mostly done with the help of satellite sensors, but also occasionally with the help of aircraft-based sensors. In particular, the large number of different satellite sensors enables continuous and large-scale recording of the earth’s surface, even in wavelength ranges that the human eye cannot detect. This provides us with unique data and a view of our Earth that helps us to record, quantify and understand the dynamics of Earth systems. Satellites, some with a daily observation interval, have been recording data for three to four decades, making it possible to derive long-term trends (e.g. temperature data from land and oceans) and forecast future developments.

History

Systematic observations of the Earth’s surface date back to the early 20th century, when aerial photographs were taken for military purposes. From the middle of the century, the first aircraft-borne radar systems and weather satellites were deployed, followed by the first optical remote sensing satellite in the early 1970s. Since then, the number of public and private earth observation satellites has increased continuously and at a rapid pace, so that in recent decades a large number of satellites have recorded a gigantic treasure trove of data that is constantly growing. At the same time, the technical capabilities of the sensors have improved with each new generation of satellites, so that the Earth’s surface can now be recorded at centimetre level and along large parts of the electromagnetic spectrum. Thanks to a series of public earth observation missions (e.g. the European Copernicus programme) and the opening up of data archives, an enormous amount of earth observation data is available, the full potential of which must be exploited.

Application and examples

As earth observation provides data with high temporal and spatial resolution, an extraordinarily diverse field of application has developed. This ranges from small-scale, local analyses to data processing on a global scale. Along the time axis, earth observation data allows us to uncover the dynamics of recent decades as well as to make predictions for the future. In particular, the high repetition rate at which the earth’s surface is recorded has led to a shift from mono- or bitemporal analyses to the evaluation of dense satellite image time series.

Thematically, broad areas of application such as environmental and nature conservation, climate (change) research, disaster prevention, weather forecasting and security can be distinguished. Earth observation provides quantitative information about our settlement areas, agricultural and forest ecosystems, the Earth’s water resources and also makes it possible to analyse remote areas such as the polar regions.

In detail, earth observation is used to observe dynamics in specific systems. For example, the continuous availability of satellite data made it possible to analyse forest damage in several years of drought for the whole of Germany (Thonfeld et al. 2022). In the polar region of Antarctica, on the other hand, radar-based Earth observation enables, among other things, the automatic extraction of glacier fronts to better understand the contribution of ice sheets to sea level rise (Baumhoer et al. 2019).

In addition to the dynamics of the natural land surface, global satellite data allow the detection and quantification of human presence on our planet in the form of a global urban footprint, which records the locations and distribution of human settlements (Esch et al. 2017).

Criticism and problems

One problem with purely satellite-based Earth observation is the validation of results. Complementary data collected in situ is necessary to verify the accuracy of observations. This is the only way to rule out misinterpretations of earth observation data.

Criticism in the field of earth observation arises, among other things, with regard to ethical values. In particular, the constantly improving resolution of satellite images raises questions about privacy, transparency, fairness and responsibility in the use of earth observation data. Sensitisation to the use and misuse of earth observation and the development of guidelines for appropriate handling will occupy the field in the future.

Another point of criticism is the use of so-called miniature satellites. Instead of one large satellite, many smaller ones are put into orbit. This increases the already high density of future space debris.

Research

The enormous number of fields of application gives rise to an equally large number of research areas in which earth observation plays a key role. In particular, questions on a global scale can now be answered from and with the perspective of satellite-based earth observation in order to explain the complex dynamics of earth systems. This goes hand in hand with the research and development of computing resources and algorithms that enable data to be analysed at this level.

As part of the bidt project “ROOT”, near-real-time earth observations of the forest in Bavaria are analysed in order to detect damage to the forest stand and support sustainable forest management.

Sources

Baumhoer, C. A. et al. (2019). Automated Extraction of Antarctic Glacier and Ice Shelf Fronts from Sentinel-1 Imagery Using Deep Learning. In: Remote Sensing 11 (21), 2529.

Esch, T. et al. (2017). Breaking new ground in mapping human settlements from space – The Global Urban Footprint. In: ISPRS Journal of Photogrammetry and Remote Sensing 134, 30–42.

Thonfeld, F. et al. (2022). A First Assessment of Canopy Cover Loss in Germany’s Forests after the 2018–2020 Drought Years. In: Remote Sens., 14, 562.