Forest fires release very much CO2, they cause the biggest natural emissions of greenhouse gas (Yue & Gao, 2018), and they destroy carbon reducers which are important for climate regulation. In particular the burning down of rain forests emits large amounts of carbon into the atmosphere. At the same time it reduces the amount of carbon dioxide the rain forests take out of the atmosphere and absorb in their biomass. Large vegetation fires in the tropical rain forest contribute significantly to the global greenhouse effect (Seiler & Crutzen, 1980; Bowman et al., 2018). But fires also threaten species-rich jungles, and they are a threat for settlements near forests. This is a good enough reason to observe such fires more closely. “FireBIRD” is the name of the thus related mission of the German Aerospace Centre – Deutsches Zentrum für Luft- und Raumfahrt (DLR). The mission concerns the TET-1 and BIROS satellites, the latter being still active.
The satellites of the FireBIRD mission are powerful yet comparably cheap small satellites – thus also heralding a new era of space travelling. Each of them has just the size of a washing machine. A look behind the scene reveals much about satellites as such and about new trends of satellite making.
Spotting fires from space is nothing new. For quite some time orbit-based systems have been active which are capable of identifying or detecting fires almost “live”. Also weather satellites may support the detection of active fires. Furthermore there are many satellite sensors which may be used for identifying already burnt areas. The data of these satellites, however, have the disadvantage that the assessment of the emissions caused by fires is only very inaccurate because “in retrospect” any assessment of the amount of burnt vegetational-organic material based on the size of burnt areas is necessarily only very rough. Such uncertainties when it comes to estimating emissions caused by vegetation fires are supposed to be reduced by way of improved observation by satellites.
The number of space sensors suitable for discovering active fires is limited. But even smaller is the number of sensors capable of a quantitative and “live” assessment of active fires, thus the numbers of those being capable of providing information about important qualities of fires or about “fire attributes” such as size, intensity or temperature. One very interesting fire attribute is the power radiated upwards by fires, the so called fire radiative power (FRP). The FRP is a measure for the intensity of vegetation fires such as forest or bush fires. About 15 years ago it was internationally recognized as a climate-relevant physical variable. Since about 10 years ago it has also been one of the so called essential climate variables (ECVs), that is the “fire disturbance” ECV. ECVs are physical, chemical or biological variables or a group of interconnected variables which contribute significantly to characterising the climate of the Earth. Currently there exist 54 of these “crucial” climate variables.
The introduction of such “crucial” climate variables resulted, among others, from the incoherence of data gained by way of Earth observation. Thus, application experts of the European space organisation ESA and other international institutions of Earth observation introduced a system which makes data comparable and allows for defining new sets of climate data for certain climate variables – indeed the ECVs. The data sets of these new climate variables are open and easily accessible. The “fire disturbance” ECV consists of three components: “burnt area”, “active fire detection” (localisation of active fires), and the above described “fire radiative power” (FPR), that is the radiation intensity of a vegetation fire.
If fire observation oriented at the “fire disturbance” climate variable is supposed to happen by help of a satellite and to cover all three components (burnt area, active fire, radiation power), the making of satellites and sensors is confronted with a complex task because all three aspects must be combined. It is thus important to watch the Earth by help of several “eyes” at the same time and to add up the different “views” to be able to effectively assess a fire.
The unique feature of the FireBIRD satellites is their capability of very close observation. They are real “fire magnifiers”. They deliver more refined data than other fire observation systems such as the MODerate Resolution Imaging Spectroradiometer (MODIS, NASA), the Visible Infrared Imaging Radiometer Suite (VIIRS, NOAA/NASA) or the Sea Land Surface Temperature Radiometer (SLSTR, European Copernicus Programme).
FireBIRD´s “fire magnifier” potential is used e. g. for the validation of radiation power or FRP values which are gained by help of the coarse-resolution MODIS, VIIRS, and SLSTR sensors of other satellites. Thus, the FireBIRD satellites are capable of having a closer look there where the other systems do not recognize the fire situation precisely enough or in a sufficiently “refined” way. Under favourable conditions, the sensor system of the FireBIRD satellites is even capable of detecting a larger camp fire.
The DLR´s orbital hawk eyes can even help with actual firefighting. One example is the use of FireBIRD photos during the devastating Paradise Fires in California in November, 2018. According to information by reinsurer Munich Re, this largest forest fire in Californian history caused a damage of 12 billion Dollars (Munich RE, 2019). This was the biggest insured loss worldwide in 2018. Still days after the devastating fires at Paradise more than 5,000 firemen were fighting the fires in the vicinity. At that time the DLR was able to give precise help, by way of daily updated maps of fires and their radiation energy. Derived from satellite data, it was possible to provide custom-made information (see Table 1). Among this there counted colour-scaled maps allowing for a swift, detailed and in each case updated view from space at fires and their temperatures as well as of actively burning areas.
Since November, 2019, BIROS observes selected fires in Australia. For fire observation, Australia reaches back mostly to the geo-stationary HIMAWARI satellite, with a resolution of about 1 kilometre. It is intended to compare, in cooperation with the Australian colleagues, the current FRP fire data provided by BIROS with those provided by HIMAWARI and to assess the latter more closely, e. g. concerning the derived FRP values.
Each satellite mission is thoroughly organised, to take care that it runs as smoothly as possible. Satellite missions are becoming cheaper, however still they require a considerable amount of resources which must be handled in a responsible way. Each satellite mission must be carefully planned and organisationally accompanied. Accordingly, each mission is structured into certain mission segments, independently of the size of the satellites. One distinguishes between the user segment, the start segment, the space segment, and the ground segment.
In case of scientific missions such as FireBIRD, the user segment is the scientific community which evaluates the data and uses them in the context of its research. Such missions always have a scientific head of mission, the “Principal Investigator”, in short: “PI”. As early as at the beginning of a mission the person holding this position defines the tasks of the different mission segments. Also, by way of “project reviews”, the PI takes care that his/her specifications are implemented as exactly as possible.
It is obvious that the start is an important segment of a mission´s course. To the start segment there also belongs the carefully chosen “launch service provider”, i. e. the provider of the space launcher and the launch site with the appropriate checkout rooms. In charge of the start of TET-1 was the space organisation of the Russian Federation, “Roskosmos” which, on July 12th, 2012, launched TET-1 “piggyback”, together with a main satellite, by a Soyuz rocket from the Baikonur space station in Kazakhstan. In June, 2016, BIROS was taken “piggyback”, together with a large Indian satellite, to the orbit by an Indian space launcher starting from Satish Dhawan Space Centre in India. Thus, in this case the “launch service provider” was the Indian ISRO space agency. Dependent on the elements of the start segment is e. g. the satellite´s initial orbit, from where it navigates to its actual orbit.
The space segment includes all aspects of the satellite´s movement in space. The TET-1 and BIROS fire observation satellites circle, as mini-constellations, around the Earth on polar orbits. This has the advantage that the Earth rotates away from them and they cover even high latitudes. The range of vision of the satellite sensors allows for a more or less broad recording of the terrestrial surface. Among the ground segment there count data reception and mission control. Among the latter there count the technical supervision of the satellites, their control by help of transmitting antennas and the data reception by help of receiving antennas.
Basically, a satellite is kind of space bus, transporting different kinds of payloads in to orbit around the Earth. And like any bus being ready to use is entered by passengers, also a satellite bus is “entered” by payloads. Accordingly, the expert term for the basic structure of a satellite is “satellite bus”, among others. The satellite bus provides the basic structure for different kinds of systems allowing for running the satellite and its payloads in space.
Such a satellite bus is equipped with sub-systems which are important for navigation, communication, energy supply, thermal control, as well as one or several board computers for position control, and computers for on-board data processing as well as for maintaining “on-board autonomy”. The sub-system for position control allows for turning around the satellite as a whole and, for example, for aiming the camera at a certain target on Earth. To this position-control system there belong star cameras and solar sensors. “On-board autonomy” takes care that the satellite need not be permanently controlled from the ground. The TET-1 and BIROS FireBIRD satellites are based on one and the same satellite bus, i. e. from the outside they look very similar to each other.
The “TechnologieErprobungsträger-1“, made by Astro- und Feinwerkstechnik GmbH in Berlin-Adlershof (in short: “Astrofein“) in cooperation with other companies, was given this name because initially it was exactly this: an orbital platform by help of which different technologies boarded as payloads were supposed to be tested. After, in 2013, this eleven-years test period had been over, the TET-1 satellite was exclusively used for running the sensor system for fire observation.
The same sensor system is on board of the BIROS satellite which was made at the DLR´s Institut für Optische Sensorsysteme in Adlershof. BIROS means Berlin InfraRed Optical System. Both the Astro- und Feinwerkstechnik GmbH and the DLR Institute are in the immediate vicinity of the academic site of Berlin-Adlershof. TET-1 was launched in 2012 and has been used by the DLR since 2013, in the context of the FireBIRD mission. BIROS was launched in 2016 and since then forms a mini-constellation with TET-1.
Both satellites are mostly, although not completely, clad in “golden coats”. These coats consist of 23 layers of thermal blankets and provide a super-isolation, to take care that in the satellite´s interior the temperature is as steady as possible. For further thermal regulation there are heat-emission or radiator surfaces without coats in those parts as not being exposed to the sun. There, the heat produced in the interior by running the satellite itself can be emitted. Thus, these satellites look somewhat like Christmas presents wrapped in golden foil – the folded-out solar arrays for energy supply looking like the “ribbons”.
Only its payload makes a satellite meaningful. Otherwise it would just be an empty transport system in the orbit. The three-elements sensor system of TET-1 and BIROS is characterised by observing the Earth by spectral bands which are important for fire observation, thus providing data which can be depicted on a map.
To allow for fire observation from the orbit, several important technological problems must be solved. Basically, fire observation happens by identifying a difference in the infrared range between a fire and the non-burning environment. A fire sensor aiming at the Earth works similar to a digital camera – only that it works within the range of infrared wavelengths. For such a satellite sensor the medium-infrared difference between the radiation caused by fires and that of the neighbouring, non-burning background is so big that a pixel for which the fire makes only 1 per cent of the surface provides a much stronger signal than a neighbouring background pixel without a fire. The medium infrared spectral band, where a big heat difference can be observed, is also called MIR band (MIR = Medium InfraRed). The MIR signal of a pixel of 300 m x 300 m on the ground, identifying a fire front of 300 m width, is clearly different from those pixels untouched by the fire. Has thus the fire been reliably identified? Well, not really.
One problem resulting from fire observation from space is “false alarm”, which is also known from Earth. False alarms in the context of fire observation are caused on the one hand by sand or rocky surfaces heated up by the sun and on the other by the sun being reflected by water surfaces. Both kinds of false alarm cause a pixel signal within the MIR band which is like that of a sub-pixel fire.
To solve this problem one makes use of measurements within other spectral bands. Of particular interest in this context is one certain thermal infrared band, the so called TIR Band. Making use of the TIR band allows for recognizing heated surfaces as false alarms. Furthermore, by help of the TIR Band fires can be assessed more exactly, i. e. the bi-spectral MIR/TIR sensor provides data which are necessary for concluding on different attributes of a fire, such as area, temperature and FRP. Thus, one does not only see that that there is a fire, but one it also informed about physical details of that fire.
To distinguish fires from sunglints, a third spectral band is needed which must either be in the visible red spectral range (VIS-red) or in the “Near InfraRed/NIR” and whose signal is related to the MIR Band. Furthermore, the NIR Band allows for easily recognizing clouds in daylight, which is important, as fire analysis always happens in cloud-free areas. Both TET-1 and BIROS are provided with a sensor system (Fig. 1, No. 8) covering the most important bands (MIR, TIR, VIS, NIR) while at the same time being capable of depicting all of them at the same time and the same place on Earth. This as well as the spatial resolution of ca. 300 m allows for fire observation with hawk's eyes from a height of 500-600 km.
From what has been said above we already understand that TET-1 and BIROS see the Earth in a way which is completely different from that of an astronaut looking down from ISS, whose eyes only perceive reflected sunlight of a wavelength of 0.4 to 0.6 micrometres. Fig. 5 shows spectral diagrams of four different sources, the radiation intensity coming up from the Earth in each case being depicted according to the radiation wavelength (λ = „Lambda“). Spectral diagrams are also used in astronomy, for example to analyse solar spectra. Thus, the figure depicts the radiation intensities of four different objects as a curve as it can be “seen” from above our atmosphere – and through its “radiation windows”. Highlighted in grey are the four spectral bands of satellite sensor technology which are of particular interest for fire assessment.
The red dashed curve gives the radiation density of a fire, making 1% of a sensor pixel (the remaining 99% are not burning vegetation). The yellow dotted curve gives the radiation density of a sun reflex, making 1% of a sensor pixel (the remaining 99% are water surfaces reflecting the sun). The white dashed-dotted curve gives the radiation density of sandy or rocky ground heated up to 320 K or 47° C, and the green curve gives the radiation density of non-burning vegetation with a temperature of 300 k or 27° C.
The reader should not be confused by the deep “valleys” of the radiation courses which are also visible in the diagram. They come from the fact that in certain wavelengths the atmosphere absorbs radiation. These absorption lines are due to the chemical composition of our planet´s atmosphere and are most of all caused by steam. They do not play any significant role for fire assessment.
Another advantage of the sensor system of the FireBIRD satellites is that they cover an extremely wide range of the dynamics of incoming signals. Many other IR sensors of satellites are capable of detecting a fire, but they do not allow for providing detailed information about fire attributes such as a fire´s temperature, burning area, length and intensity of fire fronts or radiancy, because they are “saturated” and thus “cut” or “cap” the signal. Consequently, the saturation of the sensor system distorts the signal and thus also any information about the intensity of a fire the satellite might transmit. There are only a few satellite sensors which are not saturated in case of extended fires with temperatures of more than 1,000 Kelvin (= 727° C) while at the same time giving the temperature of the “cold” area around a fire with an accuracy of at least half a degree. However, such a degree of resolution is needed to define the upward radiancy of a fire and to thus-based estimate the carbon emission of a fire with much more accuracy than possible from assessing burnt areas. Accurate information about carbon dioxide emissions is relevant particularly for climate models.
One particularity of the BIROS satellite is that, as a “mobile payload”, it carried the BEESAT 4 mini-satellite of Technische Universität Berlin. This is a so called Cube Sat which, by help of a special mechanism, was released in the orbit. This small Cube satellite has a size of just about 10cm x 10cm x 10 cm. BEESAT 4 was used for the DLR´s so called AVANTI experiment in the context of the FireBIRD mission, in the course of which BIROS moved away from BEESAT 4, to then come closer again. The approach to BEESAT 4 was much more successful than originally intended – BIROS happened to come as close as 40 m to the mini-satellite again. Releasing such a miniature satellite under conditions of zero gravity is no easy manoeuvre, among others as thus-connected forces may make the released object spin around.
Another challenge was the constant and independent tracking of the mini-satellite by help of onboard cameras. To keep its accompanying satellite in sight, BIROS had to be able to independently change both its orientation and its speed and to adjust both to the mini-satellite´s course and speed.
The TET-1 and BIROS satellites were planned in the context of an “affordable space mission” project of the DLR. This strategy is also typical for many privately owned space enterprises and their business models, which are currently edging into the market and intend to make money from space travel. Famous are in particular US-American enterprises such as Elon Musk´s SpaceX or Jeff Bezos´s Blue Origin.
For example, for its communication with the additional BEESAT satellite BIROS is provided with an antenna fixed to the exterior of the satellite. It is simply a common measuring tape made of steel which was purchased at a DIY market in Berlin but works quite extraordinarily. In much generalising terms this means that costs are reduced by not using the most expensive elements. Completely “space-proof”, tested elements are much more expensive than so called “commercial off the shelf” elements (“COTS”). Of course, making use of COTS elements increases the risks of a satellite´s functioning in the orbit. These risks are minimised by way of intensive pre-tests under space conditions, by way of redundancy concepts as well as by intelligent and autonomous onboard algorithms.
Furthermore, both FireBIRD satellites are shipped “piggyback”, as additional payloads, to the much bigger main satellites in the orbit. This is a common and cost-saving method. As an addition to a main payload, such as a huge, heavy satellite, the space launcher also takes minor payloads onboard, to make optimum use of its transport capacity and the possible launching mass and thus also of the invested resources.
Cost-saving yet high-performance miniature satellites – this is one of the trends of recent space travel, also called “NewSpace”, to increase its advertising appeal. Thus seen, the FireBIRD mission proves to be important not only when it comes to its contribution to Earth observation and the detecting of fires and their carbon dioxide emissions. Rather, it is a good example of a clearly visible structural change in satellite construction. This change is characterised by private enterprises with new business models joining the sector of space travel, by the increasing cooperation of state-funded research institutions and private economy as well as by much more cost-saving missions with small and miniature satellites.
Astro- und Feinwerkstechnik GmbH. (2010, 14. Juli). Satellite TET-1 (Issue 1) [www.astrofein.com]. Opened on 22.11.2019.
Bowman, D. M. J. S., Balch, J .K., Artaxo, P., Bond, W. J., Carlson, J. M., Cochrane, M. A. ... Pyne, S. J. (1999). Fire in the Earth System. Science, 324(5926), 481-484. doi:10.1126/science.1163886
Booklet: FireBIRD. A DLR satellite system for forest fires and early fire detection. (o.J.). Deutsches Zentrum für Luft- und Raumfahrt – DLR, Oberpfaffenhofen.
Dozier, J. (1981). A method for satellite identification of surface temperature fields of subpixel resolution. Remote Sensing of Environment, 11, 221-229. doi:10.1016/0034-4257(81)90021-3
Goldammer, J. G. (Ed.). (2013). Vegetation Fires and Global Change. Challenges for Concentrated International Action. A White Paper directed to the United Nations and International Organizations (Global Fire Monitoring Center – GFMC). Remagen-Oberwinter: Kessel Publishing House.
Hollmann, R., Merchant, C. J., Saunders, R., Downy, C., Buchwitz, M., Cazenave, A., ... Wagner, W. (2013). The ESA Climate Change Initiative: Satellite Data Records for Essential Climate Variables. Bulletin of the American Meteorological Society, 94(10), 1541-1552. doi:10.1175/BAMS-D-11-00254.1
Ley, W., Wittmann, K. & Hallmann, W. (Eds.). (2019). Handbuch der Raumfahrttechnik (5., updated and extended edition). München: Hanser.
Lorenz, E., Halle, W., Fischer, C., Mettig, N. & Klein, D. (2017). Recent Results of the Firebird Mission (37th International Symposium on Remote sensing of Environment, 8-12 may, Tshwane, South Africa). The International Archives of Photogrammetry, Remote sensing and Spatial information Sciences, 42(4/w6), 105-111.
Maibaum, O., Montenegro, S. & Terzibaschian, T. (2008). Robustness as Key to Success for Space Missions. In A. Schuster (Hrsg.), Robust Intelligent Systems (S. 232-248). London: Springer. doi:10.1007/978-1-84800-261-6_11
Munich RE. (2019, 1. August). Extreme storms, wildfires and droughts cause heavy nat cat losses in 2018 [www.munichre.com]. Opened on 11.12.2019.
Oertel, D. (2001). Fernerkundung von Feuern: Satelliten-Monitoring von Hoch-Temperatur-Ereignissen. Raumfahrt Concret, (4), 15-17.
Oertel, D., Briess, K., Lorenz, E., Skrbek, W. & Zhukov, B. (2002). Fire Remote Sensing by the Small Satellite on Bi-spectral Infrared Detection (BIRD). Photogrammetrie, Fernerkundung & Geoinformation, (5), 341-350.
Robinson, J. M. (1991). Fire from space: Global fire evaluation using infrared remote sensing. International Journal of Remote Sensing, 12(1), 3-24. doi:10.1080/01431169108929628
Seiler, W. & Crutzen, P. J. (1980). Estimates of gross and net fluxes of carbon between the biosphere and the atmosphere from biomass burning. Climatic Change, 2(3), 207-247. doi:10.1007/BF00137988
Wooster, M. J., Xu, W., & Nightingale, T. (2012). Sentinel-3 SLSTR active fire detection and FRP product: Pre-launch algorithm development and performance evaluation using MODIS and ASTER datasets. Remote Sensing of Environment, 120, 236-254. doi:10.1016/j.rse.2011.09.033
Yue, X.-L, Gao, Q.-X. (2018). Contributions of natural systems and human activity to greenhouse gas emissions. Advances in Climate Change Research, 9(4), 243-252. doi:10.1016/j.accre.2018.12.003
Zhukov, B., Briess, K., Lorenz, E., Oertel, D. & Skrbek, W. (2005). Detection and analysis of high-temperature events in the BIRD mission. Acta Astronautica, 56(1-2), 65-71. doi:10.1016/j.actaastro.2004.09.014
Zhukov, B., Lorenz, E., Oertel, D., Wooster, M. & Roberts, G. (2006). Spaceborne detection and characterisation of fires during the bi-spectral infrared detection (BIRD) experimental small satellite mission (2001 – 2004). Remote Sensing of Environment, 100, 29-51. doi:10.1016/j.rse.2005.09.019
EOWEB® GeoPortal (EGP) is a multi-mission earth observation data portal of DLR. Earth observation data from the German Satellite Data Archive (D-SDA) can be ordered here. The FireBIRD data are mainly received from DLR's ground stations in Neustrelitz and are processed, archived and made available for worldwide scientific use at DLR's German Remote Sensing Data Center. The FireBIRD satellites are operated and controlled by the German Space Operations Center (GSOC) of DLR's Space Operations facility in Oberpfaffenhofen.
Pyrocene: „We must live with fire“ (Interview with Johann Goldammer in the Süddeutsche Zeitung)
Webseite of the German Aerospace Center DLR about the FireBIRD mission
Published: 24.12.2019, Vol. 6.
Cite as: Oertel, D., Terzibaschian, T. & Halle, W. (2019, December 24th). FireBIRD sees forest fires with the eyes of an hawk. Earth System Knowledge Platform [www.eskp.de], 6. doi:10.2312/eskp.007