Although the described AP monitoring system is still just a proof of concept, there are already several directions I can point at that could be targeted for improvement.
The first and more prominent is definitely the generalisation to three dimensions.
The system was intentionally designed to perform two-dimensional tomographic re-constructions. However, there is no reason why it cannot be extended to the third dimension.
Another improvement that can be pursued is the use of an artificial light source.
This would be a fairly complex alteration, since it would probably entail the introduc-tion of a second drone unit. As the rooftop experiment has shown, alignment is crucial.
If a second drone is in fact included, a lot of effort must be directed towards ensuring this is not a problem.
In its current design state, the system cannot be used in a truly autonomous fashion.
There must always be a human in the loop, even if just to trigger the drone to perform a measurement cycle. This makes it a poor continuous measurement system. A truly useful device, on a day-to-day basis, would be programmable, and capable of operating on its own. This would require another improvement: the development of an automatic take-off, landing and charging system.
The final general improvement I would like to bring forth is different from the others, as it is the only one that does not imply hardware changes. Integrating the proposed system with a Geographic Information System (GIS). An integration with a GIS platform would be a very important development. It is very common, nowadays, for
6 . 4 . P R O D U C T I M P R OV E M E N T S
the kind of entity that might be interested in purchasing an Air Pollution monitoring system such as this to have a lot of data in a GIS. The ability to cross their data with real time atmospheric pollution information can be of relevance to their management teams, and also a very convincing selling point.
On a more particular note, and one on which I have spent much more time than any of the aforementioned improvements, I would like to discuss some changes to the Tomosim software. Just as the DOAS library, which is described in Section 4.1.3, the Tomosim simulation was programmed using the OOP paradigm. And also like the DOAS library, the SOLID principles of OOP were generally observed. The global flowcharts of the software’s operation can be viewed in Figure 4.9 and Figure 4.10.
As the application grew and I continued to work on it, it became quite clear that the original architecture was compromising the tool’s performance and the code would need to be heavily overhauled. Although the time-frame of this work did not allow this, I was able to devise a basic new architecture for this software, which is represented in Figure 6.2.
Experiment +gamma: int +delta: int -system_matrix: np.lilmatrix -create_system_matrix +run
+eval
CrossSection +name: str -convolve -interpolate
+instrument_function_convolution Spectrometer +spectral_resolution +calibration_file +calib_coeffs -calibration_data -window -load_calibration_data +calibration_window
Reconstructor +name: str +rules: dict 1:n +run
1:1
1:n Phantom
+cs: CrossSection +density: Float +spectrometer: Spectrometer
1:n Space
size: Tuple[int] or int -space: np.meshgrid
1:1
RecEvaluator +name: str +eval
1:n
Figure 6.2: First draft of the UML diagram for the new architectural model of the Tomosim library. Note the usage of several classes from the DOASlibrary described in Section 4.1.3.
Ideally, Tomosim would be an extension of the more generic DOAS library. The only reason why it is not is purely historical: Tomosim’s development was started be-fore the former library. Therebe-fore, the first part of any refactoring exercise should be to mend this error. The new architecture is characterised by an increase in modular-ity. This is achieved through the introduction of the new classes "Reconstructor"and
"RecEvaluator", which are composed into an also new class called "Experiment". The introduction of the two first classes greatly increases Tomosim’s flexibility, as one would now be allowed to introduce any new reconstruction technique or evaluation
CH A P T E R 6 . C O NC LU D I NG R E M A R K S A N D F U RT H E R D E V E L O P M E N T S
method, without ever having to touch code already in place. To ensure this functional-ity, it is highly advisable that these two classes are built using the Template Design Pattern [37], with structural enforcing in the form of key abstract methods. This new design also favours the implementation of a mix between the factory design pattern and the strategy design pattern. This would result in factory methods being used to create Experiment objects based on some input parameters that would determine the types of reconstruction and / or evaluation performed by said object.
Improving the system’s architecture is a necessary and very important develop-ment, but the refactoring effort must also comprehend the performance side of the application. Currently, Tomosim’s spends most of its running time discretising the ROI. This is expected. However, since Siddon’s algorithm is not implemented using arrays, but instead makes heavy use of Python lists and for-loop-based iterations, it is absolutely imperative that this component is optimised
B i b l i o g r a p h y
[1] 2 Degrees Institute.CO2 Levels. 2020. u r l: https://www.co2levels.org/ (visited on 04/14/2020) (cit. on p. 11).
[2] A. P. do Ambiente.QualAR. 2019. u r l: https://qualar.apambiente.pt/ (visited on 09/20/2021) (cit. on pp. 88–90).
[3] Amphenol Optics Fiber.SMA Connectors , Adapters & Receptacles. 2002 (cit. on p. 58).
[4] A. Andersen. “Simultaneous Algebraic Reconstruction Technique (SART): A superior implementation of the ART algorithm”. In: Ultrasonic Imaging 6.1 (Jan. 1984), pp. 81–94. i s s n: 01617346. d o i: 10.1016/0161- 7346(84)90008- 7.
u r l: https:// linkinghub.elsevier.com/ retrieve/ pii/ 0161734684900087 (cit. on p. 49).
[5] M. N. Asl and A. Sadremomtaz. “Analytical image reconstruction methods in emission tomography”. In: Journal of Biomedical Science and Engineering 06.01 (2013), pp. 100–107. i s s n: 1937-6871. d o i: 10.4236/jbise.2013.61013. u r l:
http://www.scirp.org/journal/doi.aspx?DOI=10.4236/jbise.2013.61013 (cit. on p. 46).
[6] Avantes. AvaSpec-Mini: Small and powerful OEM spectrometer. 2019. u r l:
https://www.avantes.com (cit. on pp. 31, 32).
[7] C. Beekman. “Differential Optical Absorption Spectroscopy of Trace Gas Species and Aerosols in the Upper Ohio River Valley”. PhD thesis. Ohio State University, 2010 (cit. on pp. 40, 41).
[8] M. L. Bell, D. L. Davis, and T. Fletcher. “A retrospective assessment of mortality from the london smog episode of 1952: The role of influenza and pollution”. In:
Urban Ecology: An International Perspective on the Interaction Between Humans and Nature6.1 (2008), pp. 263–268. d o i: 10.1007/978-0-387-73412-5_15 (cit. on p. 4).
[9] P. Bourdeau.The Dobris Assessment. Tech. rep. Copenhagen: European Envi-ronment Agency (cit. on pp. 5, 6).
B I B L I O G R A P H Y
[10] T. Bourdrel et al. “Cardiovascular effects of air pollution”. In: Archives of Car-diovascular Diseases 110.11 (Nov. 2017), pp. 634–642. i s s n: 18752136. d o i:
10.1016/j.acvd.2017.05.003. u r l: https://linkinghub.elsevier.com/retrieve/pii/S1 875213617301304 (cit. on p. 8).
[11] B. W. H. Brattain and J. Bardeent. “The Bell System Surface Properties of Germanium”. In:American Telephone and Telegraph Company32 (1952), pp. 1–
41 (cit. on p. 24).
[12] R. T. Brinkmann. “Rotational Raman scattering in planetary atmospheres”. In:
Astrophys J154 (1968), pp. 1087–1093. i s s n: 0004-637X. d o i: 10.1086/149827 (cit. on p. 34).
[13] R. D. Brook. “Cardiovascular effects of air pollution”. In:Clinical Science115.5-6 (2008), pp. 175–187. i s s n: 01435221. d o i: 10.1042/CS20070444 (cit. on p. 8).
[14] P. P. Bruyant. “Analytic and iterative reconstruction algorithms in SPECT.” In:
Journal of nuclear medicine : official publication, Society of Nuclear Medicine 43.10 (2002), pp. 1343–58. u r l: http://www.ncbi.nlm.nih.gov/pubmed/12368373 (cit. on p. 49).
[15] P. P. Bruyant. “Analytic and iterative reconstruction algorithms in SPECT.” In:
Journal of nuclear medicine : official publication, Society of Nuclear Medicine 43.10 (2002), pp. 1343–58. i s s n: 0161-5505. u r l: http://jnm.snmjournals.org/
cgi/content/abstract/43/10/1343%7B%5C%%7D5Cnhttp://jnm.snmjournals.org/
cgi/content/full/43/10/1343%7B%5C%%7D5Cnhttp://www.ncbi.nlm.nih.gov/
pubmed/12368373 (cit. on pp. 44, 49).
[16] M. Buchwitz et al. “Global carbon monoxide as retrieved from SCIAMACHY by WFM-DOAS”. In: Atmospheric Chemistry and Physics 4.7 (Sept. 2004), pp. 1945–1960. i s s n: 16807316. d o i: 10.5194/ acp- 4- 1945- 2004. u r l: www.
atmos-chem-phys.org/acp/4/1945/ (cit. on p. 39).
[17] M. Buchwitz, V. V. Rozanov, and J. P. Burrows. “A near-infrared optimized DOAS method for the fast global retrieval of atmospheric CH 4 , CO, CO 2 , H 2 O, and N 2 O total column amounts from SCIAMACHY Envisat-1 nadir radiances”. In: Journal of Geophysical Research: Atmospheres105.D12 (June 2000), pp. 15231–15245. i s s n: 01480227. d o i: 10 . 1029 / 2000JD900191. u r l:
http://doi.wiley.com/10.1029/2000JD900191 (cit. on p. 39).
[18] CABI. Air Pollution: Sources, Impacts and Controls. Ed. by P. Saxena and V.
Naik. 1st. Vol. 53. 9. Oxfordshire, UK: CAB International, 2019, pp. 1689–1699.
i s b n: 9781786393890 (cit. on pp. 8, 11–14).
[19] N. Casaballe et al. “Comparison Between Regularized Optimization Algorithms for Tomographic Reconstruction of Plume Cross Sections in the Atmosphere”.
In: Earth and Space Science 4.12 (Dec. 2017), pp. 723–736. i s s n: 23335084.
d o i: 10.1002/2017EA000341. u r l: http://doi.wiley.com/10.1002/2017EA000341 (cit. on p. 20).
B I B L I O G R A P H Y
[20] K. Chance and R. L. Kurucz. “An improved high-resolution solar reference spectrum for earth’s atmosphere measurements in the ultraviolet, visible, and near infrared”. In:Journal of Quantitative Spectroscopy and Radiative Transfer 111.9 (2010), pp. 1289–1295. i s s n: 00224073. d o i: 10.1016/j.jqsrt.2010.01.036.
u r l: http://dx.doi.org/10.1016/j.jqsrt.2010.01.036 (cit. on p. 43).
[21] A. G. Clark. Industrial Air Pollution Monitoring. 1997. i s b n: 9789401071437.
d o i: 10.1007/978-94-009-1435-3 (cit. on pp. 23, 24).
[22] CREE.XLamp ® XHP50.2 LEDs Datasheet. 2021 (cit. on p. 72).
[23] T. Danckaert et al. QDOAS. 2015. u r l: http://uv-vis.aeronomie.be/software/
QDOAS/ (cit. on pp. 34, 40, 62).
[24] M. Defrise, P. E. Kinahan, and C. J. Michel. Positron Emission Tomography. Ed. by D. L. Baley et al. London: Springer London, 2005, pp. 63–91 (cit. on pp. 44, 49, 81).
[25] L. A. Dewerd and M. Kissick.The Phantoms of Medical and Health Physics. Ed.
by L. A. DeWerd and M. Kissick. Biological and Medical Physics, Biomedical Engineering. New York, NY: Springer New York, 2014, p. 286. i s b n: 978-1-4614-8303-8. d o i: 10.1007/978-1-4614-8304-5. u r l: http://www.springer.com/series/3 740%20http://link.springer.com/10.1007/978-1-4614-8304-5 (cit. on p. 67).
[26] S. Dhall et al. “A review on environmental gas sensors: Materials and technolo-gies”. In: Sensors International2.May (2021), p. 100116. i s s n: 26663511. d o i:
10.1016/j.sintl.2021.100116. u r l: https://doi.org/10.1016/j.sintl.2021.100116 (cit. on pp. 23, 24).
[27] Diamond FIber Optics.F-SMA Standard Datasheet(cit. on p. 58).
[28] DJI.DJI E1200 Pro Tuned Propulsion System. 2015. u r l: https://www.dji.com/
pt/e1200 (visited on 12/28/2020) (cit. on p. 58).
[29] EEA.Air pollution in Europe 1990–2004. Tech. rep. 2. 2007 (cit. on p. 7).
[30] EEA. Air quality in Europe. Tech. rep. 28. 2016, p. 83. d o i: 10.2800/ 80982 (cit. on pp. 5, 7).
[31] EEA.European Environmental Agency. 2019. u r l: https://www.eea.europa.eu (visited on 09/23/2019) (cit. on pp. 14, 15, 22, 27).
[32] EEA. Copernicus Atmosphere Monitoring Service. 2021. u r l: https://regional.
atmosphere . copernicus . eu / index . php ? category = ensemble % 7B % 5C & %7 Dsubensemble=hourly%7B%5C_%7Densemble%7B%5C&%7Ddate=LAST%7 B%5C&%7Dcalculation-model=ENSEMBLE%7B%5C&%7Dspecies=o3%7B%5 C & %7Dlevel = SFC % 7B % 5C & %7Doffset = 000 % 7B % 5C & %7DTSPD % 7B % 5 C_%7D101%7B%5C_%7DR0=081a6daeeeab200024e9c0dbba34cbc62ccaa638c6 559cde56946a3607065386fc2daaf0f729 (cit. on p. 27).
[33] D. M. Etheridge et al. Historical CO2 record from the Law Dome DE08, DE08-2 and DSS ice cores. Tech. rep. Aspendale, Victoria, Australia: Division of Atmospheric Research, CSIRO. u r l: http : / / cdiac . ornl . gov / ftp / trends / co2 /lawdome.combined.dat (cit. on p. 11).
B I B L I O G R A P H Y
[34] M. Fox. Quantum optics. Oxford, UK: Oxford University Press, 2006. i s b n:
9780198566724 (cit. on p. 74).
[35] P. Frank and M. A. Ottoboni. The Dose Makes The Poison: A Plain Language Guide to Toxicology. 3rd Editio. Hoboken, New Jersey: John Wiley & Sons, Inc, 2011. i s b n: 9780470381120 (cit. on p. 7).
[36] E. Frins et al. “Tomographic multiaxis-differential optical absorption spec-troscopy observations of Sun-illuminated targets: a technique providing well-defined absorption paths in the boundary layer.” In: Appl. Opt. 45.24 (2006), pp. 6227–6240. i s s n: 0003-6935. d o i: 10.1364/AO.45.006227 (cit. on pp. 20–22, 38).
[37] E. Gamma et al. Design Patterns: Elements of Reusable Object-Oriented Soft-ware. Addison-Wesley Professional, 1994. i s b n: 0201633612 (cit. on p. 96).
[38] F. C. Goldizen, P. D. Sly, and L. D. Knibbs. “Respiratory effects of air pollution on children”. In:Pediatric pulmonology51.1 (2016), pp. 94–108. i s s n: 10990496.
d o i: 10.1002/ppul.23262 (cit. on p. 8).
[39] J. F. GRAINGER and J. RING. “Anomalous Fraunhofer Line Profiles”. In:
Nature193.4817 (Feb. 1962), pp. 762–762. i s s n: 0028-0836. d o i: 10.1038/19376 2a0. u r l: http://www.nature.com/doifinder/10.1038/193762a0 (cit. on p. 34).
[40] C. Guerreiro, J. Soares, and A. G. Ortiz. Air quality in Europe — 2019 re-port. 5. Luxembourg: Publications Office of the European Union, 2019. i s b n:
9789294800886. d o i: 10.2800/822355. u r l: papers2://publication/uuid/1D25F4 1B-C673-4FDA-AB71-CC5A2AD97FDD (cit. on p. 5).
[41] R. Gunderman. Essential Radiology: Clinical Presentation, Pathophysiology, Imaging, 2nd ed.2nd ed. Thieme, 2006. i s b n: 9781588900821 (cit. on p. 43).
[42] A. Hartl, B. C. Song, and I. Pundt. “Atmospheric Chemistry and Physics 2-D reconstruction of atmospheric concentration peaks from horizontal long path DOAS tomographic measurements: parametrisation and geometry within a discrete approach”. In:Atmos. Chem. Phys6.3 (Mar. 2006), pp. 847–861. i s s n:
1680-7324. d o i: 10.5194/acpd- 5- 11781- 2005. u r l: http://www.atmos- chem-phys.net/6/847/2006/%20www.atmos-chem-phys.net/6/847/2006/ (cit. on p. 20).
[43] A. Hartl et al. “2d tomographic reconstruction of trace gas distributions from long-path DOAS measurements: General approach, validation and outlook on an experiment on an urban site”. In: Proceedings, 31st International Sympo-sium on Remote Sensing of Environment, ISRSE 2005: Global Monitoring for Sustainability and Security2 (2005) (cit. on p. 20).
[44] A. Harzing. Publish or Perish. u r l: http://www.harzing.com/pop.htm (cit. on p. 19).
[45] E. Hecht.Optics. 5th. Pearson, 2016. i s b n: 0133977226 (cit. on pp. 29, 30).
[46] G. T. Herman. “Image Reconstruction From Projections”. In:Real-Time Imag-ing1.1 (1995), pp. 3–18. i s s n: 10772014. d o i: 10.1006/rtim.1995.1002 (cit. on p. 44).
B I B L I O G R A P H Y
[47] G. T. Herman. Fundamentals of Computerized Tomography. Advances in Pat-tern Recognition. London: Springer London, 2009. i s b n: 978-1-85233-617-2.
d o i: 10.1007/978-1-84628-723-7. u r l: http://link.springer.com/10.1007/978-1-8 4628-723-7 (cit. on p. 44).
[48] G. T. Herman, A. Lent, and S. W. Rowland. “ART: Mathematics and applications.
A report on the mathematical foundations and on the applicability to real data of the algebraic reconstruction techniques”. In:Journal of Theoretical Biology 42.1 (1973). i s s n: 10958541. d o i: 10.1016/0022-5193(73)90145-8 (cit. on pp. 44, 47).
[49] IEA. Data & Statistics - International Energy Agency. u r l: https://www.iea.
org / data - and - statistics ? country = WORLD % 7B % 5C & %7Dfuel = Energy % 20 supply%7B%5C&%7Dindicator=Total%20primary%20energy%20supply%20 (TPES)%20by%20source (visited on 04/17/2020) (cit. on p. 13).
[50] International Agency for Research on Cancer. Outdoor Air Pollution. IARC Monograph 109. Vol. 109. 2016, p. 454. i s b n: 9789283201755 (cit. on p. 11).
[51] S. Ismail and E. V. Browell. Lidar: Differential Absorption Lidar. Second Edi.
Vol. 3. Elsevier, 2015, pp. 277–288. i s b n: 9780123822260. d o i: 10.1016/B978-0-12-382225-3.00204-8. u r l: http://dx.doi.org/10.1016/B978-0-12-382225-3.00204 -8 (cit. on p. 25).
[52] M. Johansson et al. “Tomographic reconstruction of gas plumes using scanning DOAS”. In:Bulletin of Volcanology71.10 (Dec. 2009), pp. 1169–1178. i s s n: 0258-8900. d o i: 10.1007/s00445-009-0292-8. u r l: http://link.springer.com/10.1007 /s00445-009-0292-8 (cit. on pp. 20, 21).
[53] A. Kak and M. Slaney.Principles of Computerized Tomographic Imaging. Soci-ety of Industrial and Applied Mathematics, 2001 (cit. on pp. 43, 44, 47–49).
[54] A. C. Kak and M. Slaney.Principles of Computerized Tomographic Imaging. 1.
Society for Industrial and Applied Mathematics, Jan. 2001, pp. 49–112. i s b n:
978-0-89871-494-4. d o i: 10.1137/1.9780898719277. u r l: http://epubs.siam.org/
doi/book/10.1137/1.9780898719277 (cit. on pp. 44, 48).
[55] Kassay. RAM2000 OP FTIR Brochure. 2022. u r l: http:// www.kassay.com/
newsite/wp-content/uploads/2018/03/RAM2000-GII-Brochure.pdf (cit. on p. 25).
[56] D. Kazantsev et al. “TomoPhantom, a software package to generate 2D–4D analytical phantoms for CT image reconstruction algorithm benchmarks”. In:
SoftwareX7 (Jan. 2018), pp. 150–155. i s s n: 2352-7110. d o i: 10.1016/J.SOFTX.2 018.05.003. u r l: https://www.sciencedirect.com/science/article/pii/S2352711018 300335?via%7B%5C%%7D3Dihub (cit. on p. 67).
[57] Y. J. Kim and U. Platt.Advanced Environmental Monitorin. 2007. i s b n: 9781402063633 (cit. on pp. 25, 27).
[58] R. T. Ku, E. D. Hinkley, and J. O. Sample. “Long-Path Monitoring of Atmo-spheric Carbon Monoxide with a Tunable Diode Laser System”. In: Applied Optics14.4 (Apr. 1975), p. 854. i s s n: 0003-6935. d o i: 10.1364/AO.14.000854.
u r l: https://opg.optica.org/abstract.cfm?URI=ao-14-4-854 (cit. on p. 25).
B I B L I O G R A P H Y
[59] T. Laepple et al. “Longpath DOAS tomography on a motorway exhaust gas plume: numerical studies and application to data from the BAB II campaign”.
In: Atmospheric Chemistry and Physics Discussions4.3 (May 2004), pp. 2435–
2484. i s s n: 1680-7375. d o i: 10.5194/ acpd- 4- 2435- 2004. u r l: www.atmos-chem-phys.org/acp/4/1323/%20http://www.atmos-chem-phys-discuss.net/4/243 5/2004/ (cit. on p. 20).
[60] A. Merlaud. “Development and use of compact instruments for tropospheric investigations based on optical spectroscopy from mobile platforms”. PhD The-sis. Louvain: Faculté des Sciences - Université Catholique de Louvain, 2013.
i s b n: 978-2-87558-128-0. u r l: http : / / books . google . com / books ? hl = en % 7 B % 5C & %7Dlr = %7B % 5C & %7Did = inXVSyR82zwC % 7B % 5C & %7Doi = fnd%7B%5C&%7Dpg=PR3%7B%5C&%7Ddq=Development+and+use+of+
compact+instruments+for+tropospheric+investigations+based+on+optical+
spectroscopy+from+mobile+platforms+sciences%7B%5C&%7Dots=VdebeDBQMc%
7B%5C&%7Dsig=4G0eVEvthJXuqd8WI3IWjjVXuXc (cit. on pp. 31, 33–35, 40).
[61] K. U. Mettendorf, A. Hartl, and I. Pundt. “An indoor test campaign of the tomography long path differential optical absorption spectroscopy technique”.
In: (2005). d o i: 10.1039/b511337g (cit. on p. 20).
[62] K. U. Mettendorf, A. Hartl, and I. Pundt. “An indoor test campaign of the tomography long path differential optical absorption spectroscopy technique”.
In: J. Environ. Monit.8.2 (2006), pp. 279–287. i s s n: 1464-0325. d o i: 10.1039 /B511337G. u r l: http://xlink.rsc.org/?DOI=B511337G (cit. on p. 20).
[63] J. G. Murphy, S. O’Driscoll, and N. J. Smith. “Multipath DOAS for tomographic measurements”. In:Opto-Ireland 2002: Optics and Photonics Technologies and Applications. Ed. by T. J. Glynn. Vol. 4876. Mar. 2003, p. 875. d o i: 10.1117/12 .463960. u r l: http://proceedings.spiedigitallibrary.org/proceeding.aspx?doi=10 .1117/12.463960 (cit. on p. 20).
[64] W. Neumann.Fundamentals of Dispersive Optical Spectroscopy Systems. Belling-ham, Washington: SPIE—The International Society for Optical Engineering, 2014. i s b n: 9780819498243 (cit. on pp. 31, 32).
[65] Newport Fibre Optics. Fiber Optics : Fiber Preparation and Fiber Connectors (cit. on p. 58).
[66] S. O’Driscoll and N. J. Smith. “Tomographic-DOAS: From sensing to visualisa-tion of urban street canyon pollutants”. In:Advances in Air Pollution13 (2003), pp. 681–691 (cit. on p. 20).
[67] S. O’Driscoll, J. G. Murphy, and N. J. Smith. “Computed tomography of air pollutants in street canyons”. In: Spie4876 (2003), pp. 958–967. d o i: 10.1117/1 2.463985 (cit. on p. 20).
[68] M. Office. The Great Smog of 1952. 2019. u r l: https://www.metoffice.gov.uk/
weather/learn-about/weather/case-studies/great-smog (visited on 09/23/2019) (cit. on p. 4).
[69] Omegon.Omegon Mighty Mak 60. 2021. (Visited on 12/04/2021) (cit. on p. 61).
B I B L I O G R A P H Y
[70] U. Platt and J. Stutz.Differential Optical Absorption Spectroscopy - Principles and Applications. Ed. by R. Guzzi et al. Heidelberg, Germany: Springer-Verlag Heidelberg Berlin, 2008. i s b n: 9783540211938 (cit. on pp. 4, 7, 31, 32, 34, 35, 37, 38).
[71] W. H. Press et al. Numerical Recipes: The Art of Scientific Computing. 3rd Editio. 2007. i s b n: 9788578110796. arXiv: arXiv:1011.1669v3 (cit. on p. 35).
[72] I. Pundt et al. “Mapping of tropospheric trace gas concentration distributions from ground and aircraft by DOAS-tomography (Tom-DOAS)”. In: () (cit. on pp. 20, 80).
[73] I. Pundt et al. “Measurements of trace gas distributions using Long-path DOAS-Tomography during the motorway campaign BAB II: Experimental setup and
results for NO 2”. In:Atmospheric Environment39.5 (2005), pp. 967–975. i s s n:
13522310. d o i: 10.1016/j.atmosenv.2004.07.035 (cit. on pp. 20, 21).
[74] I. Pundt et al. “Measurements of trace gas distributions using Long-path DOAS-Tomography during the motorway campaign BAB II: experimental setup and
results for NO2”. In:Atmospheric Environment39.5 (Feb. 2005), pp. 967–975.
i s s n: 13522310. d o i: 10.1016/j.atmosenv.2004.07.035. u r l: http://linkinghub.
elsevier.com/retrieve/pii/S1352231004009628 (cit. on p. 20).
[75] I. Pundt. “DOAS tomography for the localisation and quantification of anthro-pogenic air pollution”. In: Analytical and Bioanalytical Chemistry385.1 (Apr.
2006), pp. 18–21. i s s n: 1618-2642. d o i: 10 . 1007 / s00216 - 005 - 0205 - 4. u r l:
http://link.springer.com/10.1007/s00216-005-0205-4 (cit. on p. 20).
[76] J. Radon. “On the determination of functions from their integral values along certain manifolds”. In:IEEE Transactions on Medical Imaging5.4 (Dec. 1986), pp. 170–176. i s s n: 0278-0062. d o i: 10 . 1109 / TMI . 1986 . 4307775. u r l: http : //ieeexplore.ieee.org/document/4307775/ (cit. on p. 43).
[77] V. Rozanov et al. “Sciatran - a new radiative transfer model for geophysical appli-cations in the 240–2400 NM spectral region: the pseudo-spherical version”. In:
Advances in Space Research29.11 (June 2002), pp. 1831–1835. i s s n: 02731177.
d o i: 10.1016/S0273- 1177(02)00095- 9. u r l: https://linkinghub.elsevier.com/
retrieve/pii/S0273117702000959 (cit. on p. 39).
[78] L. A. Shepp and B. F. Logan. “The Fourier reconstruction of a head section”.
In: IEEE Transactions on Nuclear Science21.3 (June 1974), pp. 21–43. i s s n:
0018-9499. d o i: 10.1109/TNS.1974.6499235. u r l: http://ieeexplore.ieee.org/
document/6499235/ (cit. on p. 67).
[79] L. A. Shepp and Y. Vardi. “Maximum Likelihood Reconstruction for Emission Tomography”. In: IEEE Transactions on Medical Imaging 1.2 (Oct. 1982), pp. 113–122. i s s n: 0278-0062. d o i: 10.1109/ TMI.1982.4307558. u r l: https:
//ieeexplore.ieee.org/document/4307558/ (cit. on p. 49).
[80] R. L. Siddon. “Fast calculation of the exact radiological path for a three-dimensional CT array”. In:Medical Physics12.2 (Mar. 1985), pp. 252–255. i s s n: 00942405.
d o i: 10.1118/1.595715. u r l: http://doi.wiley.com/10.1118/1.595715 (cit. on p. 50).
B I B L I O G R A P H Y
[81] W. J. Smith.Modern Optical Engineering - The Design of Optical Systems. 3rd.
Carisbad, California: McGraw-Hill, 2000. i s b n: 0071363602 (cit. on p. 30).
[82] C. Stachniss, C. Plagemann, and A. J. Lilienthal. “Learning gas distribution models using sparse Gaussian process mixtures”. In:Autonomous Robots26.2-3 (Apr. 2009), pp. 187–202. i s s n: 0929-5593. d o i: 10.1007/s10514- 009- 9111- 5.
u r l: http://link.springer.com/10.1007/s10514-009-9111-5 (cit. on p. 67).
[83] J. Stutz and U. Platt. “Numerical analysis and estimation of the statistical error of differential optical absorption spectroscopy measurements with least-squares methods”. In: 35.30 (1996), pp. 6041–6053 (cit. on pp. 69, 80).
[84] J. Stutz et al. “A novel dual-LED based long-path DOAS instrument for the measurement of aromatic hydrocarbons”. In: Atmospheric Environment 147 (Dec. 2016), pp. 121–132. i s s n: 13522310. d o i: 10.1016/j.atmosenv.2016.09.054.
u r l: http:// linkinghub.elsevier.com/ retrieve/ pii/ S1352231016307713 (cit. on pp. 20, 22).
[85] P. Tans et al. Atmospheric CO2 concentrations (ppm) derived from in situ air measurements at Mauna Loa Observatory, Hawaii. Tech. rep. Mauna Loa, Hawaii, USA: NOAA/ESRL and Scripps Institution of Oceanography. u r l: ftp:
//aftp.cmdl.noaa.gov/products/trends/co2/co2%7B%5C_%7Dannmean%7B%5 C_%7Dmlo.txt (cit. on p. 11).
[86] D. Triantafyllou and N. A. Aspragathos. Intelligent Robotics and Applications. Ed. by S. Jeschke, H. Liu, and D. Schilberg. Vol. 7101. Lecture Notes in Com-puter Science PART 1. Berlin, Heidelberg: Springer Berlin Heidelberg, 2011, pp. 509–519. i s b n: 978-3-642-25485-7. d o i: 10.1007/978-3-642-25486-4. u r l:
http://link.springer.com/10.1007/978-3-642-25486-4 (cit. on p. 58).
[87] TSF and Lusa. Erupção do Vulcão dos Capelinhos foi há 60 anos. 2017. u r l:
https : / / www. tsf . pt / sociedade / foi ha 60 anos a erupcao do vulcao dos -capelinhos-8798755.html (visited on 04/11/2020) (cit. on pp. 9, 10).
[88] UNICEF. “Levels & Trends in Child Mortality: Report 2013-Estimates devel-oped by the UN Inter-agency Group for Child Mortality Estimation”. In: Unice-f/Who/Wb/Un(2013), pp. 1–32. i s s n: 19326203. d o i: 10.1371/journal.pone.01 44443 (cit. on p. 8).
[89] R. Valente de Almeida, N. Matela, and P. Vieira. “TomoSim: A Tomographic Simulator for Differential Optical Absorption Spectroscopy”. In: Drones 5.1 (Dec. 2020), p. 3. i s s n: 2504-446X. d o i: 10.3390/drones5010003. u r l: https:
//www.mdpi.com/2504-446X/5/1/3 (cit. on p. 48).
[90] R. Valente de Almeida and P. Vieira. “Forest Fire Finder – DOAS application to long-range forest fire detection”. In:Atmospheric Measurement Techniques10.6 (June 2017), pp. 2299–2311. i s s n: 1867-8548. d o i: 10.5194/amt-10-2299-2017.
u r l: https://www.atmos-meas-tech.net/10/2299/2017/ (cit. on pp. 1, 23, 32, 34).
[91] D. Vallero.Fundamentals of air pollution. 5th ed. Oxford, UK: Academic Press - Elsevier, 2014. i s b n: 9780124017337 (cit. on pp. 3, 5, 8, 9, 11, 22, 23, 25, 26).
[92] P. Vieira and J. Matos. System for Automatic Detection of Forest Fires Through Optical Spectroscopy. 2008 (cit. on p. 1).
B I B L I O G R A P H Y
[93] P. Vieira, J. Matos, and M. Mendes. Sistema para a detecção automática de incêndios florestais por espectroscopia óptica. 2007 (cit. on p. 1).
[94] P. Virtanen et al. “SciPy 1.0: fundamental algorithms for scientific computing in Python”. In:Nature Methods17.3 (Mar. 2020), pp. 261–272. i s s n: 1548-7091.
d o i: 10.1038/s41592-019-0686-2. u r l: http://www.nature.com/articles/s41592-019-0686-2 (cit. on p. 80).
[95] World Bank.World Bank Datasets. u r l: data.worldbank.org (visited on 04/17/2020) (cit. on p. 13).
[96] F. Zimmermann et al. “PRECISE POSITIONING of UAVS - DEALING with CHALLENGING RTK-GPS MEASUREMENT CONDITIONS during AUTO-MATED UAV FLIGHTS”. In: ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences4.2W3 (2017), pp. 95–102. i s s n: 21949050.
d o i: 10.5194/isprs-annals-IV-2-W3-95-2017 (cit. on p. 58).
This document was created using the (pdf/Xe/Lua)LATEX processor, based on the NOVAthesis template, developed at the Dep. Informática of FCT-NOVA by João M. Lourenço. [1]
[1] J. M. Lourenço.The NOVAthesis LATEX Template User’s Manual. NOVA University Lisbon. 2021. URL: https://github.com/joaomlourenco/novathesis/raw/master/template.pdf (cit. on p. 105).
A
F o r e s t F i r e F i n d e r : D OA S
a p p l i cat i o n t o l o ng - r a ng e f o r e s t
f i r e d e t e c t i o n
Atmos. Meas. Tech., 10, 2299–2311, 2017 https://doi.org/10.5194/amt-10-2299-2017
© Author(s) 2017. This work is distributed under the Creative Commons Attribution 3.0 License.
Forest Fire Finder – DOAS application to long-range forest fire detection
Rui Valente de Almeida1,2and Pedro Vieira1,2
1NGNS, Ingenious Solutions, Rua Cidade de Évora, 11, 2660-022 Loures, Portugal
2Faculty of Science and Technology, NOVA University of Lisbon, Caparica Campus, 2829-516 Caparica, Portugal Correspondence to:Pedro Vieira ([email protected])
Received: 23 September 2016 – Discussion started: 14 November 2016 Revised: 11 April 2017 – Accepted: 8 May 2017 – Published: 22 June 2017
Abstract. Fires are an important factor in shaping Earth’s ecosystems. Plant and animal life, in almost every land habi-tat, are at least partially dependent on the effects of fire. How-ever, their destructive force, which has often proven uncon-trollable, is one of our greatest concerns, effectively result-ing in several policies in the most important industrialised regions of the globe.
This paper aims to comprehensively characterise the For-est Fire Finder (FFF), a forFor-est fire detection system based mainly upon a spectroscopic technique called differential op-tical absorption spectroscopy (DOAS). The system is de-signed and configured with the goal of detecting higher-than-the-horizon smoke columns by measuring and comparing scattered sunlight spectra. The article covers hardware and software, as well as their interactions and specific algorithms for day mode operation. An analysis of data retrieved from several installations deployed in the course of the last 5 years is also presented.
Finally, this paper features a discussion on the most promi-nent future improvements planned for the system, as well as its ramifications and adaptations, such as a thermal imaging system for short-range fire seeking or environmental quality control.
1 Introduction
Fire is a process by which elements chemically combine with oxygen, releasing energy (as heat and light) and smoke into the surrounding environment. Fires are an important factor in shaping Earth’s ecosystems. Plant and animal life, in many
land habitats, are at least partially dependent on the effects of fire (Food and Agriculture Organisation , FAO).
The use of fire by hominids predates civilisation by thou-sands of years and, in today’s society, there are almost no areas of technology or scientific knowledge that do not in-volve fire in one way or another. However, fire’s destructive power is undeniable.
Forest fires are among the great concerns of the present day in industrialised countries. Research regarding wildfires has been targeted by many countries and unions worldwide in an effort to minimise the negative impact these events imply.
According to the Intergovernmental Panel on Climate Change (IPCC), climate change is expected to increase global temperatures and change rainfall patterns, leading to an increased risk of fire (IPCC, 2012). This means that the number of registered fires throughout the world is expected to increase, a phenomenon the world must be ready to ad-dress.
In the European Union, the Horizon2020 research pro-gramme states that there must be a union-wide investment in research concerning forest protection and recovery from fires. In the past, the FP7 programme had sponsored the de-velopment of an automatic forest fire detection system called FireSense, an investment of over EUR 2.5 million (European Comission, 2012).
The United States Forest Service acknowledge the impor-tance of understanding wildland fire dynamics, running a net-work of research centres solely dedicated to the study of this subject. Research endeavours take 6 % of the service’s an-nual budget, which is currently directed primarily towards fire suppression (United States Forest Department, 2015).
2300 R. Valente de Almeida and P. Vieira: FFF – Forest Fire Finder Australia is another geographic region where wildfires
have had a great impact. As a response, its government has created the Bushfire and Natural Hazards Cooperative Re-search Centre. The institution builds upon more than 10 years of experience dealing with Australian bushfires and aims to produce internationally recognised research regard-ing the study and modellregard-ing of wildfires in Australia and New Zealand (BNHCRC, 2016).
In spite of this global investigation effort regarding fires and their behaviour, every year, material losses as a result of fires ascend to billions of dollars and thousands of lives are lost in the same way. This leads to a strong increase in the size of the fire protection market, including passive and active detection platforms, which is expected to grow at a cumulative aggregate growth rate of 11.53 % from 2014 to 2020 (Research and Markets, 2016).
2 State of the art
In recent years, several methods have been developed in an attempt to automatically and reliably detect forest fires.
These systems differ primarily in their strategic approach to the issue at hand, creating three main categories:
– Satellite monitoring techniques: satellite data have been used for fire monitoring purposes since the late 20th century. The MODIS (MODerate resolution Imaging Spectroradiometer) and AVHRR (Advanced Very High Resolution Radiometer) sensors, deployed respectively in the Aqua/Terra and NOAA satellites, have had ex-tensive use in this regard. However, their low temporal resolution (2 and 4 times per 24 h, respectively) make them poor candidates for fire detection uses. Geosta-tionary satellites overcome this difficulty by continu-ously scanning a single, very large geographic region.
They have, nevertheless, a low spatial resolution of 1 km, which means that small fires are difficult for them to detect (Manyangadze, 2009).
– Wireless network sensing: the wireless sensor network approach to fire detection is completely different from the other two categories. Instead of having a single vice patrolling the target region, these systems are de-signed on the capabilities of a high number of extremely small battery-operated sensor boards that can communi-cate among themselves (Alkhatib, 2014; Liyang et al., 2005).
The sensor boards are equipped with several sensors, from temperature and humidity to luminance detectors.
In spite of their great fire detection capabilities, these networks present various drawbacks, such as their very limited individual range of detection and their 2-year lifetime or the fact that their remains might imply an environmental issue (Alkhatib, 2014).
– Large-area remote sensing: this family of systems is signed with the goal of minimising the number of de-ployed devices in a given target region. Their architec-ture implies the use of an optical principle in order to de-tect smoke or flames, whether optical cameras or spec-trometers.
There are already several commercially available systems, such as the Forest Fire Finder (FFF; the main subject of this paper), FireWatch, ForestWatch, AlarmEYE or Eyefi SPARC. Although these are commercial products, the avail-able information is sparse and many times outdated, so a true comparison not only is beyond the scope of the article but also would require more research efforts. Nevertheless, it is important to briefly describe the operating principles of the more prominent systems.
– ForestWatch, developed in South Africa by EnviroVi-sion Solutions, uses optical object recognition software, coupled to a very specific camera system. It detects smoke during the day and the flame glow during the night, at a maximum distance of 24 km in every di-rection, in a semi-automatic fashion. It is probably the most commercially successful system, with more than 300 currently operating towers (Envirovision Solutions, 2015; Hough, 2007).
– FireWatch is a commercial system operated and sold by IQ Wireless Gmbh, in Germany. The system uses opti-cal sensors and object recognition algorithms to detect smoke at a maximum distance of 15 km. It is important to mention that the FireWatch system is not a fully auto-matic fire detection platform, requiring a control room to operate correctly (IQ-Wireless, 2016).
– The FFF was developed in Lisbon, in a partnership be-tween the NOVA University of Lisbon and NGNS-IS, Ltd., in 2006. This patented system uses a spectroscopic technique to assess the atmosphere and detect smoke columns (NGNS-IS, 2016). During the night, the sys-tem changes its operation mode and relies solely on im-age processing to detect a fire’s glow. Its maximum rate detection range is of 15 km, and it acts with complete autonomy, requiring minimal human intervention (see Fig. 1).
The FFF’s most significant advantage over its rivals is its low number of false alarms (typically one or two per week). This comes from the fact that the system’s smoke-detection capa-bilities do not rely on image processing. This in turn means that reliable detections can be achieved by a smaller number of deployed devices (only one, three for triangulation). How-ever, more reliable alarms imply less human intervention, which translates into less financial expenditure over time.
The FFF system is the only one to use an optical spec-troscopy technique to detect fire through smoke presence in
R. Valente de Almeida and P. Vieira: FFF – Forest Fire Finder 2301
Figure 1.The Forest Fire Finder system (NGNS-IS, 2016).
real time. Since the analysis is carried out in an outdoor sce-nario, the process is not as straightforward as in laboratory experiments. This article addresses only the spectroscopic techniques used in the system’s daytime operation mode.
3 The technique
The FFF system makes use of a spectroscopic technique called differential optical absorption spectroscopy (DOAS).
This is a well-established and widely used technique in the field of atmospheric studies (Platt and Stutz, 2007).
There are two main categories of DOAS experiment as-semblies, with different goals and capabilities:
– Active systems, of which a simple illustration is pre-sented in Fig. 2, are characterised by relying on an artificial light source for their measurements. A spec-trometer at the end of the light path performs spectro-scopic detection. Active DOAS techniques are very sim-ilar to traditional in-lab absorption spectroscopy tech-niques (Platt and Stutz, 2007);
– Passive DOAS techniques, illustrated in Fig. 3, use natu-ral light sources, such as the Sun and the moon, in their measurement process. An optical system is pointed in certain elevation and azimuth angles and sends the cap-tured light into a spectrometer, connected to a computer.
The system returns the total value of the light absorp-tion in its path (Platt and Stutz, 2007; Merlaud, 2013).
Since the FFF system is basically a passive DOAS
sys-tem, we will centre our discussion on this category from this point forward.
DOAS itself is based on Lambert–Beer’s law, which can be written as (Platt and Stutz, 2007)
I (λ)=I0(λ)·exp(−σ (λ)·c·L) , (1) whereλis the wavelength of the emitted light;I (λ)is the light intensity as measured by the system;I0(λ) is the in-tensity of the light as emitted by the source; andσ (λ)is the absorption cross section of absorber, which is wavelength de-pendent; cis the concentration of the absorber we want to measure.
This law allows the definition of optical thickness (τ) (Platt and Stutz, 2007):
τ (λ)=ln I0(λ)
I (λ)
=σ (λ)·c·L. (2)
In a laboratory setting, Eq. (1) or (2) can be used to di-rectly calculate an absorber’s concentration, provided there is knowledge of its cross section. In the open atmosphere, however, absorption spectroscopy techniques are far more complex. On one hand,I0(λ)is not accessible since we mea-sure from inside the medium we want to meamea-sure. On the other hand, there are several environmental and instrumen-tal effects that influence measurement results. These effects include the following (Platt and Stutz, 2007).
– Rayleigh scattering is due to small molecules present in the atmosphere and is heavily influenced by wavelength (hence the blue colour of the sky).