The Method for Determining Forest Characteristics Based on Earth Remote Sensing Materials, Forest Management Data and the k-NN Algorithm (Case Study of Lodeynopol’skoe Forest District of Leningrad Region)

Complete text of the article:

Download article (pdf, 2.1MB )

UDС

630*5:528.85:630*9

DOI:

10.17238/issn0536-1036.2019.4.45

Abstract

The Russian system of National (State) Forest Inventory (NFI) valid from 2007 is methodologically imperfect and serves as a target of criticism and discussion. Insufficient attention paid to remote sensing should be attributed to the system’s weaknesses. A possible way of improving the NFI system is the use of automatic classification of forests’ characteristics based on materials of Earth remote sensing. One of the advanced automatic methods for forest remote sensing materials classification is k-NN or k-nearest neighbors algorithm, which have been successively used in the NFIs in other countries. It is based on regression between the spectral response characteristics of pixels with known forest characteristics and remaining pixels of the image. Questions of practical application of this method for the purposes of NFI were first raised and studied by Finnish researchers in the 1990s. Over the past two decades, a considerable amount of research in this area has been carried out in different countries. The purpose of our research is to assess the feasibility of using the k-NN method for determining the generalized characteristics of forests on the example of the Lodeynopol’skoe forest district in Leningrad region. The forest district area is 401,866 ha. It consists of 16 forest sub-districts. A regular network was created at a pitch of 1×1 km in order to form a set of trial plots for the classification by the means of GIS-technologies within the forest district. Round-shaped plots with a radius of 10 m or forest inventory compartments were used as trial plots located in the network nodes. Forest description for both types of plots was taken from the forest management data. Landsat-8 images were used for the classification (spectral channels: GREEN, RED, NIR, SWIR 2). The Landsat-8 images were automatically classified according to a number of forest inventory characteristics: average growing stock per 1 ha; middle class of bonitet; average relative density; share of coniferous and deciduous plantations area. The results of the k-NN classification were compared with the forest management materials. Systematic errors in the assessment of growing stocks, completeness and bonitet for the territory of Lodeynopol’skoe forest district made up less than 5 %. The obtained results confirm the potential of further study of conceptual and practical issues of the k-NN method application for determining forest characteristics. The development of this direction can contribute to the improvement of the Russian State Forest Inventory methodology. 

Authors

D.M. Chernikhovskii, Candidate of Agriculture, Assoc. Prof.; ResearcherID: I-7020-2016, ORCID: 0000-0002-6375-3080
A.S. Alekseev, Doctor of Geography, Prof.; ResearcherID: F-6891-2010, ORCID: 0000-0001-8795-2888

Affiliation

St. Petersburg State Forest Engineering University named after S.M. Kirov, Institutskiy per., 5, Saint Petersburg, 194021, Russian Federation; e-mail: cherndm2006@yandex.ru, a_s_alekseev@mail.ru

Keywords

k-nearest neighbors algorithm, remote sensing methods, forest inventory, automatic classification of images

For citation

Chernikhovskii D.M., Alekseev A.S. The Method for etermining Forest Characteristics Based on Earth Remote Sensing Materials, Forest Management Data and the k-NN Algorithm (Case Study of Lodeynopol’skoe Forest District of Leningrad Region). Lesnoy Zhurnal [Forestry Journal], 2019, no. 4, pp. 45–65. DOI: 10.17238/issn0536-1036.2019.4.45

References

1. Alekseev A.S. Theoretical Foundations of the State (National) Forest Inventory. Lesnoye khozyaystvo, 2009, no 4, pp. 31–33.
2. Alekseev A.S. Methodological Foundations of Organization and Conducting the State Forest Inventory. Proceedings of the All-Russian Conf. “The Use of Materials of the State Forest Inventory for Environmental Protection”, Bryansk, October 9–10, 2013. Moscow, Roslesinforg Publ., 2013, pp. 87–96.
3. Arkhipov V.I. Special Features of the National Forest Inventory. What Kind of Forest Inventory does Russia Need? Proceedings of the All-Russian Conf. “The Use of Materials of the State Forest Inventory for Environmental Protection”, Bryansk, October 9–10, 2013. Moscow, Roslesinforg Publ., 2013, pp. 105–112.
4. Belova E.I., Ershov D.V. Assessing Reforestation on Clear Cuts Based on Landsat Time Series. Lesovedenie [Russian Journal of Forest Science], 2015, no. 5, pp. 339–345.
5. Gol’tsev V., Tolonen T., Syunev V., Dalin B., Gerasimov Yu., Karvinen S. Wood Harvesting and Logistics in Russia – Focus on Research and Business Opportunities: Final Report of the Research Project. Ed. by S. Karvinen. Working Papers of the Finnish Forest Research Institute. 2012, iss. 221. 159 p. Available at: http://www.metla.fi/julkaisut/ workingpapers/2012/mwp221.htm (Accessed 16.02.19).
6. Zhirin V.M., Knyazeva S.V., Eydlina S.P. Application of Space Images for Reforestation Dynamics Evaluating on the Plane Territories. Sovremennye problemy Earth from space], 2011, vol. 8, no. 2, pp. 208–216.
7. Zhirin V.M., Knyazeva S.V., Eydlina S.P. The Eco-Dynamical Study of Forest-Forming Process with Space Imagery. Lesovedenie [Russian Journal of Forest Science], 2013, no. 5, pp. 76–85.
8. Zhirin V.M., Knyazeva S.V., Eydlina S.P. Dynamics of Spectral Brightness for Species/Age Structure of Groups of the Forest Types on LANDSAT Satellite Images. Lesovedenie [Russian Journal of Forest Science], 2014, no. 5, pp. 3–12.
9. Zhirin V.M., Knyazeva S.V., Eydlina S.P. Influence of Forest Canopy Morphology and Relief on Spectral Characteristics of Taiga Forests. Issledovanie Zemli iz Kosmosa, 2016, no. 5, pp. 10–20. DOI: 10/7868/S0205961416050080
10. Zhirin V.M., Knyazeva S.V., Eydlina S.P., Zukert N.V. Seasonal Informative Properties of Multispectral High-Resolution Space Images for Studying of Forest Species and Age Dynamics. Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa [Current problems in remote sensing of the Earth from space], 2012, vol. 9, no. 1, pp. 87–94.
11. Kresnov V.V., Strakhov V.V., Filipchuk A.V. National Forest Inventory in Foreign Countries. Lesokhozyaystvennaya informatsiya [Forestry information], 2008, vol. 10-11, pp. 53–58.
12. Forest Management Regulation of the Lodeynopol’skoe Forest District of Leningrad Region. FSBI «ROSLESINFORG», Branch of the FSBI «ROSLESINFORG» «SEVZAPLESPROEKT». Saint Petersburg, 2017. 306 p.
13. Malysheva N.V. Automated Interpretation of Aerospace Images of Forest Stands: Educational Textbook. Moscow, MSTU Publ., 2012. 154 p.
14. Neshataev M.V., Neshataev V.Yu. Combinated Method for Vegetation Mapping (On the Example of the Lapland Reserve). Izvestia Sankt-Peterburgskoj Lesotekhnicheskoj Akademii [News of the Saint Petersburg State Forest Technical Academy], 2012, iss. 201. pp. 29–40.
15. Perepechina Yu.I., Glushenkov O.I., Glushenkov I.S. State Forest Inventory of Russian Forests. Lesotekhnicheskiy zhurnal [Forestry Engineering Journal], 2014, no. 2, pp. 60–67. DOI: 10.12737/4508
16. Perepechina Yu.I., Glushenkov O.I., Korsikov R.S. Automated Detection of Forest Cover and Forest Characteristics Kolichestvennykh of Municipal Shebekinskiy District of the Belgorod Region on Space. Lesokhozyaystvennaya informatsiya [Forestry information], 2017, no. 4, pp. 85–93. DOI: 10.24419/LHI.2304-3083.2017.4.09
17. Savchenko A.A., Vyvodzev N.V. Evaluate the Possible Application Remote Sensing Data for Monitoring the Health and Forest State Forests. Uchenye zametki TOGU [Scientists notes PNU], 2015, vol. 6, no. 4, pp. 658–661. Available at: http://pnu.edu.ru/media/ejournal/articles-2015/TGU_6_237.pdf (Accessed 16.02.19).
18. Sidorenkov V.M., Doroshenkova E.V., Lopatin E.V., Ryabtsev O.V., Sidorenkova E.M. Zoning of the Udmurtija Republic Territory by Game Resource Habitat Categories Based on Landsat 8 Oli-Tirs Satellite Survey Data. Lesotekhnicheskiy zhurnnal [Forestry Engineering Journal], 2015, vol. 5, no. 3(19), pp. 84–93. DOI: 10.12737/14156
19. Soldatenkov A.A. Interpretation of Structure of Forest Vegetation in Conditions of Mid-Mountain Relief. Vestnik Adygeyskogo gosudarstvennogo universiteta. Seriya 4: Estestvenno-matematicheskie i tekhnicheskie nauki [The Bulletin of Adyghe State University. Series 4: Natural-Mathematical and Technical Sciences], 2014, iss. 1(133), pp. 127–130.
20. Soromotin A.V., Brodt L.V. Monitoring of Vegetation Cover during the Development of Oil and Gas Fields According to the Landsat Multispectral Survey Data. Vestnik Tyumenskogo gosudarstvennogo universiteta. Ekologiya i prirodopol’zovanie [Tyumen State University Herald. Natural Resource Use and Ecology], 2018, vol. 4, no. 1, pp. 37–49. DOI: 10.21684/2411-7927-2018-4-1-37-49
21. Terehov A.G., Makarenko N.G., Pak I.T. Automatic Classification Algorithm of QuickBird Images in the Problem of Evaluating of Forest Completeness. Komp’yuternaya optika [Computer Optics], 2014, vol. 38, iss. 3, pp. 580–583. DOI: 10.18287/0134-2452-2014-38-3-580-583
22. Tolkach I.V., Kravchenko O.V., Ozhich O.S., Tarkan A.V., Il’yuchik M.A. Regularities of Variability of the Spectral Brightness Canopy of the Main Forest Speacies of Belarus on Images of Scanner ADS 100. Trudy BGTU. Seriya 1: Lesnoe khozyaystvo, prirodopol’zovanie i pererabotka vozobnovlyayemykh resursov [Proceedings of BSTU. Series 1: Forestry. Environmental management. Reprocessing of renewable resources], 2017, no. 2(198), pp. 43–49.
23. Farber S.K., Bryuhanov N.V. Mass Forest Survey and State Forest Inventory Data: Specification of the Discrepancies, Causes, Analysis. Sibiriskiy lesnoy zhurnal [Siberian Journal of Forest Science], 2014, no. 5, pp. 16–28.
24. Chernykh D.V., Biryukov R.Yu., Zolotov D.V., Vagner A.A. Anthropogenic
Modifications and Transformations of Landscapes in the Basin of the Kasmala River: Classification and Dynamics Based on the Remote Sensing Data. Vestnik altaiskoy nauki, 2014, no. 1(19), pp. 233–240.
25. Sharikalov A.G., Yakutin M.V. The Analysis of Taiga Ecosystems Condition Applying Automatic Decoding Method. Izvestiya Altayskogo gosudarstvennogo universiteta [Izvestiya of Altai State University], 2014, iss. 3-1(83), pp. 123–127. DOI: 10.14258/izvasu(2014)3.1-22
26. Yaroshenko A.Yu., Vladimirova N.A., Kobyakov K.N. Suggestions on Optimizing the State Forest Inventory in Turms of Using Its Results for Environmental Protection. Proceedings of the All-Russian Workshop “The Use of Materials of the State Forest Inventory for Environmental Protection”, Bryansk, October 9–10, 2013. Moscow, Roslesinforg Publ., 2013, pp. 82–86.
27. Beaudoin A., Bernier P.Y., Guindon L., Villemaire P., Guo X.J., Stinson G., Bergeron T., Magnussen S., Hall R.J. Mapping Attributes of Canada’s Forests at Moderate Resolution through kNN and MODIS Imagery. Canadian Journal of Forest Research, 2014, vol. 44(5), pp. 521–532. DOI: 10.1139/cjfr-2013-0401
28. Congedo L. Semi-Automatic Classification Plugin Documentation. Release 6.2.0.1. 2019. 212 p. DOI: 10.13140/RG.2.2.29474.02242/1
29. Forest Inventory: Methodology and Applications. Ed. by A. Kangas, M. Maltamo. Netherlands, Springer, 2006. 362 p. DOI: 10.1007/1-4020-4381-3
30. Franco-Lopez H., Ek A.R., Bauer M.E. Estimation and Mapping of Forest Stand Density, Volume, and Cover Type Using the k-Nearest Neighbors Method. Remote Sensing of Environment, 2001, vol. 77, iss. 3, pp. 251–274. DOI: 10.1016/S0034-4257(01)00209-7
31. Gjertsen A.K. Accuracy of Forest Mapping Based on Landsat TM Data and a kNN-Based Method. Proceedings of ForestSat 2005, Borås, May 31 – June 3, 2005. Ed. by H. Olsson. Borås, Skogsstyrelsen, 2005, pp. 7–11.
32. Gjertsen A.K. Accuracy of Forest Mapping Based on Landsat TM Data and a kNN-Based Method. Remote Sensing of Environment, 2007, vol. 110, iss. 4, pp. 420–430. DOI: 10.1016/j.rse.2006.08.018
33. Haapanen R., Ek A. Software and Instructions for kNN Applications in Forest Resources Description and Estimation. Department of Forest Resources Staff Paper Series. St. Paul, MN, University of Minnesota, 2001, no. 152. 19 p.
34. Haapanen R., Ek A.R., Bauer M.E., Finley A.O. Delineation of Forest/Nonforest Land Use Classes Using Nearest Neighbor Methods. Remote Sensing of Environment, 2004, vol. 89, iss. 3, pp. 265–271. DOI: 10.1016/j.rse.2003.10.002
35. Katila M., Tomppo E. Selecting Estimation Parameters for the Finnish Multisource National Forest Inventory. Remote Sensing of Environment, 2001, vol. 76, iss.1, pp. 16–32. DOI: 10.1016/S0034-4257(00)00188-7
36. Koch B. Remote Sensing Supporting National Forest Inventories NFA. Knowledge Reference for National Forest Assessments. Rome, FAO, 2015, pp. 77–92. Available at: http://www.fao.org/3/a-i4822e.pdf (Accessed 21.02.19).
37. Koukal T., Suppan F., Schneider W. The Impact of Relative Radiometric Calibration on the Accuracy of kNN-Predictions of Forest Attributes. Proceedings of ForestSat 2005, Borås, May 31 – June 3, 2005. Ed. by H. Olsson. Borås, Skogsstyrelsen, 2005, pp. 17–21.
38. McInerney D., Pekkarinen A., Haakana M. Combining Landsat ETM+ with Field Data for Ireland’s National Forest Inventory – A Pilot Study for co. Clare. Proceedings of ForestSat 2005, Borås, May 31 – June 3, 2005. Ed. by H. Olsson. Borås, Skogsstyrelsen, 2005, pp. 12–16.
39. McRoberts R.E. Remote Sensing Support for the National Forest Inventory of the United States of America. Proceedings of ForestSat 2005, Borås, May 31 – June 3, 2005. Ed. by H. Olsson. Borås, Skogsstyrelsen, 2005, pp. 1–6.
40. McRoberts R.E., Tomppo E.O. Remote Sensing Support for National Forest Inventories. Remote Sensing of Environment, 2007, vol. 110, iss. 4, pp. 412–419. DOI: 10.1016/j.rse.2006.09.034
41. Meng Q., Cieszewski C.J., Madden M., Borders B.E. K Nearest Neighbor Method for Forest Inventory Using Remote Sensing Data. GIScience & Remote Sensing, 2007, vol. 44, iss. 2, pp. 149–165. DOI: 10.2747/1548-1603.44.2.149
42. Nilsson M., Holm S., Reese H., Wallerman J., Engberg J. Improved Forest Statistics from the Swedish National Forest Inventory by Combining Field Data and Optical Satellite Data Using Post-Stratification. Proceedings of ForestSat 2005, Borås, May 31–June 3, 2005. Ed. by H. Olsson. Borås, Skogsstyrelsen, 2005, pp. 22–26.
43. Özsakabaşi F. Classification of Forest Areas by k Nearest Neighbor Method: Case Study, Antalya. M.Sc. in Geodetic and Geographic Information Thesis. Antalya, Middle East Technical University, 2008. 101 p.
44. Supervised kNN Classifier (kNN). Updated: 22.05.2017. 19 p. Availabe at: https://github. com/m6ev/kNN/blob/master/kNN_Documentation.pdf(Accessed 21.02.19).
45. Tomppo E.O. The Finnish Multi-Source Inventory. Proceedings of ForestSat 2005, Borås, May 31 – June 3, 2005. Ed. by H. Olsson. Borås, Skogsstyrelsen, 2005, pp. 27–37.
46. Tomppo E., Czaplewski R., Mäkisara K. The Role of Remote Sensing in Global Forest Assessment. Forest Resources Assessment – WP no. 61. Rome, FAO, 2002. Available at: http://www.fao.org/docrep/006/ad650e/AD650E00.htm#TopOfPage (Accessed 21.02.19).
47. Tomppo E., Haakana M., Katila M., Peräsaari J. Multi-Source National Forest Inventory: Methods and Applications. Netherlands, Springer, 2008. 373 p. DOI: 10.1007/978-1-4020-8713-4
48. Tomppo E., Halme M. Using Coarse Scale Forest Variables as Ancillary Information and Weighting of Variables in k-NN Estimation: A Genetic Algorithm Approach. Remote Sensing of Environment, 2004, vol. 92, iss. 1, pp. 1–20. DOI: 10.1016/j.rse.2004.04.003

Received on February 21, 2019