View or download the article (Rus)

About the Authors

Mikhail A. Korets

Sukachev Institute of Forest SB RAS,
Akademgorodok str., 50/28, 660036, Krasnoyarsk, Russia,

Siberian Federal University,
Svobodniy str., 79, 660041, Krasnoyarsk, Russia

Viktor M. Skudin

“ROSLESINFORG” “Vostsibleproekt”,
Krupskaya str., 42, 660062, Krasnoyarsk, Russia,


Conventionally forest inventory polygons mapping procedure is done manually by an aerial red-green-near infrared airborne imagery analysis. A forest inventory expert with help of the selective field plot data assigns forest stand attributes to polygons. This manual approach is time-consuming and highly subjective. The current availability of multispectral satellite imagery, digital elevation models of fine resolution and modern software allows implementing of the atomized and object based approach of forest inventory polygons mapping.

We elaborate and test the new approach in the framework of forest inventory tasks for a number of forest management areas (“Kuznetsk Alatau” reserve, “Sayano-Shushenskaya” reserve and forest lease area “Karat”) with total area more than 800 000 ha and location in central and southern Siberia. The Trimble eCognition Developer 8, Scanex Image Processor (Thematic Pro) and ESRI ArcGIS 10 software were used.

We used multiband satellite imagery (RS-composites) of Rapideye (5 spectral bands, 5 m resolution) and WorldView-2 (4 spectral bands R-G-B-NIR with spatial resolution of 1.8 m and fused to panchromatic band resolution of 0.5 m) as main data source for forest attributes mapping. As an additional reference data, we used old achieved forest inventory maps of forest inventory polygons and blocks boundaries, as well as recent groundtruth field data of 2013–2016.

We applied the 20m-resolution ASTER GDEM 2 raster digital elevation model to estimate and map the forest growing conditions. A two-layered images (DEM-composites) were calculated with layers of sin(a) и cos(a), where a—is an aspect angle of each pixel.

Both RS- and DEM-composites were automatically segmented for two assigned levels of spatial heterogeneity. Then we combined RS- and DEM-based segments into layer of initial forest inventory polygons and classified them with help of the maximum likelihood method, based on training samples of reference field sites. The resulting attributes tables of generalized inventory polygons were compiled as a combination of polygon attributes of initial detailed level. Finally, the output inventory polygons were spatially filtered, smoothed, snapped to the block boundaries and renumber according to the forest inventory regulations.

This automated method takes less time and reduces the human factor influence to compare it with traditional one. It may be rapidly repeated and adjusted to other test sites. Moreover, produced inventory polygons geometry and attributes meet Russia’s traditional forest inventory requirements.


forest inventory polygons mapping, GIS, RS, DEM.


  1. Bock M., Xofis P., Mitchley J., Rossner G., Wissen M. Object-oriented methods for habitat mapping at multiple scales—Case studies from Northern Germany and Wye Downs, UK. Journal for Nature Conservation. 2005. V. 13. P. 75–89. DOI: 10.1016/j.jnc.2004.12.002.
  2. Kolesnikov B.P. Siberian pine forests of the Russian Far East. M.; L.: Nauka, 1956. 261 p. (in Russian).
  3. Korets M.A., Danilova I.V., Cherkashin V.P. Remote indication of forest landscape structure, In: Regional Sustainable Forestry Problems. Sukachev Institute of Forest, Krasnoyarsk, 2007. P. 52–68 (in Russian).
  4. Korets M.A. The use of object-based image analysis for mapping forest inventory stands based on satellite imagery and digital elevation model. International conference proceedings “Regional problems of remote sensing of the Earth”, 2014. Krasnoyarsk: SFU, 2014. P. 294–298 (in Russian).
  5. Korets M., Ryzhkova V., Danilova I., Prokushkin A. Vegetation cover mapping based on remote sensing and digital elevation model data, Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci. 2016. V. XLI-B8. P. 699–704. DOI: 10.5194/isprs-archives-XLI-B8-699-2016.
  6. Minayeva L.U. Forest inventory regulation of the Russia. V. 1. M.: Voenizdat, 1995. No 6. 274 p. (in Russian).
  7. Mkrtchan A.S. Atomized mapping of the landscape units on the base of relief classification with help of GIS. International conference proceedings. M.: MGU, 2006. P. 203–208 (in Russian).
  8. Richards J., Xiuping J. Remote sensing digital image analysis: an introduction. Birkhäuser, Basel, 2005. Switzerland. 439 р.
  9. Ryzhkova V.A. Southern taiga forest regeneration dynamics. In: Forest Ecosystems of the Yenisei Meridian. Novosibirsk, 2002. P. 146–157 (in Russian).
  10. Ryzhkova V.A. Regeneration of subtaiga and mountain taiga forests. In: Regional Sustainable Forestry Problems. Sukachev Institute of Forest Publications. Krasnoyarsk, 2007. P. 78–90 (in Russian).
  11. Ryzhkova V., Danilova I. GIS-based classification and mapping of forest site condition and vegetation. BOSQUE, 2012. V. 33(3). P. 293–297.

For citation: Korets M.A., Skudin V.M. AUTOMATED APPROACH FOR MAPPING OF FOREST INVENTORY POLYGONS ON THE BASE OF SPACE IMAGERY AND DIGITAL ELEVATION MODEL Proceedings of the International conference “InterCarto. InterGIS”. 2018;24(2):94–105