Agroforestry Systems

, Volume 88, Issue 5, pp 865–875 | Cite as

Identification of windbreaks in Kansas using object-based image analysis, GIS techniques and field survey

  • Kabita GhimireEmail author
  • Mike W. Dulin
  • Robert L. Atchison
  • Douglas G. Goodin
  • J. M. Shawn Hutchinson


Windbreaks are valuable resources in conserving soils and providing crop protection in Great Plains states in the US. Currently, Kansas has no up-to date inventory of windbreaks. The goal of this project was to assist foresters with future windbreak renovation planning and reporting, by outlining a series of semi-automated digital image processing methods that rapidly identify windbreak locations. There were two specific objectives of this research. First, to develop semi-automated methods to identify the location of windbreaks in Kansas, this can be applied to other regions in Kansas and the Great Plains. We used a remote sensing technique known as object-based image analysis (OBIA) to classify windbreaks visible in the color aerial imagery of National Agriculture Imagery Program. We also combined GIS techniques and field survey to complement OBIA in generating windbreak inventory. The techniques successfully located more than 4500, windbreaks covering an approximate area of 2500, hectares in 14 Kansas counties. The second purpose of this research is to determine how well the results of the automated classification schemes match with other available windbreak data and the selected sample collected in the field. The overall accuracy of OBIA method was 58.97 %. OBIA combined with ‘heads up’ digitizing and field survey method yielded better result in identifying and locating windbreaks in the studied counties with overall accuracy of 96 %.


Shelterbelts Soil conservation Crop protection Kansas Great plains 



Funding for this project was provided by the Kansas Forest Service through the USDA State and Private Forestry Western Competitive Resource Allocation grant program.


  1. Atchison RL, Daniels R, Martinson E (2010) Kansas forest action plan (KFAP). Kansas forest resource assessment and strategy. Kansas State University, ManhattanGoogle Scholar
  2. Baatz M, Benz U, Dehghani S, Heynen M, Höltje A, Hofmann P, Lingenfelder I, Mimler M, Sohlbach M, Weber M, Willhauck G (2004) eCognition Professional: User Guide 4. Munich: definiens-imaging. Accessed 27 Feb 2014
  3. Benz UC, Hofmann P, Willhauck G, Lingenfelder I, Heynen M (2004) Multi-resolution, object-oriented fuzzy analysis of remote sensing data for GIS-ready information. J Photogramm Remote Sens 58:239–258CrossRefGoogle Scholar
  4. Blaschke T (2010) Object-based image analysis for remote sensing. ISPRS J Photogramm Remote Sens 65:2–16CrossRefGoogle Scholar
  5. Brandle JR, Hodges L, Zhou XH (2004) Windbreaks in North American agricultural systems. Agrofor Syst 61:65–78Google Scholar
  6. Cable TT (1999) Nonagricultural benefits of windbreaks in Kansas. Great Plain Res: A J Nat Soc Sci 417:41–55Google Scholar
  7. Castonguay TL, Hansen MH (1984) Wooded stripes and windbreaks in Kansas, 1981. Research Notes. NC313. North Central Forest Experiment Station. USDA. Saint PaulGoogle Scholar
  8. Congalton RG (1991) A review of assessing the accuracy of classifications of remotely sensed data. Remote Sens Environ 37:35–46CrossRefGoogle Scholar
  9. Congalton RG, Green K (1999) Assessing the accuracy of remotely sensed data: principles and practices. Lewis Publishers, Boca Raton, p 137pGoogle Scholar
  10. Croker T (1991) The Great Plains shelterbelt. Artistic Printers, GreenevilleGoogle Scholar
  11. Dodgson NA (1992) Image re-sampling. Technical report: Number 261. University of Cambridge, CambridgeGoogle Scholar
  12. Droze WH (1977) Trees, prairies, and people: a history of tree planting in the plains states. USDA Forest Service and Texas Woman’s University Press, Denton 331Google Scholar
  13. ESRI (2011) ArcGIS desktop: release 10. Environmental Systems Research Institute, RedlandsGoogle Scholar
  14. Fitzpatrick-Lins K (1981) Comparison of sampling procedures and data analysis for a land-use and land-cover map. Photogramm Eng Remote Sens 47(3):343–351Google Scholar
  15. Goodin DG, Harrington J Jr, Rundquist B (2002) Land cover change and associated trends in surface reflectivity and vegetation index in southwest Kansas: 1972–1992. Geocarto Int 17(1):45–52CrossRefGoogle Scholar
  16. Grala RK, Tyndall JC (2010) Impact of field windbreaks on visual appearance of agricultural lands. Agrofor Syst 80:411–422CrossRefGoogle Scholar
  17. Harrington LMB (2001) Attitudes toward climate change: major emitters in southwestern Kansas. Clim Res 16(113–122):341Google Scholar
  18. Hsu C-W, Chang C-C, Lin C-J (2008). A practical guide to support vector classification. Department of Computer Science, National Taiwan University, Taipei, Taiwan. Accessed 5 May 2014
  19. Huang X, Zhang L (2008) An adaptive mean shift analysis approach for object extraction and classification from urban hyperspectral imagery. IEEE Trans Geosci Remote Sens 46(12):4173–4185CrossRefGoogle Scholar
  20. ITT Visual Information Solutions 2007. An interval based attribute ranking technique. ITT-VIS White Paper, JulyGoogle Scholar
  21. Jager G, Benz U (2000) Measures of classification accuracy based on fuzzy similarity. IEEE Trans Geosci Remote Sens 38(3):1462–1467CrossRefGoogle Scholar
  22. Jensen JR (2005) Introductory digital image processing: a remote sensing perspective, 3rd edn. Prentice Hall, Upper Saddle RiverGoogle Scholar
  23. Koch DE, Mohler RL, Goodin D (2007) Stratifying land use/land cover for spatial analysis of disease ecology and risk: an example using object-based classification techniques. Geospatial Health 2(1):15–28PubMedGoogle Scholar
  24. Kort J, Stefner C (2007) A protocol for shelterbelt assessment related to its function. When trees and crops get together. In: Proceedings of the 10th North American agroforestry conference, Québec City June 10–13Google Scholar
  25. Natural Resource Inventory (NRI), NRCS (2010) Soil Erosion in Cropland. Accessed Aug 2013
  26. Parker JA, Kenyon RV, Troxel DE (1983) Comparison of interpolating methods for image resampling. IEEE Trans Med Imaging 2(1):31–39PubMedCrossRefGoogle Scholar
  27. Prairie States Forest Project (PSFP) (1937) Forestry for the great plains. USDA Forest Service, LincolnGoogle Scholar
  28. Quam V, Johnson L, Wight B, Brandle JR (1994) Windbreaks for livestock operations. Natural resources. University of Nebraska Extension EC 94 -1766-XGoogle Scholar
  29. Quattrochi DA, Walsh SJ, Jensen JR, Ridd MK (2003) Remote sensing. In: Gaile G, Wilmot CJ (eds) Geography in America, vol Chapter 25. Oxford University Press, OxfordGoogle Scholar
  30. Radoux J, Defourny P (2007) A quantitative assessment of boundaries in automated forest stand delineation using very high resolution imagery. Remote Sens Environ 110:468–475CrossRefGoogle Scholar
  31. Read R (1958) The Great Plains shelterbelt in 1954: a re-evaluation of field windbreaks planted between 1935 and 1942 and a suggested research program. Great Plains Agric Counc No. 16Google Scholar
  32. Richards JA, Xiuping J (2005) Remote sensing digital image analysis: an introduction, 4th edn. Springer-Verlag, BerlinGoogle Scholar
  33. Robinson DJ, Redding NJ, Crisp DJ (2002) Implementation of a fast algorithm for segmenting SAR imagery. Scientific and Technical Report, Defense Science and Technology Organization, Australia, January 1, 2002Google Scholar
  34. Rutherford P, Rapoza L (2008) An evaluation of an object-oriented paradigm for land use/land cover classification. Prof Geogr 60(1):87–100CrossRefGoogle Scholar
  35. SAF (2008) Windbreak. Accessed 15 May 2014
  36. Sorenson CJ, Marotz GA (1977) Changes in shelterbelt mileage statistics over four decades in Kansas. J Soil Water Conserv 32(6):276–281Google Scholar
  37. Tian J, Chen DM (2007) Optimization in multi-scale segmentation of high- resolution satellite images for artificial feature recognition. Int J Remote Sens 28(20):4625–4644CrossRefGoogle Scholar
  38. Tsai YH, Stow D, Weeks J (2011) Comparison of object-based image analysis approaches to mapping new buildings in accra, ghana using multi-temporal quickbird satellite imagery. Remote Sens 3:2707–2726CrossRefGoogle Scholar
  39. Tyndall J (2009) Characterizing pork producer demand for shelterbelts to mitigate odor: an Iowa case study. Agrofor Syst 77:205–221CrossRefGoogle Scholar
  40. United States Department of Agriculture (USDA). 1980. Field windbreak removals in five Great Plains States, 1970–1975. Government Printing Office, Washington, DCGoogle Scholar
  41. United States Department of Agriculture (USDA). 1994. Summary report 1992 National Resources Inventory. U.S. Government Printing Office, Washington, DCGoogle Scholar
  42. United States Department of Agriculture (USDA). 2007. Summary report 2007 National Resources Inventory. U.S. Government Printing Office, Washington, DCGoogle Scholar
  43. Williams K, Davis D (2013). National agriculture imagery program (NAIP) information sheet. United States Department of Agriculture. Accessed 27 Feb 2014
  44. Wiseman GS, Kort J, Walker DJ, Powers JS, Stefner CN (2007) Quantification of shelterbelt characteristics using high resolution imagery. When tree and crops get together. In: Proceedings of the 10th north American agroforestry conference, Quebec City, Canada, June 10–13Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2014

Authors and Affiliations

  • Kabita Ghimire
    • 1
    Email author
  • Mike W. Dulin
    • 2
  • Robert L. Atchison
    • 3
  • Douglas G. Goodin
    • 1
  • J. M. Shawn Hutchinson
    • 1
  1. 1.Department of GeographyKansas State UniversityManhattanUSA
  2. 2.US Army Corps of EngineersKansas CityUSA
  3. 3.Kansas Forest ServiceKansas State UniversityManhattanUSA

Personalised recommendations