A two-scale approach for estimating forest aboveground biomass with optical remote sensing images in a subtropical forest of Nepal

Abstract

Forests account for 80% of the total carbon exchange between the atmosphere and terrestrial ecosystems. Thus, to better manage our responses to global warming, it is important to monitor and assess forest aboveground carbon and forest aboveground biomass (FAGB). Different levels of detail are needed to estimate FAGB at local, regional and national scales. Multi-scale remote sensing analysis from high, medium and coarse spatial resolution data, along with field sampling, is one approach often used. However, the methods developed are still time consuming, expensive, and inconvenient for systematic monitoring, especially for developing countries, as they require vast numbers of field samples for upscaling. Here, we recommend a convenient two-scale approach to estimate FAGB that was tested in our study sites. The study was conducted in the Chitwan district of Nepal using GeoEye-1 (0.5 m), Landsat (30 m) and Google Earth very high resolution (GEVHR) Quickbird (0.65 m) images. For the local scale (Kayerkhola watershed), tree crowns of the area were delineated by the object-based image analysis technique on GeoEye images. An overall accuracy of 83% was obtained in the delineation of tree canopy cover (TCC) per plot. A TCC vs. FAGB model was developed based on the TCC estimations from GeoEye and FAGB measurements from field sample plots. A coefficient of determination (R ²) of 0.76 was obtained in the modelling, and a value of 0.83 was obtained in the validation of the model. To upscale FAGB to the entire district, open source GEVHR images were used as virtual field plots. We delineated their TCC values and then calculated FAGB based on a TCC versus FAGB model. Using the multivariate adaptive regression splines machine learning algorithm, we developed a model from the relationship between the FAGB of GEVHR virtual plots with predictor parameters from Landsat 8 bands and vegetation indices. The model was then used to extrapolate FAGB to the entire district. This approach considerably reduced the need for field data and commercial very high resolution imagery while achieving two-scale forest information and FAGB estimates at high resolution (30 m) and accuracy (R ² = 0.76 and 0.7) with minimal error (RMSE = 64 and 38 tons ha⁻¹) at local and regional scales. This methodology is a promising technique for cost-effective FAGB and carbon estimations and can be replicated with limited resources and time. The method is especially applicable for developing countries that have low budgets for carbon estimations, and it is also applicable to the Reducing Emissions from Deforestation and Forest Degradation (REDD +) monitoring reporting and verification processes.

Keywords

Forest aboveground biomass / Google Earth imagery / Multi-scale remote sensing / Virtual plot / Optical imagery

References

[1] Abubakari Z, Molen P, Bennett RM, Kuusaana ED. Land consolidation, customary lands, and Ghana’s Northern Savannah Ecological Zone: an evaluation of the possibilities and pitfalls. Land Use Policy, 2016, 54(2): 386-398.

[2] Avitabile V, Baccini A, Friedl MA, Schmullius C. Capabilities and limitations of Landsat and land cover data for aboveground woody biomass estimation of Uganda. Remote Sens Environ, 2012, 117(3): 366-380.

[3] Baccini A, Goetz SJ, Walker WS, Laporte NT, Sun M, Sulla D. Estimated carbon dioxide emissions from tropical deforestation improved by carbon-density maps. Nat Clim Change, 2012, 2(1): 182-185.

[4] Baig MHA, Zhang LF, Tong Shuai, Tong QX. Derivation of a tasselled cap transformation based on Landsat 8 at- satellite reflectance. Remote Sens Lett, 2014, 5(5): 423-431.

[5] Barbier N, Couteron P, Proisy C, Malhi Y, Gastellu JP. The variation of apparent crown size and canopy heterogeneity across lowland Amazonian forests. Glob Ecol Biogeogr, 2010, 19(1): 72-84.

[6] Bastin JF, Barbier N, Couteron P, Adams B, Shapiro A, Bogaert J, Cannière C. Aboveground biomass mapping of African forest mosaics using canopy texture analysis: toward a regional approach. Ecol Appl, 2014, 24(8): 1984-2001.

[7] Breiman L. Random forests. Mach Learn, 2001, 45(1): 5-32.

[8] Chalise D, Rawal N, Shriwastac CP (2014) Characterization and mapping of soils of Chitwan. https://www.researchgate.net/publication/291832686_CHARACTERIZATION_AND_MAPPING_OF_SOILS_OF_CHITWAN. Accessed 16 May 2017

[9] Chojnacky DC, Heath LS. Estimating down deadwood from FIA forest inventory variables in Maine. Environ Pollut, 2002, 116(1): 25-30.

[10] Culvenor DS. TIDA: an algorithm for the delineation of tree crowns in high spatial resolution remotely sensed imagery. Comput Geosci, 2002, 28(1): 33-44.

[11] Cutler MEJ, Boyd DS, Foody GM, Vetrivel A. Estimating tropical forest biomass with a combination of SAR image texture and Landsat TM data: an assessment of predictions between regions. ISPRS J Photogramm Remote Sens, 2012, 70(4): 66-77.

[12] DeFries R, Achard F, Brown S, Herold M, Murdiyarso D, Schlamadinger B, Souza C. Earth observations for estimating greenhouse gas emissions from deforestation in developing countries. Environ Sci Policy, 2007, 10(4): 385-394.

[13] Dorais A, Cardille J. Strategies for incorporating high-resolution google earth databases to guide and validate classifications : understanding deforestation in borneo. Remote Sens, 2011, 3(6): 1157-1176.

[14] Eckert S. Improved forest biomass and carbon estimations using texture measures from worldView-2 satellite data. Remote Sens, 2012, 4(12): 810-829.

[15] Elshayal (2017) Elshayal Smart GIS software. http://elshayal-smart.en.lo4d.com/. Accessed 18 May 2017

[16] FAO. Global forest resources assessment 2010, 2010, Rome: FAO 122

[17] Frazier RJ, Coops NC, Wulder MA, Kennedy R. Characterization of aboveground biomass in an unmanaged boreal forest using Landsat temporal segmentation metrics. ISPRS J Photogramm Remote Sens, 2014, 92(6): 137-146.

[18] Fuchs H, Magdon P, Kleinn C, Flessa H. Estimating aboveground carbon in a catchment of the Siberian forest tundra: combining satellite imagery and field inventory. Remote Sens Environ, 2009, 113(3): 518-531.

[19] Gates DM. Climate change and the response of forests. Int J Remote Sens, 1990, 11(7): 1095-1107.

[20] Gibbs HK, Brown S, Niles JO, Foley J. Monitoring and estimating tropical forest carbon stocks: making REDD a reality. Environ Res Lett, 2007, 2(4): 45-49.

[21] Gilani H, Gautam SK, Murthy MSR, Koju UA, Uddin K, Karky B. Monitoring the performance of community forestry to achieve REDD+ Goals through geospatial methods. Int Arch Photogramm Remote Sens Spat Inf Sci, 2014, 8(1): 1295-1300.

[22] Glenn NF, Neuenschwander A, Vierling LA, Spaete L, Shinneman DJ, McIlroy SK. Landsat 8 and ICESat-2: performance and potential synergies for quantifying dryland ecosystem vegetation cover and biomass. Remote Sens Environ, 2016, 185(11): 233-242.

[23] Goetz SJ, Baccini A, Laporte NT, Johns T, Walker W, Kellndorfer J, Sun M. Mapping and monitoring carbon stocks with satellite observations: a comparison of methods. Carbon Balance Manage, 2009, 4(1): 112-121.

[24] Goodman RC, Phillips OL, Baker TR. The importance of crown dimensions to improve tropical tree biomass estimates. Ecol Appl, 2014, 24(4): 680-698.

[25] Guneralp I, Filippi AM, Randall J. Estimation of floodplain aboveground biomass using multispectralremote sensing and nonparametric modelling. Int J Appl Earth Obs Geoinf, 2014, 33(1): 119-126.

[26] Hanes JM (2014) Biophysical applications of satellite remote sensing. http://doi.org/10.1007/978-3-642-25047-7. Accessed 8 May 2017

[27] Healey SP, Yang Z, Cohen WB, Pierce DJ. Application of two regression-based methods to estimate the effects of partial harvest on forest structure using Landsat data. Remote Sens Environ, 2006, 101(91): 115-126.

[28] Hu Q, Wu WB, Xia T, Yu QY, Yang P, Li ZG, Song Q. Exploring the use of google earth imagery and object-based methods in land use/cover mapping. Remote Sens, 2013, 5(11): 6026-6042.

[29] Husch B, Beers TW, Kershaw JA. Forest mensuration, 2003, New York: Wiley.

[30] Hussin YA, Gilani H. Mapping carbon stocks in community forests of Nepal using high spatial resolution satellite images, 2011, Patan: ICIMOD, Sustainable Mountain Development 125 135

[31] Hussin YA, Gilani H, Leeuwen L, Murthy MSR, Shah R, Baral S, Qamer FM. Evaluation of object-based image analysis techniques on very high-resolution satellite image for biomass estimation in a watershed of hilly forest of Nepal. Appl Geomatics, 2014, 6(1): 59-68.

[32] Ismail R, Mutanga O. A comparison of regression tree ensembles: Predicting Sirex noctilio induced water stress in Pinus patula forests of KwaZulu-Natal, South Africa. Int J Appl Earth Obs Geoinf, 2010, 12(1): 45-51.

[33] Kamusoko C, Gamba J, Murakami H. Mapping woodland cover in the Miombo ecosystem: a comparison of machine learning classifiers. Land, 2014, 3(2): 524-540.

[34] Kankeu RS, Jean D, Richard S, Marle N. Quantifying post logging biomass loss using satellite images and ground measurements in Southeast Cameroon. J For Res, 2016, 27(6): 1415-1426.

[35] Kapfer PM, Streby HM, Gurung B, Simcharoen A, McDougal CC, Smith JLD. Fine-scale spatio-temporal variation in tiger Panthera tigris diet: effect of study duration and extent on estimates of tiger diet in Chitwan National Park, Nepal. Wildl Biol, 2011, 17(3): 277-285.

[36] Karlson M, Ostwald M, Reese H, Sanou J, Tankoano B, Karlson M, Tankoano B. Mapping tree canopy cover and aboveground biomass in Sudano-Sahelian woodlands using landsat 8 and random forest. Remote Sens, 2015, 7(8): 10017-10041.

[37] Karna YK, Hussin YA, Gilani H, Bronsveld MC, Murthy MSR, Qamer FM, Baniya CB. Integration of WorldView-2 and airborne LiDAR data for tree species level carbon stock mapping in Kayar Khola watershe, Nepal. Int J Appl Earth Obs Geoinf, 2015, 38(6): 280-291.

[38] Ke YH, Quackenbush LJ (2008) Comparison of individual tree crown detection and delineation methods. In: ASPRS 2008 annual conference proceedings, 11. http://www.asprs.org/a/publications/proceedings/portland08/0038.pdf. Accessed 28 May 2017

[39] Ke YH, Quackenbush LJ. A review of methods for automatic individual tree-crown detection and delineation from passive remote sensing. Int J Remote Sens, 2011, 32(17): 4725-4747.

[40] Kim SR, Kwak DA, Lee WK, Son Y, Bae W, Yoo S. Estimation of carbon storage based on individual tree detection in Pinus densiflora stands using a fusion of aerial photography and LiDAR data. Sci China. Life Sci, 2010, 53(7): 885-897.

[41] Landsat (2017) Using the USGS Landsat 8 Product. https://landsat.usgs.gov/using-usgs-landsat-8-product. Accessed 08 Jan 2017

[42] Larsson H. Linear regressions for canopy cover estimation in Acacia woodlands using Landsat-TM, -MSS and SPOT HRV XS data. Int J Remote Sens, 2007, 48(1): 37-41.

[43] Leckie DG, Gougeon FA, Tinis S, Nelson T, Burnett CN, Paradine D. Automated tree recognition in old growth conifer stands with high resolution digital imagery. Remote Sens Environ, 2005, 94(3): 311-326.

[44] Lu DS. The potential and challenge of remote sensing—based biomass estimation. Int J Remote Sens, 2006, 27(7): 1297-1328.

[45] Lu DS, Mausel P, Brond E, Moran E. Relationships between forest stand parameters and Landsat TM spectral responses in the Brazilian Amazon Basin. For Ecol Manag, 2004, 198(3): 149-167.

[46] Ludwig A, Meyer H, Nauss T. Automatic classification of Google Earth images for a larger scale monitoring of bush encroachment in South Africa. Int J Appl Earth Obs Geoinf, 2016, 50(3): 89-94.

[47] MacDicken KG. A guide to monitoring carbon storage in forestry and agroforestry projects, 1997, Morrilton: Winnrocak Int. Special publication 182

[48] Malarvizhi K, Kumar SV, Porchelvan P. Use of high resolution google earth satellite imagery in landuse map preparation for urban related applications. Procedia Technol, 2016, 24(2): 1835-1842.

[49] Mcroberts RE, Tomppo EO, Czaplewski RL. Sampling designs for national forest assessments, 1992, Rome: FAO 121

[50] Ming QZ, Guo SR, Jiao YM. High gradient effects of forest biomass energy in mountainous region: a case of Meili Snow Mountain. Procedia Earth Planet Sci, 2011, 2(1): 315-320.

[51] Murthy MSR, Wesselman SG (2015) Multi-scale forest biomass assessment and monitoring in the Hindu Kush Himalayan region: a geospatial perspective. ICIMOD, p 16

[52] Platt RV, Schoennagel T. An object-oriented approach to assessing changes in tree cover in the Colorado Front. For Ecol Manag, 2017, 258(7): 1342-1349.

[53] Ploton P, Pélissier R, Proisy C, Flavenot T, Barbier N, Rai SN, Couteron P. Assessing aboveground tropical forest biomass using Google Earth canopy images. Ecol Appl, 2012, 22(3): 993-1003.

[54] Pot J (2016) Google Earth Pro used to cost $400 a year—here’s how to get it for free. https://www.digitaltrends.com/computing/how-to-get-google-earth-pro-for-free/. Accessed 11 Nov 2016

[55] Rahman EM, Berg M, Way MJ, Ahmed FB. Hand-held spectrometry for estimating thrips (Fulmekiola serrata) incidence in sugarcane. Geosci Remote Sens, 2009, 63(1): 268-271.

[56] Shao G. Richardson D, Castree N, Goodchild MF, Kobayashi A, Liu W, Marston RA. Optical remote sensing. International encyclopedia of geography: people, the earth, environment, and technology, 2016, Hoboken: Wiley 2390 2395

[57] Sharma ER, Pukkala T. Volume and biomass prediction equations of forest trees of Nepal, 1990, Kathmandu: Forest Survey and Statistical Division. Ministry of Forest and Soil Conservation 82 88

[58] Shih FY, Cheng SX. Automatic seeded region growing for color image segmentation. Image Vis Comput, 2005, 23(10): 877-886.

[59] Singh M, Evans D, Friess DA, Tan BS. Google Earth. Remote Sens, 2015, 7(5): 5057-5076.

[60] Smil V. Energy myths and realities: bringing science to the energy policy debate, 2010, Lanham: Government Institutes Press 182

[61] Spanner M, Johnson L, Miller J, Mccreight R, Applications SE, May N, Gong P. Remote sensing of seasonal leaf area index across the Oregon Transect Freemantle. Ecol Appl, 1994, 4(2): 258-271.

[62] Stocker TF, Qin GK, Plattner M, Tignor SK, Allen J, Boschung A, Nauels Y, Xia VB. IPCC, 2013: climate change 2013: the physical science basis. Contribution of working group I to the fifth assessment report of the intergovernmental panel on climate change, 2013, Cambridge: Cambridge University Press 1535

[63] Takahashi T, Awaya Y, Hirata Y, Furuya N, Sakai T, Sakai A. Stand volume estimation by combining low laser-sampling density LiDAR data with QuickBird panchromatic imagery in closed-canopy Japanese cedar ( Cryptomeria japonica ) plantations. Int J Remote Sens, 2010, 31(5): 1281-1301.

[64] Uddin K, Lal H, Murthy MSR, Bajracharya B, Shrestha B, Gilani H, Dango B. Development of 2010 national land cover database for the Nepal. J Environ Manag, 2014, 148(1): 82-90.

[65] Uddin K, Gilani H, Murthy MSR, Kotru R, Qamer FM. Forest condition monitoring using very-high-resolution satellite imagery in a Remote Mountain watershed in Nepal. Mt Res Dev, 2015, 35(3): 264-277.

[66] Wu WC, Pauw E, Helldén U. Assessing woody biomass in African tropical savannahs by multiscale remote sensing. Int J Remote Sens, 2013, 34(13): 4525-4549.

[67] Wu CF, Shen HH, Wang K, Shen AH, Deng JS, Gan MY. Landsat imagery-based above ground biomass estimation and change investigation related to human activities. Sustainability, 2016, 8(2): 159-165.

[68] Wulder MA, Skakun WAK. Estimating time since forest harvest using segmented landsat ETM+ imagery. Remote Sens Environ, 2004, 93(1–2): 179-187.

[69] Yan EP, Lin H, Wang GX, Sun H (2014) Multi-scale simulation and accuracy assessment of forest carbon using Landsat and MODIS data. Conference paper on Earth observation and Remote sensing, Changsha, China. http://ieeexplore.ieee.org/document/6927877/. Accessed 28 Aug 2017

[70] Yunus U, Ibrahim O, Tufan O, Serhun D. Comparison of satellite images with different spatial resolutions to estimate stand structural diversity in urban forests. J For Res, 2017, 185(11): 233-242.

[71] Zhu XL, Liu DS. Improving forest aboveground biomass estimation using seasonal Landsat NDVI time-series. ISPRS J Photogramm Remote Sens, 2015, 102(4): 222-231.