Skip to main content

Advertisement

SpringerLink
Log in

Nonparametric machine learning for mapping forest cover and exploring influential factors

  • Research Article
  • Published:
Landscape Ecology Aims and scope Submit manuscript

Abstract

Context

The contribution of forest ecosystem services to human well-being varies over space following the dynamics in forest cover. Use of machine learning models is increasing in projecting forest cover changes and investigating the drivers, yet references are still lacking for selecting machine learning models for spatial projection of forest cover patterns.

Objectives

We assessed the ability of nonparametric machine learning techniques to project the spatial distribution of forest cover and identify its drivers using a case study of Tasmania, Australia.

Methods

We developed, evaluated, and compared the performance of four nonparametric machine learning models: support vector regression (SVR), artificial neural networks (ANN), random forest (RF), and gradient boosted regression trees (GBRT).

Results

The results demonstrated that RF far outperformed the other three models in both fitting and projection accuracy, and required less computional costs. GBRT outperformed SVR and ANN in projection accuracy. However, RF exhibited serious overfitting due to the full growth of its decision trees. The influence rankings of explanatory variables on spatial patterns of forest cover were different under the four models. Land tenure type and rainfall were identified among the top four most influential variables by all four models. The ranking produced by the RF model was significantly different with topographic factors associated with land clearing and production costs (elevation and distance to timber facilities) being the two most influential variables.

Conclusions

We encourage practitioners to consider nonparametric machine learning methods, especially RF, when facing problems of complex environmental data modelling.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3

Similar content being viewed by others

References

  • ABARES (2014) Tenure of Australia's Forest (2013) v2.0. Australian Bureau of Agricultural and Resource Economics and Sciences, Canberra

  • ACLEP (2014) National soil data provided by the Australian Collaborative Land Evaluation Program

  • Andam KS, Ferraro PJ, Pfaff A, Sanchez-Azofeifa GA, Robalino JA (2008) Measuring the effectiveness of protected area networks in reducing deforestation. Proc Natl Acad Sci 105(42):16089–16094

    CAS  PubMed  PubMed Central  Google Scholar 

  • Australian Bureau of Meteorology (2015b) Climate Data Online

  • Baskent EZ, Kadiogullari AI (2007) Spatial and temporal dynamics of land use pattern in Turkey: a case study in Inegol. Landsc Urban Plan 81(4):316–327

    Google Scholar 

  • Bergstra J, Bengio Y (2012) Random search for hyper-parameter optimization. J Mach Learn Res 13:281–305

    Google Scholar 

  • Breiman L, Friedman JH, Olshen RA, Stone CJ (1984) Classification and regression trees. Wadsworth, Belmont

    Google Scholar 

  • Breiman L (1996) Bagging predictors. Mach Learn 24(2):123–140

    Google Scholar 

  • Breiman L (2001) Random forests. Mach Learn 45(1):5–32

    Google Scholar 

  • Breiman L, Friedman J, Olshen R, Stone C (1984) Classification and regression trees, Wadsworth Int. Group 37(15):237–251

    Google Scholar 

  • Brereton RG, Lloyd GR (2010) Support vector machines for classification and regression. Analyst 135(2):230–267

    CAS  PubMed  Google Scholar 

  • Bryan BA (2013) High-performance computing tools for the integrated assessment and modelling of social–ecological systems. Environ Modell Softw 39:295–303

    Google Scholar 

  • Caccetta P, Furby S, Wallace J, Wu X, Richards G, Waterworth R (2012) Long-term monitoring of australian land cover change using landsat data. In: Proceedings of the Global Forest Monitoring from Earth Observation, pp 243–258

  • Chang C-C, Lin C-J (2011) LIBSVM: a library for support vector machines. ACM Trans Intell Syst Technol (TIST) 2(3):1–27

    Google Scholar 

  • Cohen WB, Maiersperger TK, Spies TA, Oetter DR (2001) Modelling forest cover attributes as continuous variables in a regional context with Thematic Mapper data. Int J Remote Sens 22(12):2279–2310

    Google Scholar 

  • Cracknell MJ, Reading AM (2014) Geological mapping using remote sensing data: a comparison of five machine learning algorithms, their response to variations in the spatial distribution of training data and the use of explicit spatial information. Comput Geosci 63:22–33

    Google Scholar 

  • Cutler DR, Edwards TC Jr, Beard KH, Cutler A, Hess KT, Gibson J, Lawler JJ (2007) Random forests for classification in ecology. Ecology 88(11):2783–2792

    PubMed  Google Scholar 

  • Department of Agriculture and Water Resources (2013) Australia's indigenous forest estate (2013)

  • Department of the Environment (2014) Collaborative Australian Protected Areas Database (CAPAD)

  • Dong M, Bryan BA, Connor JD, Nolan M, Gao L (2015) Land use mapping error introduces strongly-localised, scale-dependent uncertainty into land use and ecosystem services modelling. Ecosyst Serv 15:63–74

    Google Scholar 

  • Dormann CF, Elith J, Bacher S, Buchmann C, Carl G, Carré G, Marquéz JRG, Gruber B, Lafourcade B, Leitão PJ, Münkemüller T, McClean C, Osborne PE, Reineking B, Schröder B, Skidmore AK, Zurell D, Lautenbach S (2013) Collinearity: a review of methods to deal with it and a simulation study evaluating their performance. Ecography 36(1):27–46

    Google Scholar 

  • Du G, Shin KJ, Yuan L, Managi S (2018) A comparative approach to modelling multiple urban land use changes using tree-based methods and cellular automata: the case of Greater Tokyo Area. Int J Geograph Inf Sci 32(4):757–782

    Google Scholar 

  • Fan RE, Chen PH, Lin C-J (2005) Working set selection using second order information for training support vector machines. J Mach Learn Res 6:1889–1918

    Google Scholar 

  • Ferretti-Gallon K, Busch J (2014) What drives deforestation and what stops it? A meta-analysis of spatially explicit econometric studies. A Meta-Analysis of Spatially Explicit Econometric Studies

  • Foley JA, DeFries R, Asner GP, Barford C, Bonan G, Carpenter SR, Chapin FS, Coe MT, Daily GC, Gibbs HK, Helkowski JH, Holloway T, Howard EA, Kucharik CJ, Monfreda C, Patz JA, Prentice IC, Ramankutty N, Snyder PK (2005) Global consequences of land use. Science 309(5734):570–574

    CAS  PubMed  Google Scholar 

  • Forest Practices Authority (2017) State of the forests Tasmania 2017, booklet by the Forest Practices Authority, Hobart, Tasmania

  • Freeman EA, Moisen GG, Coulston JW, Wilson BT (2015) Random forests and stochastic gradient boosting for predicting tree canopy cover: comparing tuning processes and model performance. Can J For Res 46(3):323–339

    Google Scholar 

  • Freitas SR, Hawbaker TJ, Metzger JP (2010) Effects of roads, topography, and land use on forest cover dynamics in the Brazilian Atlantic Forest. For Ecol Manag 259(3):410–417

    Google Scholar 

  • Freund Y (1996) Schapire RE Experiments with a new boosting algorithm. Icml 96:148–156

    Google Scholar 

  • Friedman JH (2001) Greedy function approximation: a gradient boosting machine. Ann Stat 25:1189–1232

    Google Scholar 

  • Friedman JH (2002) Stochastic gradient boosting. Comput Stat Data Anal 38(4):367–378

    Google Scholar 

  • Gallant J, Wilson N, Dowling T, Read A, Inskeep C (2011) SRTM-derived one second digital elevation models version 1.0. Geoscience Australia, Canberra

    Google Scholar 

  • Gao L, Bryan BA (2016) Incorporating deep uncertainty into the elementary effects method for robust global sensitivity analysis. Ecol Modell 321:1–9

    Google Scholar 

  • Gao L, Bryan BA (2017) Finding pathways to national-scale land-sector sustainability. Nature 544(7649):217–222

    CAS  PubMed  Google Scholar 

  • Gao L, Bryan BA, Liu J, Li W, Chen Y, Liu R, Barrett D (2017) Managing too little and too much water: robust mine-water management strategies under variable climate and mine conditions. J Clean Prod 162:1009–1020

    Google Scholar 

  • Gao L, Bryan BA, Nolan M, Connor JD, Song X, Zhao G (2016) Robust global sensitivity analysis under deep uncertainty via scenario analysis. Environ Modell Softw 76:154–166

    Google Scholar 

  • Gao L, Ding Y-S, Ren L-H (2004) A novel ecological network-based computation platform as grid middleware system. Int J Intell Syst 19(10):859–884

    Google Scholar 

  • Geurts P, Ernst D, Wehenkel L (2006) Extremely randomized trees. Mach Learn 63(1):3–42

    Google Scholar 

  • Giriraj A, Irfan-Ullah M, Murthy MSR, Beierkuhnlein C (2008) Modelling spatial and temporal forest cover change patterns (1973–2020): a case study from South Western Ghats (India). Sensors 8(10):6132–6153

    PubMed  PubMed Central  Google Scholar 

  • GISCA (2001) Accessibility/Remoteness Index of Australia (ARIA). Canberra

  • Gleason CJ, Im J (2012) Forest biomass estimation from airborne LiDAR data using machine learning approaches. Remote Sens Environ 125:80–91

    Google Scholar 

  • Hansen MC, DeFries RS (2004) Detecting long-term global forest change using continuous fields of tree-cover maps from 8-km advanced very high resolution radiometer (AVHRR) data for the years 1982–99. Ecosystems 7(7):695–716

    Google Scholar 

  • Hastie T, Tibshirani R, Friedman J (2009) Unsupervised learning. The elements of statistical learning. Springer, New York, pp 485–585

    Google Scholar 

  • Hu X, Wu C, Hong W, Qiu R, Li J, Hong T (2014) Forest cover change and its drivers in the upstream area of the Minjiang River, China. Ecol Indic 46:121–128

    Google Scholar 

  • Huang C, Song K, Kim S, Townshend JR, Davis P, Masek JG, Goward SN (2008) Use of a dark object concept and support vector machines to automate forest cover change analysis. Remote Sens Environ 112(3):970–985

    Google Scholar 

  • Huang X, Gao L, Crosbie RS, Zhang N, Fu G, Doble R (2019) Groundwater recharge prediction using linear regression, multi-layer perception network, and deep learning. Water 11(9):1879

    Google Scholar 

  • Jain RK, Smith KM, Culligan PJ, Taylor JE (2014) Forecasting energy consumption of multi-family residential buildings using support vector regression: investigating the impact of temporal and spatial monitoring granularity on performance accuracy. Appl Energy 123:168–178

    Google Scholar 

  • Jiang Z, Mallants D, Peeters L, Gao L, Soerensen C, Mariethoz G (2019) High-resolution paleovalley classification from airborne electromagnetic imaging and deep neural network training using digital elevation model data. Hydrol Earth Syst Sci 23(6):2561–2580

    Google Scholar 

  • Jones DA, Wang W, Fawcett R (2009) High-quality spatial climate data-sets for Australia. Aust Meteorol Oceanogr J 58(4):233–248

    Google Scholar 

  • Keerthi SS, Shevade SK, Bhattacharyya C, Murthy KRK (2001) Improvements to Platt's SMO algorithm for SVM classifier design. Neural Computat 13(3):637–649

    Google Scholar 

  • Kingma DP, Ba J (2014) Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980

  • Kohavi R (1995) A study of cross-validation and bootstrap for accuracy estimation and model selection. In: Proceedings of the Ijcai. vol 14. Montreal, Canada, pp 1137–1145

  • Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. In: Advances in neural information processing systems, pp 1097–1105

  • Kumar R, Nandy S, Agarwal R, Kushwaha S (2014) Forest cover dynamics analysis and prediction modeling using logistic regression model. Ecol Indic 45:444–455

    Google Scholar 

  • LeCun Y, Bengio Y, Hinton G (2015) Deep learning. Nature 521(7553):436–444

    CAS  PubMed  Google Scholar 

  • LeCun YA, Bottou L, Orr GB, Müller K-R (2012) Efficient backprop. Neural networks: Tricks of the trade. Springer, New York, pp 9–48

    Google Scholar 

  • Lehmann EA, Wallace JF, Caccetta PA, Furby SL, Zdunic K (2013) Forest cover trends from time series Landsat data for the Australian continent. Int J Appl Earth Observ Geoinf 21:453–462

    Google Scholar 

  • Leinenkugel P, Wolters ML, Kuenzer C, Oppelt N, Dech S (2014) Sensitivity analysis for predicting continuous fields of tree-cover and fractional land-cover distributions in cloud-prone areas. Int J Remote Sens 35(8):2799–2821

    Google Scholar 

  • Lin S, Jiang Y, He J, Ma G, Xu Y, Jiang H (2010s) Changes in the spatial and temporal pattern of natural forest cover on Hainan Island from the 1950s to the 2010s: implications for natural forest conservation and management. PeerJ 5:e3320

    PubMed  PubMed Central  Google Scholar 

  • Lin Y-P, Chu H-J, Wu C-F, Verburg PH (2011) Predictive ability of logistic regression, auto-logistic regression and neural network models in empirical land-use change modeling–a case study. Int J Geographys Inf Sci 25(1):65–87

    Google Scholar 

  • Ludwig M, Morgenthal T, Detsch F, Higginbottom TP, Valdes ML, Nauß T, Meyer H (2019) Machine learning and multi-sensor based modelling of woody vegetation in the Molopo Area, South Africa. Remote Sens Environ 222:195–203

    Google Scholar 

  • Makinano-Santillan MM, Santillan JR, Paringit EC (2001) Merging landsat image information with georeferenced biophysical and socio-economical datasets to describe forest cover change in a philippine province

  • Marcos-Martinez R, Bryan BA, Connor JD, King D (2017) Agricultural land-use dynamics: assessing the relative importance of socioeconomic and biophysical drivers for more targeted policy. Land Use Policy 63:53–66

    Google Scholar 

  • Marcos-Martinez R, Bryan BA, Schwabe KA, Connor JD, Law EA (2018) Forest transition in developed agricultural regions needs efficient regulatory policy. For Policy Econ 86:67–75

    Google Scholar 

  • Marcos-Martinez R, Bryan BA, Schwabe KA, Connor JD, Law EA, Nolan M, Sánchez JJ (2019) Projected social costs of CO2 emissions from forest losses far exceed the sequestration benefits of forest gains under global change. Ecosyst Serv 37:100935

    Google Scholar 

  • Marinoni O, Navarro Garcia J, Marvanek S, Prestwidge D, Clifford D, Laredo LA (2012) Development of a system to produce maps of agricultural profit on a continental scale: an example for Australia. Agric Syst 105(1):33–45

    Google Scholar 

  • Mas J-F, Puig H, Palacio JL, Sosa-López A (2004) Modelling deforestation using GIS and artificial neural networks. Environ Modell Softw 19(5):461–471

    Google Scholar 

  • Mayfield H, Smith C, Gallagher M, Hockings M (2017) Use of freely available datasets and machine learning methods in predicting deforestation. Environ Modell Softw 87:17–28

    Google Scholar 

  • McMichael A, Scholes R, Hefny M, Pereira E, Palm C, Foale S (2005) Linking ecosystem services and human well-being. In: Capistrano D, Samper K, Cristián L, Marcus J, Raudsepp-Hearne C (eds) Ecosystems and human well-being: multi-scale assessments, MIllenium Ecosystem Assessment Series 4. Island Press, Washington DC, pp 43–60

    Google Scholar 

  • Micheletti N, Foresti L, Robert S, Leuenberger M, Pedrazzini A, Jaboyedoff M, Kanevski M (2014) Machine learning feature selection methods for landslide susceptibility mapping. Math Geosci 46(1):33–57

    Google Scholar 

  • Nahib I, Suryanta J (2017) Forest cover dynamics analysis and prediction modelling using logistic regression model (case study: forest cover at Indragiri Hulu Regency, Riau Province). In: IOP Conference Series: Earth and Environmental Science. vol 54. IOP Publishing, p 012044

  • Opitz D, Maclin R (1999) Popular ensemble methods: an empirical study. J Artif Intell Res 11:169–198

    Google Scholar 

  • Palmate S, Pandey A, Kumar D, Pandey R, Mishra S (2017) Climate change impact on forest cover and vegetation in Betwa Basin, India. Appl Water Sci 7(1):103–114

    Google Scholar 

  • Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, Prettenhofer P, Weiss R, Dubourg V, Vanderplas J, Passos A, Cournapeau D, Brucher M, Perrot M, Duchesnay E (2011) Scikit-learn: machine learning in Python. J Mach Learn Res 12:2825–2830

    Google Scholar 

  • Peeters LJM, Pagendam DE, Crosbie RS, Rachakonda PK, Dawes WR, Gao L, Marvanek SP, Zhang YQ, McVicar TR (2018) Determining the initial spatial extent of an environmental impact assessment with a probabilistic screening methodology. Environ Modell Softw 109:353–367

    Google Scholar 

  • Pijanowski BC, Brown DG, Shellito BA, Manik GA (2002) Using neural networks and GIS to forecast land use changes: a land transformation model. Comput Environ Urban Syst 26(6):553–575

    Google Scholar 

  • Platt JC (1999) 12 fast training of support vector machines using sequential minimal optimization. In: Schölkopf B, Burges CJC, Smola AJ (eds) Advances in kernel methods. MIT Press, Cambridge, pp 185–208

    Google Scholar 

  • Pressey R, Whish G, Barrett T, Watts M (2002) Effectiveness of protected areas in north-eastern New South Wales: recent trends in six measures. Biol Conserv 106(1):57–69

    Google Scholar 

  • Russell SJ, Norvig P (2016) Artificial intelligence: a modern approach. Pearson Education Limited, Malaysia

    Google Scholar 

  • Schwieder M, Leitão P, Suess S, Senf C, Hostert P (2014) Estimating fractional shrub cover using simulated EnMAP data: a comparison of three machine learning regression techniques. Remote Sens 6(4):3427–3445

    Google Scholar 

  • Simard M, Pinto N, Fisher JB, Baccini A (2011) Mapping forest canopy height globally with spaceborne lidar. J Geophys Res 116(G4):G04021

    Google Scholar 

  • Smola AJ, Schölkopf B (2004) A tutorial on support vector regression. Stat Comput 14(3):199–222

    Google Scholar 

  • Sola J, Sevilla J (1997) Importance of input data normalization for the application of neural networks to complex industrial problems. IEEE Trans Nucl Sci 44(3):1464–1468

    Google Scholar 

  • Sun H, Wang Q, Wang G, Lin H, Luo P, Li J, Zeng S, Xu X, Ren L (2018) Optimizing kNN for mapping vegetation cover of arid and semi-arid areas using landsat images. Remote Sens 10(8):1248

    Google Scholar 

  • Taieb SB, Hyndman RJ (2014) A gradient boosting approach to the Kaggle load forecasting competition. Int J Forecast 30(2):382–394

    Google Scholar 

  • Wang X, Huang H, Gong P, Biging GS, Xin Q, Chen Y, Yang J, Liu C (2016) Quantifying multi-decadal change of planted forest cover using airborne LiDAR and Landsat imagery. Remote Sens 8(1):62

    Google Scholar 

  • Wei B, Hao K, Gao L, Tang X (2020) Bio-inspired visual integrated model for multi-label classification of textile defect images. IEEE Trans Cognit Dev Syst. https://doi.org/10.1109/TCDS.2020.2977974

    Article  Google Scholar 

  • Were K, Bui DT, Dick ØB, Singh BR (2015) A comparative assessment of support vector regression, artificial neural networks, and random forests for predicting and mapping soil organic carbon stocks across an Afromontane landscape. Ecol Indic 52:394–403

    CAS  Google Scholar 

  • Wheeler D, Hammer D, Kraft R, Dasgupta S, Blankespoor B (2013) Economic dynamics and forest clearing: a spatial econometric analysis for Indonesia. Ecol Econ 85:85–96

    Google Scholar 

  • Ye L, Gao L, Marcos-Martinez R, Mallants D, Bryan BA (2019) Projecting Australia's forest cover dynamics and exploring influential factors using deep learning. Environ Modell Softw 119:407–417

    Google Scholar 

  • Zhou W, Huang G, Pickett STA, Cadenasso ML (2011) 90 years of forest cover change in an urbanizing watershed: spatial and temporal dynamics. Landsc Ecol 26(5):645

    Google Scholar 

Download references

Acknowledgements

B. Liu and B. Li were supported by the Fundamental Research Funds for the Central Universities (20CX05006A). L. Gao was supported by a CSIRO Julius award and a CSIRO 2018/19 Land and Water Appropriation Project. The authors thank Dr. Xin Huang for his technical support in the map visualization work. The authors would also like to thank two anonymous reviewers for their constructive comments, which have been very helpful for improving this manuscript.

Author information

Authors and Affiliations

  1. College of Information and Control Engineering, China University of Petroleum (East China), Qingdao, 266580, China

    Bao Liu & Baoan Li

  2. CSIRO, Waite Campus, Urrbrae, SA, 5064, Australia

    Lei Gao

  3. School of Business, Shandong Normal University, Jinan, 250014, China

    Lei Gao

  4. CSIRO, Black Mountain, ACT, 2601, Australia

    Raymundo Marcos-Martinez

  5. Centre for Integrative Ecology, Deakin University, Melbourne Burwood Campus, Burwood, VIC, 3125, Australia

    Brett A. Bryan

Authors
  1. Bao Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Lei Gao
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Baoan Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Raymundo Marcos-Martinez
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Brett A. Bryan
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Lei Gao.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary file1 (DOCX 129 kb)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Liu, B., Gao, L., Li, B. et al. Nonparametric machine learning for mapping forest cover and exploring influential factors. Landscape Ecol 35, 1683–1699 (2020). https://doi.org/10.1007/s10980-020-01046-0

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10980-020-01046-0

Keywords

Search

Navigation