Predicting the spectral information of future land cover using machine learning

ABSTRACT Application of machine learning models to study land-cover change is typically restricted to the change detection of categorical, i.e. classified, land-cover data. In this study, our aim is to extend the utility of such models to predict the spectral band information of satellite images. A Random Forests (RF)-based machine learning model is trained using topographic and historical climatic variables as inputs to predict the spectral band values of high-resolution satellite imagery across two large sites in the western United States, New Mexico (10,570 km2), and Washington (9400 km2). The model output is used to obtain a true colour photorealistic image and an image showing the normalized difference vegetation index values. We then use the trained model to explore what the land cover might look like for a climate change scenario during the 2061–2080 period. The RF model achieves high validation accuracy for both sites during the training phase, with the coefficient of determination (R2) = 0.79 for New Mexico site and R2 = 0.73 for Washington site. For the climate change scenario, prominent land-cover changes are characterized by an increase in the vegetation cover at the New Mexico site and a decrease in the perennial snow cover at the Washington site. Our results suggest that direct prediction of spectral band information is highly beneficial due to the ability it provides for deriving ecologically relevant products, which can be used to analyse land-cover change scenarios from multiple perspectives.

[1]  Dorothy K. Hall,et al.  Detection of earlier snowmelt in the Wind River Range, Wyoming, using Landsat imagery, 1972–2013 , 2015 .

[2]  J. Kerr,et al.  From space to species: ecological applications for remote sensing , 2003 .

[3]  David M. Wolock,et al.  STATSGO soil characteristics for the conterminous United States , 1997 .

[4]  M. Floyd,et al.  Effects of Historic Livestock Grazing on Vegetation at Chaco Culture National Historic Park, New Mexico , 2003 .

[5]  Janet Franklin,et al.  Mapping land-cover modifications over large areas: A comparison of machine learning algorithms , 2008 .

[6]  Katharine Hayhoe,et al.  Past and projected future changes in snowpack and soil frost at the Hubbard Brook Experimental Forest, New Hampshire, USA , 2010 .

[7]  A. Kiviste,et al.  A conceptual model of forest stand development based on permanent sample-plot data in Estonia , 2005 .

[8]  Mark A. Friedl,et al.  Using prior probabilities in decision-tree classification of remotely sensed data , 2002 .

[9]  J. L. Parra,et al.  Very high resolution interpolated climate surfaces for global land areas , 2005 .

[10]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[11]  E Brown de Colstoun,et al.  National Park vegetation mapping using multitemporal Landsat 7 data and a decision tree classifier , 2003 .

[12]  J. L. Foster,et al.  HEI SIGNIFICANCE OF THE DATE OF SNOW DISAPPEARANCE ON ITHE ARCTIC TUNDRA AS A POSSIBLE INDICATOR OF CLIMATE CHANGE , 1989 .

[13]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[14]  Radford M. Neal.  Pattern Recognition and Machine Learning , 2007, Technometrics.

[15]  Robert S. Stone,et al.  Earlier spring snowmelt in northern Alaska as an indicator of climate change , 2002 .

[16]  Ranga B. Myneni,et al.  Remote sensing of vegetation and land-cover change in Arctic Tundra Ecosystems , 2004 .

[17]  Christina L. Tague,et al.  Modeling the eco-hydrologic response of a Mediterranean type ecosystem to the combined impacts of projected climate change and altered fire frequencies , 2009 .

[18]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

[19]  C. Schmullius,et al.  Effects of soil freezing and thawing on vegetation carbon density in Siberia: A modeling analysis with the Lund‐Potsdam‐Jena Dynamic Global Vegetation Model (LPJ‐DGVM) , 2007 .

[20]  Ian H. Witten,et al.  Data mining: practical machine learning tools and techniques with Java implementations , 2002, SGMD.

[21]  Marcel Abendroth,et al.  Data Mining Practical Machine Learning Tools And Techniques With Java Implementations , 2016 .

[22]  Cameron L. Aldridge,et al.  Forecasting sagebrush ecosystem components and greater sage-grouse habitat for 2050: Learning from past climate patterns and Landsat imagery to predict the future , 2015 .

[23]  T. Dawson,et al.  Predicting the impacts of climate change on the distribution of species: are bioclimate envelope models useful? , 2003 .

[24]  J. Townshend,et al.  Global land cover classi(cid:142) cation at 1 km spatial resolution using a classi(cid:142) cation tree approach , 2004 .

[25]  C. Field,et al.  The velocity of climate change , 2009, Nature.

[26]  Ian H. Witten,et al.  Practical machine learning and its application to problems in agriculture , 1993 .

[27]  Weifeng Li,et al.  Comparing Machine Learning Classifiers for Object-Based Land Cover Classification Using Very High Resolution Imagery , 2014, Remote. Sens..

[28]  Cho‐ying Huang,et al.  Woody plant proliferation in North American drylands: A synthesis of impacts on ecosystem carbon balance , 2011 .

[29]  Max Bramer,et al.  Principles of Data Mining , 2013, Undergraduate Topics in Computer Science.

[30]  R. Everett,et al.  Fire history in the ponderosa pine/Douglas-fir forests on the east slope of the Washington Cascades , 2000 .

[31]  Philip W. Mote,et al.  Trends in snow water equivalent in the Pacific Northwest and their climatic causes , 2003 .

[32]  M. Kubát.  An Introduction to Machine Learning , 2017, Springer International Publishing.

[33]  Thomas R. Karl,et al.  Observed Impact of Snow Cover on the Heat Balance and the Rise of Continental Spring Temperatures , 1994, Science.

[34]  S. Schneider,et al.  Fingerprints of global warming on wild animals and plants , 2003, Nature.

[35]  C. Gouveia.  Predicting the Impacts of Climate Change on Protected Areas: A Case Study of Land Snails in Madeira Island , 2015 .

[36]  S S I T C H,et al.  Evaluation of Ecosystem Dynamics, Plant Geography and Terrestrial Carbon Cycling in the Lpj Dynamic Global Vegetation Model , 2022 .

[37]  John R. Jensen,et al.  A change detection model based on neighborhood correlation image analysis and decision tree classification , 2005 .

[38]  C. Allen.  Interactions Across Spatial Scales among Forest Dieback, Fire, and Erosion in Northern New Mexico Landscapes , 2007, Ecosystems.

[39]  I. C. Prentice,et al.  A dynamic global vegetation model for studies of the coupled atmosphere‐biosphere system , 2005 .

[40]  Martha C. Anderson,et al.  Landsat-8: Science and Product Vision for Terrestrial Global Change Research , 2014 .

[41]  I. C. Prentice,et al.  Evaluation of the terrestrial carbon cycle, future plant geography and climate‐carbon cycle feedbacks using five Dynamic Global Vegetation Models (DGVMs) , 2008 .

[42]  Kathleen Neumann,et al.  Challenges in using land use and land cover data for global change studies , 2011 .

[43]  Thomas G. Dietterich.  What is machine learning? , 2020, Archives of Disease in Childhood.

[44]  Vladimir Vapnik,et al.  An overview of statistical learning theory , 1999, IEEE Trans. Neural Networks.

[45]  M. Janssen,et al.  Multi-Agent Systems for the Simulation of Land-Use and Land-Cover Change: A Review , 2003 .

[46]  N. Pettorelli,et al.  Using the satellite-derived NDVI to assess ecological responses to environmental change. , 2005, Trends in ecology & evolution.

[47]  Nadejda A. Soudzilovskaia,et al.  Functional traits predict relationship between plant abundance dynamic and long-term climate warming , 2013, Proceedings of the National Academy of Sciences.

[48]  F. Pan,et al.  Effects of harvest on carbon and nitrogen dynamics in a Pacific Northwest forest catchment , 2013 .

[49]  S. Running,et al.  What does Remote Sensing Do for Ecology , 1991 .

[50]  Edward J. Knight,et al.  Landsat-8 Operational Land Imager Design, Characterization and Performance , 2014, Remote. Sens..

[51]  Sotiris B. Kotsiantis,et al.  Supervised Machine Learning: A Review of Classification Techniques , 2007, Informatica.

[52]  Ying LU,et al.  Decision tree methods: applications for classification and prediction , 2015, Shanghai archives of psychiatry.

[53]  Pedro M. Domingos.  A few useful things to know about machine learning , 2012, Commun. ACM.

[54]  P. Marquet,et al.  A Significant Upward Shift in Plant Species Optimum Elevation During the 20th Century , 2008, Science.

[55]  Annemarie Schneider,et al.  Monitoring land cover change in urban and peri-urban areas using dense time stacks of Landsat satellite data and a data mining approach , 2012 .

[56]  Andrei P. Sokolov,et al.  Climate Dynamics (2006) DOI 10.1007/s00382-005-0092-6 , 2005 .

[57]  Jonathan Cheung-Wai Chan,et al.  Multiple Criteria for Evaluating Machine Learning Algorithms for Land Cover Classification from Satellite Data , 2000 .

[58]  Stephan Pauleit,et al.  Modeling the environmental impacts of urban land use and land cover change—a study in Merseyside, UK , 2005 .

[59]  Steven J. Phillips,et al.  Shifts in Arctic vegetation and associated feedbacks under climate change , 2013 .

[60]  C. Brodley,et al.  Decision tree classification of land cover from remotely sensed data , 1997 .

[61]  U. Gessner,et al.  Regional land cover mapping and change detection in Central Asia using MODIS time-series , 2012 .

[62]  J. W. Winchester,et al.  The Date Of Snow Disappearance On The Arctic Tundra As Determined From Satellite, Meteorological Station And Radiometric In-situ Observations , 1991, [Proceedings] IGARSS'91 Remote Sensing: Global Monitoring for Earth Management.

[63]  Trevor Hastie,et al.  Making better biogeographical predictions of species’ distributions , 2006 .

[64]  Thomas Giesecke,et al.  Projecting the future distribution of European potential natural vegetation zones with a generalized, tree species-based dynamic vegetation model , 2012 .

[65]  Feifei Pan,et al.  Catchment hydrological responses to forest harvest amount and spatial pattern , 2011 .

[66]  Virginia H. Dale,et al.  THE RELATIONSHIP BETWEEN LAND‐USE CHANGE AND CLIMATE CHANGE , 1997 .

[67]  อนิรุธ สืบสิงห์,et al.  Data Mining Practical Machine Learning Tools and Techniques , 2014 .

[68]  C. Müller,et al.  Uncertainties in climate responses to past land cover change: First results from the LUCID intercomparison study , 2009 .

[69]  Antoine Guisan,et al.  Predictive habitat distribution models in ecology , 2000 .

[70]  M. Goulden,et al.  Rapid shifts in plant distribution with recent climate change , 2008, Proceedings of the National Academy of Sciences.

[71]  Alex S. Kutt,et al.  Focus on poleward shifts in species' distribution underestimates the fingerprint of climate change , 2013 .

[72]  N. Vuichard,et al.  A validation of heat and carbon fluxes from high-resolution land surface and regional models , 2010 .

[73]  O. Hoegh‐Guldberg,et al.  Ecological responses to recent climate change , 2002, Nature.