iForest - Biogeosciences and Forestry


A geographically weighted deep neural network model for research on the spatial distribution of the down dead wood volume in Liangshui National Nature Reserve (China)

Yuman Sun (1), Ziqi Ao (1), Weiwei Jia (1)   , Ying Chen (1), Kai Xu (2)

iForest - Biogeosciences and Forestry, Volume 14, Issue 4, Pages 353-361 (2021)
doi: https://doi.org/10.3832/ifor3705-014
Published: Jul 27, 2021 - Copyright © 2021 SISEF

Research Articles

In natural forest ecosystems, there is often abundant down dead wood (DDW) due to wind disasters, which greatly changes the size and structure of forests. Accurately determining the DDW volume (DDWV) is crucial for sustaining forest management, predicting the dynamic changes in forest resources and assessing the risks of natural disasters or disturbances. However, existing models cannot accurately express the significant spatial nonstationarity or complexity in their spatial relationships. To this end, we established a geographically weighted deep neural network (GWDNN) model that constructs a spatially weighted neural network (SWNN) through geographic location data and builds a neural network through stand factors and remote sensing factors to improve the interpretability of the spatial model of DDWV. To verify the effectiveness of this method, using 2019 data from Liangshui National Nature Reserve, we compared model fit, predictive ability and residual spatial autocorrelation among the GWDNN model and four other spatial models: an ordinary least squares (OLS) model, a linear mixed model (LMM), a geographically weighted regression (GWR) model and a deep neural network (DNN) model. The experimental results show that the GWDNN model is far superior to the other four models according to various indicators; the coefficient of determination R2, root mean square error (RMSE), mean absolute error (MAE), Moran’s I and Z-statistic values of the GWDNN model were 0.95, 1.05, 0.77, -0.01 and -0.06, respectively. In addition, compared with the other models, the GWDNN model can more accurately depict local spatial variations and details of the DDWV in Liangshui National Nature Reserve.


Down Dead Wood Volume (DDWV), Ordinary Least Squares (OLS) Model, Linear Mixed Model (LMM), Geographically Weighted Regression (GWR) Model, Deep Neural Network (DNN) Model, Geographically Weighted Deep Neural Network (GWDNN) Model

Authors’ address

Yuman Sun
Ziqi Ao
Weiwei Jia 0000-0001-7318-8997
Ying Chen
School of Forestry, Northeast Forestry University, Harbin 150040 (China)
Kai Xu
CosmosVision Robot of Anhui Zhong’an Chuanggu Technology Park, Hefei 230000 (China)

Corresponding author

Weiwei Jia


Sun Y, Ao Z, Jia W, Chen Y, Xu K (2021). A geographically weighted deep neural network model for research on the spatial distribution of the down dead wood volume in Liangshui National Nature Reserve (China). iForest 14: 353-361. - doi: 10.3832/ifor3705-014

Academic Editor

Maurizio Marchi

Paper history

Received: Nov 27, 2020
Accepted: Jun 03, 2021

First online: Jul 27, 2021
Publication Date: Aug 31, 2021
Publication Time: 1.80 months

Breakdown by View Type

(Waiting for server response...)

Article Usage

Total Article Views: 26195
(from publication date up to now)

Breakdown by View Type
HTML Page Views: 22550
Abstract Page Views: 1405
PDF Downloads: 1949
Citation/Reference Downloads: 4
XML Downloads: 287

Web Metrics
Days since publication: 1085
Overall contacts: 26195
Avg. contacts per week: 169.00

Article Citations

Article citations are based on data periodically collected from the Clarivate Web of Science web site
(last update: Feb 2023)

Total number of cites (since 2021): 4
Average cites per year: 1.33


Publication Metrics

by Dimensions ©

Articles citing this article

List of the papers citing this article based on CrossRef Cited-by.

Adamec Z, Drápela K (2017)
Comparison of parametric and nonparametric methods for modeling height-diameter relationships. iForest 10: 1-8.
CrossRef | Gscholar
Anselin L (2001)
Spatial effects in econometric practice in environmental and resource economics. American Journal of Agricultural Economics 83: 705-710.
CrossRef | Gscholar
Aquino DN, Neto OCR, Moreira MA, Teixeira AS, De Andrade EM (2018)
Use of remote sensing to identify areas at risk of degradation in the semi-arid region. Revista Ciencia Agronomica 49: 420-429.
CrossRef | Gscholar
Bassett M, Chia EK, Leonard SWJ, Nimmo DG, Holland GJ, Ritchie EG, Clarke MF, Bennett AF (2015)
The effects of topographic variation and the fire regime on coarse woody debris: Insights from a large wildfire. Forest Ecology and Management 340: 126-134.
CrossRef | Gscholar
Bates D (2005)
Fitting linear mixed models in R. R News 5: 27-30.
Online | Gscholar
Behrens T, Schmidt K, Viscarra Rossel RA, Gries P, Scholten T, MacMillan RA (2018)
Spatial modelling with Euclidean distance fields and machine learning. European Journal of Soil Science 69: 757-770.
CrossRef | Gscholar
Boonpook W, Tan Y, Xu B (2021)
Deep learning-based multi-feature semantic segmentation in building extraction from images of UAV photogrammetry. International Journal of Remote Sensing 42: 1-19.
CrossRef | Gscholar
Brunsdon C, Fotheringham AS, Charlton ME (1996)
Geographically weighted regression: a method for exploring spatial nonstationarity. Geographical Analysis 28: 281-298.
CrossRef | Gscholar
Chen L, Ren C, Zhang B, Wang Z, Xi Y (2018)
Estimation of forest above-ground biomass by geographically weighted regression and machine learning with Sentinel imagery. Forests 9 (10): 582.
CrossRef | Gscholar
Choi S, Lee S, Kang Y, Choi DY, Choi J (2020)
Use of unmanned aerial vehicle imagery and deep learning UNet to classification upland crop in small scale agricultural land. Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography 38: 671-679.
Online | Gscholar
Chojnacky DC, Heath LS (2002)
Estimating down deadwood from FIA forest inventory variables in Maine. Environmental Pollution 116 (7): S25-S30.
CrossRef | Gscholar
Costa S, Ibanez L (2005)
Can wood storage be profitable? French experience after the windstorms in 1999. Journal of Forest Economics 11: 161-176.
CrossRef | Gscholar
Deb D, Singh JP, Deb S, Datta D, Ghosh A, Chaurasia RS (2017)
An alternative approach for estimating above ground biomass using Resourcesat-2 satellite data and artificial neural network in Bundelkhand region of India. Environmental Monitoring and Assessment 189 (11): 631.
CrossRef | Gscholar
Du M, Noguchi N (2016)
Multi-temporal monitoring of wheat growth through correlation analysis of satellite images, unmanned aerial vehicle images with ground variable. IFAC-PapersOnLine 49: 5-9.
CrossRef | Gscholar
Du Z, Wang Z, Wu S, Zhang F, Liu R (2020)
Geographically neural network weighted regression for the accurate estimation of spatial non-stationarity. International Journal of Geographical Information Science 34: 1353-1377.
CrossRef | Gscholar
Fassnacht FE, Hartig F, Latifi H, Berger C, Hernandez J, Corvalan P, Koch B (2014)
Importance of sample size, data type and prediction method for remote sensing-based estimations of aboveground forest biomass. Remote Sensing of Environment 154: 102-114.
CrossRef | Gscholar
Fotheringham AS, Charlton ME, Brunsdon C (1998)
Geographically weighted regression: a natural evolution of the expansion method for spatial data analysis. Environment and Planning A 30: 1905-1927.
CrossRef | Gscholar
Green JL, Hastings A, Arzberger P, Ayala FJ, Cottingham KL, Cuddington K, Davis F, Dunne JA, Fortin M-J, Gerber L, Neubert M (2005)
Complexity in ecology and conservation: mathematical, statistical, and computational challenges. BioScience 55: 501-510.
CrossRef | Gscholar
Hagan JM, Grove SL (1999)
Coarse woody debris: humans and nature competing for trees. Journal of Forestry 97: 6-11.
Online | Gscholar
Harmon ME, Franklin JF, Swanson FJ, Sollins P, Gregory SV, Lattin JD, Anderson NH, Cline SP, Aumen NG, Sedell JR, Lienkaemper GW, Cromack Jr K, Cummins KW (1986)
Ecology of coarse woody debris in temperate ecosystems. Advances in Ecological Research 15: 133-302.
CrossRef | Gscholar
Irfan M, Koj A, Thomas H, Sedighi M (2016)
Geographical General Regression Neural Network (GGRNN) tool for geographically weighted regression analysis. In: Proceedings of the 8th International Conference on “Advanced Geographic Information Systems, Applications, and Services”. Venice (Italy), 24-28 April 2016. GEOProcessing 2016, pp. 154-159.
Online | Gscholar
Jenkins MA, Webster CR, Parker GR, Spetich MA (2004)
Coarse woody debris in managed central hardwood forests of Indiana, USA. Forest Science 50: 781-792.
Online | Gscholar
Ketkar N (2017)
Introduction to Keras. In: “Deep Learning with Python: A Hands-on Introduction” (N Ketkar ed). Apress, Berkeley, CA, USA, pp. 15-30.
CrossRef | Gscholar
Kinnucan HW (2016)
Timber price dynamics after a natural disaster: hurricane Hugo revisited. Journal of Forest Economics 25: 115-129.
CrossRef | Gscholar
Klobučar D (2010)
Using geostatistics in forest management. Sumarski List 134: 249-259.
Lavoie S, Ruel J-C, Bergeron Y, Harvey BD (2012)
Windthrow after group and dispersed tree retention in eastern Canada. Forest Ecology and Management 269: 158-167.
CrossRef | Gscholar
Lecun Y, Bengio Y, Hinton G (2015)
Deep learning. Nature 521: 436-444.
CrossRef | Gscholar
Li H, Calder CA, Cressie N (2007)
Beyond Moran’s I: testing for spatial dependence based on the spatial autoregressive model. Geographical Analysis 39: 357-375.
CrossRef | Gscholar
Li J, Jin M, Li H (2019)
Exploring spatial influence of remotely sensed PM2.5 concentration using a developed deep convolutional neural network model. International Journal of Environmental Research and Public Health 16: 454.
CrossRef | Gscholar
Liu C, Zhang L, Li F, Jin X (2014)
Spatial modeling of the carbon stock of forest trees in Heilongjiang Province, China. Journal of Forestry Research 25: 269-280.
CrossRef | Gscholar
Lu J, Zhang L (2011)
Modeling and prediction of tree height-diameter relationships using spatial autoregressive models. Forest Science 57: 252-264.
CrossRef | Gscholar
Mennis J (2006)
Mapping the results of geographically weighted regression. Cartographic Journal 43: 171-179.
CrossRef | Gscholar
Miles J (2005)
Tolerance and variance inflation factor. Encyclopedia of Statistics in Behavioral Science, North Yorkshire, UK, pp. 1-2.
Mohammadinia A, Saeidian B, Pradhan B, Ghaemi Z (2019)
Prediction mapping of human leptospirosis using ANN, GWR, SVM and GLM approaches. BMC Infectious Diseases 19 (1): 736.
CrossRef | Gscholar
Pang Y, Li Z, Ju H, Lu H, Jia W, Si L, Guo Y, Liu Q, Li S, Liu L, Xie B, Tan B, Dian Y (2016)
LiCHy: the CAF’s LiDAR, CCD and hyperspectral integrated airborne observation system. Remote Sensing 8 (5): 398.
CrossRef | Gscholar
Pfeil EK, Casacchia N, Kerns GJ, Diggins TP (2007)
Distribution, composition, and orientation of down deadwood in riparian old-growth woodlands of Zoar Valley Canyon, western New York State, USA. Forest Ecology and Management 239: 159-168.
CrossRef | Gscholar
Polo JA, Hallgren SW, Leslie DM (2013)
Effect of long-term understory prescribed burning on standing and down dead woody material in dry upland oak forests. Forest Ecology and Management 291: 128-135.
CrossRef | Gscholar
Qi YJ, Zhang YC, Wang K, He SQ, Tan W (2020)
Application of spatial regression models for forest biomass estimation in Guizhou Province, Southwest China. Applied Ecology and Environmental Research 18: 7215-7232.
CrossRef | Gscholar
Quirós-Segovia M, Condés-Ruiz S, Drápela K (2016)
Comparison of height-diameter models based on geographically weighted regressions and linear mixed modelling applied to large scale forest inventory data. Forest Systems 25: 76-86.
Online | Gscholar
Schmidhuber J (2015)
Deep learning in neural networks: an overview. Neural Networks 61: 85-117.
CrossRef | Gscholar
Spetich MA, Shifley SR, Parker GR (1999)
Regional distribution and dynamics of coarse woody debris in midwestern old-growth forests. Forest Science 45: 302-313.
Online | Gscholar
Subedi N, Zhang L, Zhen Z (2018)
Bayesian geographically weighted regression and its application for local modeling of relationships between tree variables. iForest 11: 542-552.
CrossRef | Gscholar
Sun J, Xu F, Cervone G, Gervais M, Wauthier C, Salvador M (2021)
Automatic atmospheric correction for shortwave hyperspectral remote sensing data using a time-dependent deep neural network. ISPRS Journal of Photogrammetry and Remote Sensing 174: 117-131.
CrossRef | Gscholar
Tiefelsdorf M (2002)
The saddlepoint approximation of Moran’s I’s and local Moran’s I’s reference distributions and their numerical evaluation. Geographical Analysis 34: 187-206.
CrossRef | Gscholar
Tiefelsdorf M, Boots B (1995)
The exact distribution of Moran’s I. Environment and Planning A 27: 985-999.
CrossRef | Gscholar
Wei Q, Zhang L, Duan W, Zhen Z (2019)
Global and geographically and temporally weighted regression models for modeling PM2.5 in Heilongjiang, China from 2015 to 2018. International Journal of Environmental Research and Public Health 16: 5107.
CrossRef | Gscholar
Wu S, Du Z, Wang Y, Lin T, Zhang F, Liu R (2020)
Modeling spatially anisotropic nonstationary processes in coastal environments based on a directional geographically neural network weighted regression. Science of the Total Environment 709: 136097.
CrossRef | Gscholar
Wu S, Wang Z, Du Z, Huang B, Zhang F, Liu R (2021)
Geographically and temporally neural network weighted regression for modeling spatiotemporal non-stationary relationships. International Journal of Geographical Information Science 35: 582-608.
CrossRef | Gscholar
Zhang L, Gove JH (2005)
Spatial assessment of model errors from four regression techniques. Forest Science 51: 334-346.
Online | Gscholar
Zhen Z, Li F, Liu Z, Liu C, Zhao Y, Ma Z, Zhang L (2013)
Geographically local modeling of occurrence, count, and volume of downwood in Northeast China. Applied Geography 37: 114-126.
CrossRef | Gscholar

This website uses cookies to ensure you get the best experience on our website. More info