Methods to account for spatial autocorrelation in the analysis of species distributional data: a review

Ecography - Tập 30 Số 5 - Trang 609-628 - 2007
Carsten F. Dormann, Jana McPherson, Miguel B. Araújo, Roger Bivand, Janine Bolliger, Gudrun Carl, R. Davies, Alexandre H. Hirzel, Walter Jetz, W. Daniel Kissling, Ingolf Kühn, Ralf Ohlemüller, Pedro R. Peres‐Neto, Björn Reineking, Boris Schröder, Frank M. Schurr, Robert J. Wilson

Tóm tắt

Species distributional or trait data based on range map (extent‐of‐occurrence) or atlas survey data often display spatial autocorrelation, i.e. locations close to each other exhibit more similar values than those further apart. If this pattern remains present in the residuals of a statistical model based on such data, one of the key assumptions of standard statistical analyses, that residuals are independent and identically distributed (i.i.d), is violated. The violation of the assumption of i.i.d. residuals may bias parameter estimates and can increase type I error rates (falsely rejecting the null hypothesis of no effect). While this is increasingly recognised by researchers analysing species distribution data, there is, to our knowledge, no comprehensive overview of the many available spatial statistical methods to take spatial autocorrelation into account in tests of statistical significance. Here, we describe six different statistical approaches to infer correlates of species’ distributions, for both presence/absence (binary response) and species abundance data (poisson or normally distributed response), while accounting for spatial autocorrelation in model residuals: autocovariate regression; spatial eigenvector mapping; generalised least squares; (conditional and simultaneous) autoregressive models and generalised estimating equations. A comprehensive comparison of the relative merits of these methods is beyond the scope of this paper. To demonstrate each method's implementation, however, we undertook preliminary tests based on simulated data. These preliminary tests verified that most of the spatial modeling techniques we examined showed good type I error control and precise parameter estimates, at least when confronted with simplistic simulated data containing spatial autocorrelation in the errors. However, we found that for presence/absence data the results and conclusions were very variable between the different methods. This is likely due to the low information content of binary maps. Also, in contrast with previous studies, we found that autocovariate methods consistently underestimated the effects of environmental controls of species distributions. Given their widespread use, in particular for the modelling of species presence/absence data (e.g. climate envelope models), we argue that this warrants further study and caution in their use. To aid other ecologists in making use of the methods described, code to implement them in freely available software is provided in an electronic appendix.

Từ khóa


Tài liệu tham khảo

10.2307/2532950

Anon. 2005. R: a language and environment for statistical computing. – R Foundation for Statistical Computing.

10.1007/978-94-015-7799-1

10.1111/j.1574-0862.2002.tb00120.x

10.1016/S0006-3207(00)00074-4

10.1126/science.1131758

10.2307/2404755

10.1002/(SICI)1099-095X(199803/04)9:2<175::AID-ENV294>3.0.CO;2-2

Augustin N. H., 2005, Analyzing the spread of beech canker, For. Sci., 51, 438

10.1016/0377-0427(96)00018-0

10.2307/3236222

Besag J., 1974, Spatial interaction and the statistical analysis of lattice systems, J. Roy. Stat. Soc. B, 36, 192

10.1007/BF00116466

Bivand R. 2005. spdep: spatial dependence: weighting schemes statistics and models. – R package version 0.3–17.

10.1023/A:1009601932481

10.1016/S0304-3800(01)00501-4

10.2307/2290687

10.1098/rsta.2003.1263

10.1289/ehp.6052

10.1111/j.1538-4632.1996.tb00936.x

Carey V. J. 2002. gee: generalized estimation equation solver. Ported to R by Thomas Lumley (ver. 3.13 4.4) and Brian Ripley. – <www.r‐project.org>.

10.1016/j.ecolmodel.2007.04.024

Carl G. and Kühn I. 2007b. Analyzing spatial ecological data using linear regression and wavelet analysis. – Stochast. Environ. Res. Risk Assess. in press.

10.2307/2685208

Cliff A. D., 1981, Spatial processes: models and applications

10.2307/2532039

10.1002/9781119115151

10.1111/j.1472-4642.2004.00054.x

10.1098/rspb.2006.3551

Diggle P. J., 1995, Analysis of longitudinal data

10.1111/j.1466-822X.2005.00147.x

10.1111/j.1558-5646.1998.tb02006.x

10.1046/j.1466-822X.2003.00322.x

Dobson A. J., 2002, An introduction to generalized linear models

10.1016/j.ecolmodel.2007.05.002

10.1111/j.1466-8238.2006.00279.x

10.1016/j.baae.2006.11.001

10.1016/j.ecolmodel.2006.02.015

10.2307/2532625

10.1111/j.2006.0906-7590.04596.x

10.1023/A:1021302930424

10.1111/j.1466-822X.2004.00097.x

10.1017/CBO9780511542039

Fotheringham A. S., 2002, Geographically weighted regression: the analysis of spatially varying relationships

10.1111/j.1365-2699.2006.01509.x

10.1093/biostatistics/4.1.11

10.1111/j.1467-9876.2005.00466.x

10.1016/S0024-3795(00)00031-8

10.1007/PL00011451

10.1890/0012-9658(2006)87[2603:SMIETF]2.0.CO;2

10.1016/S0304-3800(00)00354-9

10.1111/j.1461-0248.2005.00792.x

10.2307/1400400

10.1017/CBO9780511754944

Hastie T. J., 1990, Generalized additive models

10.1111/j.0906-7590.2007.05117.x

10.1198/1085711031508

10.2307/1400634

10.1023/A:1026001008598

10.2307/3109759

10.2307/1942661

Isaaks E. H., 1989, An introduction to applied geostatistics

10.1126/science.1072779

10.1111/j.1466-822X.2004.00129.x

10.1111/j.1365-2699.2005.01379.x

Kaluzny S. P., 1998, S‐plus spatial stats user's manual for Windows and Unix

10.1034/j.1600-0587.2002.250509.x

Kissling W. D. and Carl G. 2007. Spatial autocorrelation and the selection of simultaneous autoregressive models. – Global Ecol. Biogeogr. in press.

Klute D. S., 2002, Predicting species occurrences: issues of accuracy and scale, 335

10.1890/1051-0761(2003)13[1069:DPMTPA]2.0.CO;2

10.1007/s10980-006-9058-2

10.1111/j.1472-4642.2006.00293.x

10.1111/j.1469-8137.2006.01811.x

10.1890/04-0609

10.2307/1939924

10.1007/BF00048036

Legendre P., 1998, Numerical ecology

10.1034/j.1600-0587.2002.250508.x

10.1111/j.1600-0587.2000.tb00265.x

10.1093/biomet/73.1.13

10.1890/0012-9615(2002)072[0445:SAAAMI]2.0.CO;2

10.1034/j.1600-0587.2002.250505.x

10.1890/0012-9658(2006)87[2626:MWATFO]2.0.CO;2

Littell R. C., 1996, SAS system for mixed lodels

10.1111/j.1600-0587.2001.tb00494.x

10.1007/978-1-4899-3242-6

McPherson J. M., 2007, Effects of species’ ecology on the accuracy of distribution models, Ecography, 30, 135

10.1016/j.ecolmodel.2006.12.012

10.2193/0022-541X(2005)069[0933:MPOBFR]2.0.CO;2

Myers R. H., 2002, Generalized linear models

10.1038/nature03850

10.1046/j.1365-2664.2001.00604.x

Osborne P. E. et al. 2007. Non‐stationarity and local approaches to modelling the distributions of wildlife. – Div. Distribut. in press.

10.1046/j.1365-2664.1999.00436.x

10.1046/j.1466-822X.2003.00042.x

10.1034/j.1600-0587.2002.250507.x

10.1007/978-1-4419-0318-1

10.1111/j.1466-822X.2006.00237.x

10.1016/j.ecolmodel.2003.09.039

10.1111/j.1365-2699.2004.01076.x

10.1111/j.1365-2664.2006.01162.x

10.2307/2997753

10.1111/j.1095-8312.1978.tb00013.x

10.1111/j.1095-8312.1978.tb00014.x

10.1016/j.ecolmodel.2005.09.007

10.1016/S0378-1127(00)00490-4

Teterukovskiy A., 2003, Effective field sampling for predicting the spatial distribution of reindeer (Rangifer tarandus) with help of the Gibbs sampler, Ambio, 32, 568, 10.1579/0044-7447-32.8.568

10.1890/03-5247

10.1068/a310165

10.2307/143141

10.1111/j.0906-7590.2004.03732.x

10.1007/978-0-387-21706-2

10.2307/3236071

10.1016/S0378-3758(03)00111-3

10.1002/0471662682

10.1201/9781420010404

10.1126/science.1113399

Wu H. L., 1997, Modelling the distribution of plant species using the autologistic regression model, Environ. Ecol. Stat., 4, 49, 10.1023/A:1018505924603

10.1644/1545-1542(2003)084<1356:HPOFAM>2.0.CO;2

Yan J., 2002, geepack: yet another package for generalized estimating equations, R News, 2, 12

Yan J. 2004. geepack: generalized estimating equation package. – R package version 0.2–10.

Thông tin
Thông tin xuất bản

Ecography

Tập 30 Số 5

609-628

Thông tin tác giả