How to test the significance of the relation between spatially autocorrelated data at the landscape scale: A case study using fire and forest maps

Abstract To better understand the relationship between wildfire and forest regeneration in the boreal forest, we quantify their degree of relationship by means of correlation. Given that wildfires in the boreal forest can cover large areas, such correlation needs also to be computed for large areas (i.e., 33,000 km2 in northern Québec). At this landscape scale, both variables (fire and forest) show strong and significant positive spatial autocorrelation. The presence of spatial autocorrelation in the two variables, however, can affect the statistical significance and the interpretation of their degree of correlation. In this paper, we compare different approaches that have been proposed to solve this problem: a parametric test that corrects for the presence of autocorrelation by adjusting the effective sample size (Dutilleul’s modified t test); a complete randomization test; a restricted randomization test based on a toroidal shift; the Mantel test that controls for the relative spatial locations among sampling; and the partial Mantel test that controls for the spatial distances among sampling sites. A positive correlation between the two variables was found significant by the parametric test and complete randomization test, but not significant when the restricted randomization test, Dutilleul’s modified t test, and the Mantel test were used. Conversely, a negative correlation was found by the partial Mantel test. Hence, to control for the presence of spatial autocorrelation, either a restricted randomization test or the Dutilleul method is recommended, while to control for the spatial relative position of the data, the Mantel and partial Mantel tests should be used. A firm understanding of these statistical tests and their respective assumptions regarding the spatial structure of the data is crucial to any valid ecological understanding and interpretation.

[1]  P. Clifford,et al.  Modifying the t test for assessing the correlation between two spatial processes , 1993 .

[2]  Jeremy S. Fried,et al.  Predicting the impacts of global warming on wildland fire , 1992 .

[3]  C. Elfring.  Yellowstone: fire storm over fire management , 1989 .

[4]  E. Johnson.  Fire recurrence in the subarctic and its implications for vegetation composition , 1979 .

[5]  D Hémon,et al.  Assessing the significance of the correlation between two spatial processes. , 1989, Biometrics.

[6]  M. Flannigan,et al.  Predicting the effects of climate change on fire frequency in the southeastern Canadian boreal forest , 1995 .

[7]  F. Rousset,et al.  ARE PARTIAL MANTEL TESTS ADEQUATE? , 2001, Evolution; international journal of organic evolution.

[8]  J. Keith Ord,et al.  Spatial Processes Models and Applications , 1981 .

[9]  Geoffrey M. Jacquez,et al.  Quantification of the Spatial Co-Occurrences of Ecological Boundaries , 1996 .

[10]  Y. Bergeron,et al.  Changes in the understory of Canadian southern boreal forest after fire , 1993 .

[11]  R. M. Cormack,et al.  Spatial Data Analysis by Example. Volume 1: Point Pattern and Quantitative Data , 1985 .

[12]  P. Legendre.  Spatial Autocorrelation: Trouble or New Paradigm? , 1993 .

[13]  M. Fortin,et al.  Spatial autocorrelation and statistical tests in ecology , 2002 .

[14]  Noel A. C. Cressie,et al.  Statistics for Spatial Data: Cressie/Statistics , 1993 .

[15]  Norman L. Christensen,et al.  Interpreting the Yellowstone Fires of 1988Ecosystem responses and management implications , 1989 .

[16]  B. Manly.  Randomization, Bootstrap and Monte Carlo Methods in Biology , 2018 .

[17]  Pierre Legendre,et al.  Comparison of permutation methods for the partial correlation and partial mantel tests , 2000 .

[18]  M. Fortin,et al.  The Subarctic Forest–Tundra: The Structure of a Biome in a Changing Climate , 2001 .

[19]  Robert Haining,et al.  Statistics for spatial data: by Noel Cressie, 1991, John Wiley & Sons, New York, 900 p., ISBN 0-471-84336-9, US $89.95 , 1993 .

[20]  A. K. BREGT,et al.  Determination of rasterizing error a case study with the soil map of The Netherlands , 1991, Int. J. Geogr. Inf. Sci..

[21]  Wim G. M. van der Knaap,et al.  The vector to raster conversion: (mis)use in geographical information systems , 1992, Int. J. Geogr. Inf. Sci..

[22]  W. Romme,et al.  Historical Perspective on the Yellowstone Fires of 1988A reconstruction of prehistoric fire history reveals that comparable fires occurred in the early 1700s , 1989 .

[23]  M. Fortin.  Effects of sampling unit resolution on the estimation of spatial autocorrelation , 1999 .

[24]  N. Mantel.  The detection of disease clustering and a generalized regression approach. , 1967, Cancer research.

[25]  R. Balling,et al.  Climate change in Yellowstone National Park: Is the drought-related risk of wildfires increasing? , 1992 .

[26]  M. Fortin,et al.  Spatial pattern and ecological analysis , 1989, Vegetatio.

[27]  Robert R. Sokal,et al.  An investigation of three-matrix permutation tests , 1992 .

[28]  Frederik P. Agterberg,et al.  Interactive spatial data analysis , 1996 .

[29]  Noel A Cressie,et al.  Uncertainty and Spatial Linear Models for Ecological Data , 2001 .

[30]  R. Sokal,et al.  Multiple regression and correlation extensions of the mantel test of matrix correspondence , 1986 .

[31]  W. Hargrove,et al.  Effects of fire on landscape heterogeneity in Yellowstone National Park, Wyoming , 1994 .

[32]  S. Payette,et al.  Late Holocene deforestation and tree regeneration in the forest–tundra of Québec , 1985, Nature.

[33]  Dominique Arseneault,et al.  A Postfire Shift From Lichen‐Spruce to Lichen‐Tundra Vegetation at Tree Line , 1992 .

[34]  Louis Legendre,et al.  The Importance of Being Digital , 1963 .

[35]  S. Payette,et al.  RECENT FIRE HISTORY OF THE NORTHERN QUEBEC BIOMES , 1989 .