Exploring spatial non-stationarity of fisheries survey data using geographically weighted regression (GWR): an example from the Northwest Atlantic

Windle, M. J. S., Rose, G. A., Devillers, R., and Fortin, M-J. 2010. Exploring spatial non-stationarity of fisheries survey data using geographically weighted regression (GWR): an example from the Northwest Atlantic. - ICES Journal of Marine Science, 67: 145-154.Analyses of fisheries data have traditionally been performed under the implicit assumption that ecological relationships do not vary within management areas (i.e. assuming spatially stationary processes). We question this assumption using a local modelling technique, geographically weighted regression (GWR), not previously used in fisheries analyses. Outputs of GWR are compared with those of global logistic regression and generalized additive models (GAMs) in predicting the distribution of northern cod off Newfoundland, Canada, based on environmental (temperature and distance from shore) and biological factors (snow crab and northern shrimp) from 2001. Results from the GWR models explained significantly more variability than the global logistic and GAM regressions, as shown by goodness-of-fit tests and a reduction in the spatial autocorrelation of model residuals. GWR results revealed spatial regions in the relationships between cod and explanatory variables and that the significance and direction of these relationships varied locally. A k-means cluster analysis based on GWR t-values was used to delineate distinct zones of species-environment relationships. The advantages and limitations of GWR are discussed in terms of potential application to fisheries ecology.

[1]  A. Paul.  Crabs in cold water regions : biology, management, and economics , 2002 .

[2]  Calvin A. Farris,et al.  Incorporating spatial non-stationarity of regression coefficients into predictive vegetation models , 2007, Landscape Ecology.

[3]  R. Tibshirani,et al.  Generalized additive models for medical research , 1986, Statistical methods in medical research.

[4]  K. Zou,et al.  Receiver-Operating Characteristic Analysis for Evaluating Diagnostic Tests and Predictive Models , 2007, Circulation.

[5]  B. Worm,et al.  META-ANALYSIS OF COD-SHRIMP INTERACTIONS REVEALS TOP-DOWN CONTROL IN OCEANIC FOOD WEBS , 2003 .

[6]  W. C. Leggett,et al.  The importance of scale to predator-prey spatial correlations: an example of Atlantic fishes , 1990 .

[7]  Jeffrey H. Gove,et al.  Spatial Assessment of Model Errors from Four Regression Techniques , 2005 .

[8]  D. Danielssen,et al.  New perspectives on fish movement: kernel and GAM smoothers applied to a century of tagging data on coastal Atlantic cod , 2008 .

[9]  Malcolm Haddon,et al.  Modelling and quantitative methods in fisheries , 2001 .

[10]  R. Rideout,et al.  Influence of diet on growth, condition and reproductive capacity in Newfoundland and Labrador cod (Gadus morhua) : Insights from stable carbon isotopes (δ13C) , 2007 .

[11]  S. Miao,et al.  Real world ecology : large-scale and long-term case studies and methods , 2009 .

[12]  Giles M. Foody,et al.  Spatial nonstationarity and scale-dependency in the relationship between species richness and environmental determinants for the sub-Saharan endemic avifauna , 2004 .

[13]  M. Pennington,et al.  How does somatic growth rate affect otolith size in fishes , 1997 .

[14]  J. Hutchings.  Spatial and temporal variation in the density of northern cod and a review of hypotheses for the stock's collapse , 1996 .

[15]  P. McDaniel,et al.  A Geographically Weighted Regression Analysis of Douglas-Fir Site Index in North Central Idaho , 2008 .

[16]  G. Rose,et al.  Cod spawning on a migration highway in the north-west Atlantic , 1993, Nature.

[17]  J. Mennis.  Mapping the Results of Geographically Weighted Regression , 2006 .

[18]  Carl J. Walters,et al.  Nonstationarity of Production Relationships in Exploited Populations , 1987 .

[19]  Kung-Sik Chan,et al.  Spatial fisheries ecology: Recent progress and future prospects , 2008 .

[20]  S. Fotheringham,et al.  Geographically Weighted Regression , 1998 .

[21]  B. deYoung,et al.  Comparison of hydrography and circulation on the Newfoundland Shelf during Pages 19901993 with the long-term mean , 1997 .

[22]  D. Kulka,et al.  Was the Increase in Shrimp Biomass on the Northeast Newfoundland Shelf a Consequence of a Release in Predation Pressure from Cod , 2000 .

[23]  E. Dawe,et al.  Distribution and demography of snow crab (Chionoecetes opilio) males on the Newfoundland and Labrador Shelf , 2002 .

[24]  M. Wimberly,et al.  Spatial heterogeneity of climate and land-cover constraints on distributions of tick-borne pathogens , 2008 .

[25]  N. Stenseth,et al.  The Responses of Fish Populations to Ocean Climate Fluctuations , 2005 .

[26]  C. Bishop,et al.  Distribution changes and abundance of northern cod (Gadus morhua), 1981-1993. , 1997 .

[27]  David Wheeler,et al.  Multicollinearity and correlation among local regression coefficients in geographically weighted regression , 2005, J. Geogr. Syst..

[28]  M. Austin.  Species distribution models and ecological theory: A critical assessment and some possible new approaches , 2007 .

[29]  L. Zhang,et al.  Local Modeling of Tree Growth by Geographically Weighted Regression , 2004 .

[30]  G. Rose,et al.  Capelin are good for cod: can the northern stock rebuild without them? , 2002 .

[31]  G. Rose.  On distributional responses of North Atlantic fish to climate change , 2005 .

[32]  J. Rice.  3 Changes to the large marine ecosystem of the Newfoundland-Labrador shelf , 2002 .

[33]  K. Sherman,et al.  Large marine ecosystems of the North Atlantic , 2002 .

[34]  S. Suárez‐Seoane,et al.  Non‐stationarity and local approaches to modelling the distributions of wildlife , 2007 .

[35]  G. Rose.  Monitoring coastal northern cod: towards an optimal survey of Smith Sound, Newfoundland , 2003 .

[36]  R. G. Davies,et al.  Methods to account for spatial autocorrelation in the analysis of species distributional data : a review , 2007 .

[37]  Alaska Sea Grant.  Resiliency of Gadid Stocks to Fishing and Climate Change , 2006 .

[38]  S. Wood.  Generalized Additive Models: An Introduction with R , 2006 .

[39]  N. Stenseth,et al.  Marine Ecosystems and Climate Variation , 2005 .

[40]  M. Fortin,et al.  Avian Spatial Responses to Forest Spatial Heterogeneity at the Landscape Level: Conceptual and Statistical Challenges , 2009 .

[41]  G. Lilly.  The Decline, Recovery, and Collapse of Atlantic Cod (Gadus morhua) off Labrador and Eastern Newfoundland , 2008 .

[42]  Edward J. Laurent,et al.  Local spatial modeling of white-tailed deer distribution , 2006 .

[43]  M. Fortin,et al.  Spatial Analysis: A Guide for Ecologists 1st edition , 2005 .