deletions | additions       diff --git a/materials_and_methods.tex b/materials_and_methods.tex  index 4bc6685..21d9035 100644  --- a/materials_and_methods.tex  +++ b/materials_and_methods.tex 
...
\paragraph{Study area and presence data}  The geographical area of this study (20ºN to 70ºN, 10ºW to 70ºE) comprises Europe and the Irano-Turanian region. This area does not cover the entire known range of Neanderthals, which extended at least to Okladnikov in Southern Siberia \cite{Krause_2007}, but there is only one single Neanderthal record attributed to MIS 5e beyond the Caspian Sea (Ust-Izhul, Siberia, \citet{Rolland2010}), and therefore our the  data available  would be too scarce to build a reliable model for this region. We foundin the literature  up to 65 archaeological sites in the literature  attributed to Neanderthals a Neanderthal  presence during the last interglacial period (see Appendix). 35 of these presences were attributed by their authors to MIS 5e, while the remaining ones were broadly attributed to MIS 5, 5  and thereforewere  excluded from our study. To reduce spatial clustering and pseudorreplication pseudo-replication  within the presence dataset we filtered the data applying a minimum distance of 100 km between nearby locations \cite{Guisan_2005}, which reduced the sample sizedown  to 26 presence records (see Fig. 1). \paragraph{Environmental variables} 
...
We processed the digital elevation model provided by www.worldclim.com (SRTM model aggregated to 1km resolution, \cite{Farr_2007}) and derived maps representing slope, topographic wetness index, and topographic diversity (see appendix for further details) to represent factors influencing Neanderthals distribution at the local scale. All the variables were aggregated to 5 km resolution using GRASS GIS (GRASS development team 2012). To represent the Eemian shoreline the sea level was set at 7 m.a.s.l \cite{Dutton_2012}.  We assessed multicollinearity among the 22 environmental predictors (19 bioclim variables and 3 topographic variables) using a two step approach. Firstly, we computed the correlation matrix among variables, and defined 0.5 (Pearson correlation index) as a maximum correlation criterion to reject correlated predictors. Secondly we applied a variance inflation factor analysis to remove predictors being linear combinations of other predictors (R package HH, \cite{heiberger2015hh}). The final set of predictors comprised: maximum temperature of warmest month (bio5), minimum temperature of coldest month (bio6), annual precipitation (bio12), precipitation of warmest quarter (bio18), slope and topographic diversity. \paragraph{Species distribution modeling} 
...
The previous analysis only provides a single number for each variable across the whole study area, and cannot evaluate whether a particular variable is important at a particular region or not. To provide a further insight into this question and to understand how habitat suitability is shaped by the different predictors at the local scale we defined \emph{local scale} as the average home range of Neanderthals. According to \citet{Daujeard201232}, and based on the transportation of raw lithic materials, the regional mobility range of Neanderthals during the Middle Palaeolithic was around 50 kilometers. In consequence we divided the study area into cells of 50 per 50 kilometers, and for each cell with more than 30 original cells (the resolution of the habitat suitability map and the predictors) we fitted one linear regression model per environmental predictor using habitat suitability as response variable. For any given predictor, we considered the R squared to be an indicator of its importance at the local scale, and the coefficient to be an indicator of its effect over habitat suitability. We interpreted near zero coefficients linked to low habitat suitability as regions with extreme values for the given predictor, while near zero coefficients linked to high habitat suitability values were interpreted as optimum habitat. Positive coefficients indicated a positive (but still sub-optimum) effect of the given predictor over habitat suitability, while negative coefficients indicated that the predictor values were beyond the optimum (e.g. too hot, too wet). All the results with p-values higher than 0.05 were recoded to no-data to reduce noise in the following analyses. We selected 44 European localities with different values of habitat suitability (See Table 1), and applied recursive partition trees (rpart library, \cite{rpartcitation}) using habitat suitability as response variable to group them in three different ways: 1) using the values of the environmental variables as predictors, to group the localities according to their environmental similarity; 2) using the local R-squared as predictors, to group the localities depending on the importance of the predictors; 3) using the local coefficients as predictors, to group together the localities with similar effect of the environmental variables.