Simon Dedman

From Jane Elith: "I noticed a bug in predicting to rasters with factor variables. The predictions are different to what happens via a data frame." More details to follow

Have users enter their lat & lon columns instead of bounds = c(xmin,xmax,ymin,ymax)? More in-keeping with standard practice for this package

In cumulative closures, sum the HRMSY% for each species, e.g. 4th one will be ~100%, how much is the 1st? 150%? Where to put that, in legend? Metrics: add %area...

L177 do I use k for anything? Is that a problem? Only 1 badcols allowed. e.g. L77 invert baddata meh.

PerSpeciesClosedAreaMaps: need to make the scale from 1:4 (0:4?) not auto. There are 1+4 items in the legend, correctly. L309 set breaks. gbm.map allows them to be parsed in "..."?...

Label perspeciesareamap legend items w/ species names not 1:4 numbers (L387 & 402) legend

have per-species cons maps underpin the Conservation Sort in p3, rather than all 4 species underpinned by the 4-sp gbm.cons map.

manually setting heatcolours in the gbm.auto(call) passes heatcolours to gbm.auto for gbm.map but it'll override the grey colours for the B&W maps, See gbm.auto L578

Diya: Error in FUN(X[[i]], ...) : only defined on a data frame with all numeric variables 3 env vars categorical. Changed to numeric. If data need to be numeric I...

C of V map: C of V map: 20% CV is a concerningly high figure. Ideally have a colour scheme which clearly illustrates when 20% is reached/passed (Cóilín). C of...