[R-sig-eco] clamping in R dismo package
Hi all, In niche modeling with maxent through the R package Dismo, is there a way to see which areas in a projected model were subjected to clamping? I know there's an option in maxent to write clamp grid, but using that argument in the predict function in Dismo doesn't seem to do anything... Thanks, -Pascal -- Pascal Title, MSc. PhD student, Rabosky Lab http://cteg.berkeley.edu/~rabosky/Home.html Dept of Ecology and Evolutionary Biology University of Michigan pti...@umich.edu [[alternative HTML version deleted]] ___ R-sig-ecology mailing list R-sig-ecology@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
[R-sig-eco] Permanova with nested data
Hi, Apologies if this has been asked before, I looked looked through the archives and couldn't find a solution. I have plant community data from 32 woodland plots. Plots are grouped into sites (8 sites in total). Half the sites (4) are on one soil type, and half on another. At each site there are 4 plots, each under a different management regime . I want to know if the plant communities under different management regimes and on different soil types differ. My data is as follows: veg - a data frame containg the community data (abundances of each species at each plot) environ-read.csv(file.choose(),header=TRUE) str(environ) 'data.frame': 32 obs. of 4 variables: $ code: Factor w/ 32 levels cf.1,cf.3,..: 1 9 17 25 2 10 18 26 3 11 ... $ site: Factor w/ 8 levels eight,five,..: 5 5 5 5 8 8 8 8 3 3 ... $ type: Factor w/ 4 levels clearfell,plantation,..: 1 2 3 4 1 2 3 4 1 2 ... $ nvc : Factor w/ 2 levels W10,W8: 2 2 2 2 2 2 2 2 1 1 ... # site is the name of each site # type refers to the management regime # nvc refers to the soil type (actually the NVC classification) Prior to PERMANOVA I used betadisper() to test for homogeneity of multivariate dispersion. No difference in dispersion was found either between plots on different site types, or plots on different NVC types. I carried out a PERMANOVA using adonis on this data as follows: adon.mod1.bray-adonis(veg~ type*nvc, data=environ,strata=environ$site, method = bray, permutations=999) I used strata=environ$site as management regime is nested within site. adon.mod1.bray Call: adonis(formula = veg ~ type * nvc, data = environ, permutations = 999, method = bray, strata = environ$site) Terms added sequentially (first to last) Df SumsOfSqs MeanSqs F.Model R2 Pr(F) type 32.2358 0.74528 3.5742 0.24662 0.001 *** nvc11.0905 1.09045 5.2296 0.12028 0.001 *** type:nvc 30.7353 0.24509 1.1754 0.08110 0.208 Residuals 245.0044 0.20852 0.55200 Total 319.0659 1.0 Is this acceptable. Is the strata=environ$site part correct? If data points are only being permutted within sites, and each site only occurs on one soil type, is there no permutation between NVC/soil types? I've also heard that adonis does not give the correct p values when data is nested. Is this correct, and a problem in the above example? If so is there a more suitable analysis? Would a db-RDA as follows be ok?: dbRDA-capscale(veg~type*nvc+Condition(site), data=environ, distance=bray, add=TRUE) anova.cca(dbRDA, stop=999) anova.cca(dbRDA, by=terms, step=999) anova.cca(dbRDA, by=terms, step=999) Permutation test for capscale under reduced model Terms added sequentially (first to last) Model: capscale(formula = veg ~ type + nvc + Condition(site) + type:nvc, data = environ, distance = bray, add = TRUE) DfVar F N.Perm Pr(F) type 3 2.9615 3.3215999 2e-16 *** type:nvc 3 1.1299 1.2673999 0.1141 Residual 18 5.3496 However, this doesn't show a result for nvc. Why? I've also been looking into the manyglm() function in mvabund, but I don't think this can accept random effects. Any help would be greatly appreciated, Thanks, Beth Beth Atkinson PhD student Community Ecology Group School of Biological Sciences University of Bristol Woodland Road Bristol BS8 1UG ___ R-sig-ecology mailing list R-sig-ecology@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
[R-sig-eco] Vegan metaMDS: unusual first run stress values with large data set
Hello, R-Community! This is the first time writing to this group and indeed the first time using a mailing list, so please bear with me if I’ve done something wrong. I have a large species x site matrix (89 x 4831) that I want to ordinate using metaMDS in the Vegan (2.0-5) package in R (2.15.2). If I run this data frame using the Jaccard index in two or more dimensions (k1), the first run (run=0) has a relatively low stress value and the other 20 runs are much higher and have very low deviation. However, k=1 seems to work fine. Furthermore, a stress/scree plot reveals a pyramid-like shape, where the k=1 lowest stress value is low, increases rapidly for k=2 then decreases slowly as k increases. Dimensions Stress 1 0.1382185 2 0.1939509 3 0.1695375 4 0.155221 5 0.1406408 6 0.1294149 I’ve tried this with a small iteration of this data and this issue arises at k2 rather than at k1 as it is here. Anyway, this is the input and output: library(vegan) library(MASS) PSU - read.table(PSU.txt, header = TRUE, sep = ) PSU.sp - PSU[, 22:110] PSU.NMDS - metaMDS(PSU.sp, k=4, zerodist = add, distance = jaccard) Square root transformation Wisconsin double standardization Zero dissimilarities changed into 0.0006657301 Run 0 stress 0.155221 Run 1 stress 0.2548103 Run 2 stress 0.255434 Run 3 stress 0.2551382 … (Up to run 20 where run 1 through run 20 have all very similar stress values.) Call: metaMDS(comm = PSU.sp, distance = jaccard, k = 4, zerodist = add) global Multidimensional Scaling using monoMDS Data: wisconsin(sqrt(PSU.sp)) Distance: jaccard Dimensions: 4 Stress: 0.155221 Stress type 1, weak ties No convergent solutions - best solution after 20 tries Scaling: centring, PC rotation, halfchange scaling Species: expanded scores based on ‘wisconsin(sqrt(PSU.sp))’ Now, again, with k=1 this does not happen – the solution looks like any other regular NMDS run. There are no blank values in the data as they are all numbers between 0 and 100 corresponding to % cover, and every row and column sum is greater than 0. There are many sites with the same species configurations, hence the zerodist, but omitting this makes no difference to the problem at hand. The NMDS works fine if I use a subset of the data, but I have not subsetted and tested all of it. Other metric (Euclidean) and nonmetric (Bray) dissimilarity indices result in the same effect. I’ve chosen k=4 here because of the (marginal) elbow in the stress plot, but the data itself actually looks pretty good at any k value. Even though the output is reasonable, I am concerned that hitting the best solution by a large amount on the first run means something is messing up, and this concern is amplified by the strange pyramid shaped stress plot. Because metaMDS uses random starts, I don't see how this output is possible. I've scoured the help files and archives of this list and I am really now at a loss to explain this. Thank you in advance for your time and consideration! Ewan ___ R-sig-ecology mailing list R-sig-ecology@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
[R-sig-eco] Applied Environmental Data Analyses: Water Chemistry and Biota
I left academia and basic ecological research decades ago and now work with environmental data collected by companies in compliance with regulatory permit requirements. I've bought and read (mostly) all the books I could find on ecological analyses using R (all but one of Alain Zuur's books, Legendre Legendre's 2nd English edition, Mike McCarthy's Bayesian methods book, and Ben Bolker's book) but cannot find any references to 'communities' in the indices. I'd greatly appreciate pointers to sources appropriate for environmental data (which is much sloppier than ecological research data). The last time I addressed community analyses was my post-doc research which I published in Freshwater Biology in 1984. Only within the past year have my clients needed to address issues using benthic macroinvertebrate assemblages (and fish) in streams. And, since I work by myself, I've no one with whom to share ideas and discuss approaches; perhaps there's a better forum than this mail list for this. The available benthic data has little taxonomic consistency below the family level. I want to use functional feeding groups rather than taxa as the basis of comparison because those better reflect conditions in each stream (collections of biota are made only once per year), and I want to examine correlations and cause-and-effect relations between biotic assemblages and water chemistry. There are only a few fish collections, tool, in the available data. All ideas are certainly welcome! TIA, Rich -- Richard B. Shepard, Ph.D. | Integrity - Credibility - Innovation Applied Ecosystem Services, Inc. | http://www.appl-ecosys.com Voice: 503-667-4517 Fax: 503-667-8863 ___ R-sig-ecology mailing list R-sig-ecology@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
Re: [R-sig-eco] Applied Environmental Data Analyses: Water Chemistry and Biota
Hi Rich, Theres a book coming out next May by Otto Wildi called Data Analysis in Vegetation Ecology 2nd Edition which includes a lot of R code. While its aimed at vegetation ecology, a lot of the principles in the book can be generalised to community analysis in whatever form. It covers ordination, cluster analysis, some basic GLM and numerous other topics. It takes the topics in the 1st Ed and adds R code to redo the analyses, and adds a few other methods not covered in the 1st Ed, the main difference will be the code though. Book: http://eu.wiley.com/WileyCDA/Section/id-302479.html?query=Otto+Wildi 1st Ed website: http://www.wsl.ch/info/mitarbeitende/wildi/data_analysis_book/index_EN HTH Alan -- Email: aghay...@gmail.com Mobile: +41794385586 Skype: aghaynes On 6 December 2012 04:59, alfredo tello alfredote...@gmail.com wrote: Hi Rich, Check out the R package Vegan tutorial by Jari Oksanen: http://cc.oulu.fi/~jarioksa/opetus/metodi/vegantutor.pdf. Vegan has several functions which I believe should allow you to address some of your questions. I have not worked with functional groups, but I'm guessing you could treat them as you would treat taxa for statistical purposes (?). I think Vegan's function 'adonis' might be of interest to you. It allows you to partition variation in a data/distance matrix according to specified factors (e.g., water chemistry). Hopet this helps! Best, A On Wed, Dec 5, 2012 at 10:26 PM, Rich Shepard rshep...@appl-ecosys.com wrote: I left academia and basic ecological research decades ago and now work with environmental data collected by companies in compliance with regulatory permit requirements. I've bought and read (mostly) all the books I could find on ecological analyses using R (all but one of Alain Zuur's books, Legendre Legendre's 2nd English edition, Mike McCarthy's Bayesian methods book, and Ben Bolker's book) but cannot find any references to 'communities' in the indices. I'd greatly appreciate pointers to sources appropriate for environmental data (which is much sloppier than ecological research data). The last time I addressed community analyses was my post-doc research which I published in Freshwater Biology in 1984. Only within the past year have my clients needed to address issues using benthic macroinvertebrate assemblages (and fish) in streams. And, since I work by myself, I've no one with whom to share ideas and discuss approaches; perhaps there's a better forum than this mail list for this. The available benthic data has little taxonomic consistency below the family level. I want to use functional feeding groups rather than taxa as the basis of comparison because those better reflect conditions in each stream (collections of biota are made only once per year), and I want to examine correlations and cause-and-effect relations between biotic assemblages and water chemistry. There are only a few fish collections, tool, in the available data. All ideas are certainly welcome! TIA, Rich -- Richard B. Shepard, Ph.D. | Integrity - Credibility - Innovation Applied Ecosystem Services, Inc. | http://www.appl-ecosys.com Voice: 503-667-4517 Fax: 503-667-8863 __**_ R-sig-ecology mailing list R-sig-ecology@r-project.org https://stat.ethz.ch/mailman/**listinfo/r-sig-ecology https://stat.ethz.ch/mailman/listinfo/r-sig-ecology -- Alfredo Tello (http://alfredotello.com) Sustainable Aquaculture Group Institute of Aquaculture University of Stirling Scotland, UK. [[alternative HTML version deleted]] ___ R-sig-ecology mailing list R-sig-ecology@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-ecology [[alternative HTML version deleted]] ___ R-sig-ecology mailing list R-sig-ecology@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-ecology