ࡱ> VXUy /bjbj .L{{'FF<<*%t,(TTTT/.]q $$$$$$$$&P)$//$TT$ TT$ $ T nX2 $$0*% * * J * } O}}}$$}}}*% *}}}}}}}}}F f: BIO 377 Lab Exercise: Vegetation Data and Diversity Goals: Understand the patterns of diversity and species associations across the environmental gradient of the Manu Tree community data set. Specifically: (1) Understand measures of diversity and community similarity. (2) Calculate similarity indices for the Manu Tree plots. (3) Understand ordination based on similarity of occurrence and abundance. Also, you should understand how the number of species sampled changes with sample size. To do this we will look at: (1) Species-area and species-individual curves (2) Rarefaction Introduction: How similar are two communities? The question seems simple, but there are lots of ways that communities can vary, e.g. (1) Species richness (2) Species diversity (3) Compositional similarity Species richness (sorting, consolidating) Q: How many species are in the sample? How do we compare among samples of different sizes? Dominance diversity curves (charts) Evenness (formulas) Diversity indicies (formulas) Fitting: Fisher's alpha (solver) Species-individual curves (R) Data Sets: BCI (in package vegan in R), manu.txt (on course site under lab data) and hrtrees.txt (on course site). Download the text files to your R working directory (usually C:\Program Files\R\rw2001) WEB RESOURCE:  HYPERLINK "http://ordination.okstate.edu/" http://ordination.okstate.edu/ This site has information on all of the ordination techniques and distance measures you could ever want to know. An excellent resource by Mike Palmer at OSU. Today well be using the R package vegan for community analyses. Open R and load the package vegan from the packages menu. Also, load MASS. To use the package the data have to be in the form of a community matrix. The basic form is to have the species as columns and the plots as rows. The package vegan has several example data sets. One of the examples comes from the famous 50 hectare plot on Barro Colorado Island, Panama. The data set has 50 rows, one for each of the hectares in the plot, and one column for each of the 225 species found there. The data are numbers of individuals >10cm dbh. Well read a lot of primary literature based on this plot, and you have the data to play with. Paste the following into R: data(BCI) #loads BCI data dim(BCI) #gives you the dimensions of the data set, (rows, columns) BCI[1:10,20:25] #shows the data for rows 1:10 and columns 20:25 If you cant see everything at once, scroll up: R records the output. Now, look at the first 5 rows of columns 70:75. What are the species that these represent? You can also address data in R like you do in excel, but you cant see the actual cells in spreadsheet form. Here are three ways to look at the data for the species Faramea occidentalis: BCI[1:50, 71] #addresses the row and column of the data set BCI[,71] #R takes no value for rows to mean "show them all" BCI$Faramea.occidentalis #you can also use the dataset$column address form Now, to get fancy, get the total number of individuals for Faramea occidentalis in the 50 ha plot: sum(BCI$Faramea.occidentalis) #adds up all the individuals Now, how many individuals of the tree Poulsenia.armata are in the 50 ha plot? Diversity Indices: In chapter 1 you read about several different diversity indices. Each has their own strengths and weaknesses. Vegan has ways of looking at them. Find the Shannon and Fisher diversity for the BCI plot. Use the command fisher.alpha (see the HTML help or type ?fisher.alpha at the R prompt). Lets calculate Shannons index for each of the 50 hectares in the BCI plot: diversity(BCI, index = "shannon") Now, we can use those 50 hectares to get the average diversity for a hectare on Barro Colorado Island: plots<- diversity(BCI, index = "shannon") #makes an object #that the diversity values #are written to. summary(plots) #gives summary statistics for the plots median(plots) #gives the median mean(plots) #gives the mean You get the picture! Now, what is the median diversity for hectares 1:10? 40:50? Fishers alpha is a measure of diversity that takes into account variability in stem number. You can calculate that with vegan as well: fish.a<-fisher.alpha(BCI, MARGIN = 1) fish.a #shows you the values in the object "fish.a" that you made. This returns Fishers alpha for all of the hectares. Since Fishers alpha is supposed to be invariant with sample size, we can test that with the BCI data: bcitot<-apply(BCI, 2, sum) #gives you the total number of individuals for the 50 ha plot bcitot.a<- fisher.alpha(bcitot, MARGIN = 1) #calculates fishers alpha on all 50 ha combined. Advanced: If we wanted to, we could use R to do advanced things like calculate Fishers alpha for increasing numbers of hectares combined. R allows you to program using loops very much like the language C: x<-1:50 #makes a sequence of numbers 1:50 that represent the hectares a<-NULL #sets up an empty object well fill with results for (i in x){ #tells R to give i each value in the object x b<- apply(BCI[1:i,], 2, sum) #get the sum of i hectares. c<- fisher.alpha(b, MARGIN = 1) #work the fishers alpha magic on it a<- c(a,c) #stick the new result on the end of the old data } plot(x, a) #plot the data Discussion question: If you had only one plot and you wanted to compare it to another plot (using any diversity measure), how would you do it? Species richness: Another simple measure of species diversity is simply the number of species, or species richness. Again, this gives you a number, but how would you compare two plots that varied in the number of individuals that they have? In your readings for chapter 16 you came across the idea of rarefaction. This simply means taking a random sample of a smaller, standard size from the plot to compare with other plots that have the same number of individuals. So, for example, wed express our result as diversity per 50 individuals. In R we can implement it as: rar <- rarefy(BCI, 20) #gives you the species per 20 individuals sampled for each of 50 ha rarsum<-rarefy(bcitot, 20, MARGIN=2) #species per 20 from whole plot, margin is 2 because bcitot has the data as a column and not a row Species-area and species-individual curves: spa <- specaccum(BCI) plot(spa) #plots the species accumulation curve and the confidence intervals for sites. plot(spa, ci.type="poly", col="blue", lwd=2, ci.lty=0, ci.col="lightblue") #males a prettier plot Were often interested in comparing accumulation of individuals, though rather than areas: spi<-specaccum(BCI, method="rarefaction") plot(spi) Now, add the species accumulation curve for area you did before, spa: plot(spa, add=TRUE, col=4) #color number 4 is blue You can see that the answers differ slightly. Note how you can add plots. This will become important when comparing among Manu plots, or comparing Manu and BCI. ORDINATION Distance measures (open the ordination web site to use as a reference). We are very used to thinking about geographical distances between places. When we think about the similarity or differences between community, we can also think of distances or dissimilarity. To do this we need some metricmeasureof distances between communities based on their individual compositions. There are lots of ways of doing this. Well look at two simple distance measures based on compositional similarity. One is the Bray-Curtis index, or percent similarity including species abundances, and the other is Sorensons index, or percent similarity based on presence-absence data. For the 50 hectare plot, compute the distances among the rows (hectares): bc<-vegdist(BCI, method="bray", binary=FALSE) #binary=FALSE means you look at the number of individuals. TRUE would give the result for presence-absence (Sorensons index) bc The result is the lower part of a 50x50 matrix. It looks just like the distance table on a road map. Imagine instead of having the distances between Asheville and Greensboro, and Greensboro and Raleigh, etc., that you have the distance, in vegetation composition, between any pair of hectares. These distances can be used to test hypotheses. For example, what if you knew that some hectares were farther apart than others? Then you could test the hypothesis that vegetational similarity is just a function of proximity. You could take another matrix of geographic distances and look at the correlation between the two using a Mantel Test. ( HYPERLINK "http://en.wikipedia.org/wiki/Mantel_test" http://en.wikipedia.org/wiki/Mantel_test) Arranging plots by compositional distance: NMDS Just like you can take all the distances between cities in North Carolina and reconstruct a map of the state (try it!), you can take the compositional distances between communities and reconstruct a map of the plots in community composition space. The ordination technique is called Non-metric Multidimensional Scaling. You can look at compositional similarity of sites based on species, or species based on sites (the former is more common). Try it on the 50 ha plot: bci.mds<-metaMDS(BCI, distance = "bray", k = 2, trymax = 20, autotransform =TRUE, noshare = 0.1, expand = TRUE, trace = 1, plot = FALSE) #makes the object bci.mds using Bray-Curtis ordination plot(bci.mds, choices = c(1, 2), type="n") #plots the ordination axes points(bci.mds, display = c("sites", "species"))#displays both sites and species on the same plot. Try choosing just sites to reduce clutter text(bci.mds, display = c("sites", "species")) Try just using the text without the points to make it look nicer. Now that you have these techniques in your arsenal. Load the Manu and Hanging Rock data hrtrees<-read.table("hrtrees.txt", header = TRUE, sep = "\t") manu<- read.table("manu.txt", header = TRUE, sep = "\t") and answer the following questions: Diversity: What is more diverse, a hectare of terra firme forest in Manu, or a hectare on BCI? Ordination: Does geomorphology influence species composition in the Manu data set? What is the most diverse habitat in Manu? Hanging Rock and WFU Campus Tree plots: Which plot was most diverse? Do the sites fall into discrete clusters? &45    , . 3 L N O O ] " L M > ? A C    ! - K M N O m n »Ʒw h6zhCJOJQJ^JaJhD.h/ /hh@ph6z0Jjh6zU h6zh6zjh6zUh6z hYyhThYyhThw hN5:h ehN:h  hw h hN hw hNhhN5 hN5.56 M   > N O ] ^  ! " L M 0^`0gd gdN  # A B C    P2ngdN12?^_lmnrsZ[PQ4VŴŦŕŴŴŴq h6zh<CJOJQJ^JaJh<hh  h h  h6zh,TCJOJQJ^JaJh,TCJOJQJ^JaJ h6zhCJOJQJ^JaJ h6zh CJOJQJ^JaJh h6 h h h h  h 6h h hD.hD.,Z[QR34VW'^~^gd%:)Vz:;DE  PXwiiiih~CJOJQJ^JaJh~h~CJaJ h~h~h~h~h~5h~h~OJQJ^Jh"tCJOJQJ^JaJ h"th"tCJOJQJ^JaJ h"th<CJOJQJ^JaJ h6zh,Th"th%:)CJOJQJ^JaJ h6zh<CJOJQJ^JaJh<)yz:;  QD./0BL]Lgd~gd<CN-. W!Y!!!!!!!!!!Dz}l}l}[}ShJv{CJaJ hJv{h6 CJOJQJ^JaJ h6zh,TCJOJQJ^JaJ hJv{hJv{CJOJQJ^JaJh6 CJaJh6 CJOJQJ^JaJhJv{CJOJQJ^JaJ h6 h6 CJOJQJ^JaJhJv{h6 h"th~h"tCJOJQJ^JaJh~CJOJQJ^JaJ h"th"tCJOJQJ^JaJ BCtuY Z [ X!Y!!!!!!0"1"d"e"# # ###gd6 gd~!0"1"9"d"e"## ###'#]#^#k%s%%%%& &&&+&,&&&&'4)6)ûóykk]VOKh%:) hx hx hx h6zhptcCJOJQJ^JaJhx CJOJQJ^JaJ h6zh6zCJOJQJ^JaJh6zCJOJQJ^JaJhptcCJaJ h6zh6zhx h/ /h6zhx h/ /:hx h/ /5h/ /CJaJ h6zhFtp hJv{hJv{CJOJQJ^JaJh%:)CJOJQJ^JaJhJv{CJaJ h6zhJv{#^#_#%%%%&&&)))))++h,i,,?-n-o--- . .K..gdD.gd/ /6)7)C)k)m)n)o)))))))))))))++++ ,0,g,n,u,,,,,,,,,,,,-$->-D-K-n-o-Ž h6zh,Th Wh2'hg4h/ / h6zh6z hx hx h[ h/ /:hx hptc: hx :hx h/ /:hptcCJaJhx h:h6zh+!h:0Jjh:U h:h:h:jh:U-o---- . . .!.".-...].^.f.g.......... ///////////ǶǶǶǶܯܯ৚ؖؒh WhYy h W:h Whg46hQ%phg4: hg4hD. h6zh,TCJOJQJ^JaJ hD.hD.CJOJQJ^JaJhg4hD. hg4hg4hg4CJaJ h Wh WCJOJQJ^JaJ"..... / /////// ,1h/ =!"#$% DyK http://ordination.okstate.edu/yK >http://ordination.okstate.edu/DyK )http://en.wikipedia.org/wiki/Mantel_testyK Rhttp://en.wikipedia.org/wiki/Mantel_testb 2 0@P`p2( 0@P`p 0@P`p 0@P`p 0@P`p 0@P`p 0@P`p8XV~PJ_HmH nHsH tHD`D NormalCJ_HaJmH nHsH tHDA`D Default Paragraph FontRiR  Table Normal4 l4a (k (No List e@ <HTML Preformatted7 2( Px 4 #\'*.25@9CJOJQJ^JaJ6U@6 6z Hyperlink >*B*phPK![Content_Types].xmlN0EH-J@%ǎǢ|ș$زULTB l,3;rØJB+$G]7O٭V$ !)O^rC$y@/yH*񄴽)޵߻UDb`}"qۋJחX^)I`nEp)liV[]1M<OP6r=zgbIguSebORD۫qu gZo~ٺlAplxpT0+[}`jzAV2Fi@qv֬5\|ʜ̭NleXdsjcs7f W+Ն7`g ȘJj|h(KD- dXiJ؇(x$( :;˹! I_TS 1?E??ZBΪmU/?~xY'y5g&΋/ɋ>GMGeD3Vq%'#q$8K)fw9:ĵ x}rxwr:\TZaG*y8IjbRc|XŻǿI u3KGnD1NIBs RuK>V.EL+M2#'fi ~V vl{u8zH *:(W☕ ~JTe\O*tHGHY}KNP*ݾ˦TѼ9/#A7qZ$*c?qUnwN%Oi4 =3N)cbJ uV4(Tn 7_?m-ٛ{UBwznʜ"Z xJZp; {/<P;,)''KQk5qpN8KGbe Sd̛\17 pa>SR! 3K4'+rzQ TTIIvt]Kc⫲K#v5+|D~O@%\w_nN[L9KqgVhn R!y+Un;*&/HrT >>\ t=.Tġ S; Z~!P9giCڧ!# B,;X=ۻ,I2UWV9$lk=Aj;{AP79|s*Y;̠[MCۿhf]o{oY=1kyVV5E8Vk+֜\80X4D)!!?*|fv u"xA@T_q64)kڬuV7 t '%;i9s9x,ڎ-45xd8?ǘd/Y|t &LILJ`& -Gt/PK! ѐ'theme/theme/_rels/themeManager.xml.relsM 0wooӺ&݈Э5 6?$Q ,.aic21h:qm@RN;d`o7gK(M&$R(.1r'JЊT8V"AȻHu}|$b{P8g/]QAsم(#L[PK-![Content_Types].xmlPK-!֧6 0_rels/.relsPK-!kytheme/theme/themeManager.xmlPK-!0C)theme/theme/theme1.xmlPK-! ѐ' theme/theme/_rels/themeManager.xml.relsPK] 'LV!6)o-/!#$ B#./ "% Nm6!n!!'XX8@0(  B S  ? OLE_LINK1 OLE_LINK2$$'$$'r9r9r9r9r9r9r9r9r9r9r9r9r9r9r9r9r9r9r9[ LGUett " ""'     b SP_o{{"""' 9*urn:schemas-microsoft-com:office:smarttagsState9*urn:schemas-microsoft-com:office:smarttagsplace8*urn:schemas-microsoft-com:office:smarttagsCityB*urn:schemas-microsoft-com:office:smarttagscountry-region= *urn:schemas-microsoft-com:office:smarttags PlaceName= *urn:schemas-microsoft-com:office:smarttags PlaceType 7  ps n " : LSzJMux+1#)-35>  #########$C$J$n$u$$$D%K% &&& &K&O&R&\&&&' os   P T 2 6  " 4>&'/^e~HJux16NY##i$n$$$?%D%%% &&K&&&&'333333333333333333333333333333333333333333333>O#CFv 4W~Nx0BWt 5!!%?%K&& ''' n^ _ l m 6!!$$$$!&"&-&.&]&^&f&g&'''_2 m>+=D EA ,Lw7# 6 x j$ * Y6 y }S1|-" W L,D.) =[ &q"S#q$1%fL%2'5'#(UT(~%)8)%:)O+./ /1i1023a5'7?M<h=;@C B 2BkHC(EsEtEWFtF*wF|hGjqGI*%Ia JJ3J/M#xP*TsVVWX.YsY?\I]3z]e^*`rIaptcfdelh_fixj kpknn?,nJyn%pQ%pFtpp tVtsft @uiduONwAx[x[zJv{Z|  Ce o]g4ts1d<^|]c{^|Yyt T5Um6i"tS`j =x U!RsTYLPd92TN[n\ Jf hci_[ B"KR6zT]Yg_~C(iK OhroxzKI4wXnr,Tx;c0 * N ?P03^;lV ^~Q[C*).;7:mb Y7U ''@&&&&'@UnknownG*Ax Times New Roman5Symbol3. *Cx Arial?= *Cx Courier New;(SimSun[SOACambria Math"qhV&2 "H"H"24'' 3HX ?2!xx AToday we ll be using the R package  vegan for community analyses Wake ForestWFU2011Oh+'0 $8 HT t   DToday well be using the R package vegan for community analyses Wake Forest Normal.dotmWFU20114Microsoft Office Word@b@hW+@]x@I"՜.+,D՜.+,< hp  Wake Forest UniversityH' BToday well be using the R package vegan for community analyses TitleH 8@ _PID_HLINKSA K7)http://en.wikipedia.org/wiki/Mantel_test|uhttp://ordination.okstate.edu/  !"#$%&()*+,-.0123456789:;<=>?@ABCDFGHIJKLNOPQRSTWRoot Entry F0}XYData '1Table/ *WordDocument.LSummaryInformation(EDocumentSummaryInformation8MCompObjr  F Microsoft Word 97-2003 Document MSWordDocWord.Document.89q