install_github("jaroyle/oSCR.com") library(oSCR) library(raster) library(sp) library(GlobalOptions) library(rgdal) library(car) library(carData) library(devtools) library(ggplot2) library(viridis) library(plotly) library(hrbrthemes) library(tidyverse) ##### OPEN FILES ##### edf= read.csv(file.choose(), header=TRUE, sep=";") tdf_PM= read.csv(file.choose(), header=TRUE, sep=";") tdf_TT = read.csv(file.choose(), header=TRUE, sep=";") tdf_RF = read.csv(file.choose(), header=TRUE, sep=";") tdf_PF = read.csv(file.choose(), header=TRUE, sep=";") #check data head(tdf_PM) head(tdf_TT) head(edf) tail(edf) str(edf$Trap) str(tdf_PM$Trap) str(tdf_TT$Trap) # create the scrFrame from edf and tdf margay_data <- data2oscr(edf = edf, sess.col = which(colnames(edf) %in% "Session"), id.col = which(colnames(edf) %in% "Id"), occ.col = which(colnames(edf) %in% "Occasion"), trap.col = which(colnames(edf) %in% "Trap"), sex.col = which(colnames(edf) %in% "Sex"), sex.nacode = "U", K=c(62,62,62,56), tdf = list(tdf_junto_RF,tdf_junto_PM,tdf_junto_TT,tdf_junto_PF), ntraps = c(20,21,20,20), trapcov.names=c("birds","dogs","small_mammals", "dist_sensor", "sensor","cows","ocelot","humans", "mapbiomas", "cats","pets"), tdf.sep = "/", remove.zeros = T, remove.extracaps = F) par(mfrow=c(1,1),mar=c(6,6,4,2)) plot(margay_data.sf) sf.BPWR <- subFrame(margay_data.sf,subs=1) plot(sf.BPWR) sf.promata <- subFrame(margay_data.sf,subs=2) plot(sf.promata) sf.teutonia<- subFrame(margay_data.sf,subs=3) plot(sf.teutonia) sf.PFNF<- subFrame(margay_data.sf,subs=4) plot(sf.PFNF) margay_data.ss <- make.ssDF(scrFrame = margay_data.sf, buffer=2000, res =250) plot(margay_data.ss,margay_data.sf) plot(margay_data.ss) ############ EXCLUDING WATER AREAS (33- MapBiomas) #### JUST SELECETED OTHERS CATEGORIES #### rast8 = MapBiomas rast_MapBiomas <-raster(file.choose()) rast_MapBiomas_new <- stack(rast_MapBiomas) plot(rast_MapBiomas) #function to extract raster to ssDF columm cov.name = "MapBiomas" extract.rast(margay_data.ss, rast_MapBiomas_new, mult = 1, cov.name = "MapBiomas", func = mean) { for(i in 1:length(margay_data.ss)) { tmpS <- margay_data.ss[[i]][, c("X", "Y")]*mult #separate the X and Y tmpS <- data.frame(x=tmpS$X,y=tmpS$Y) #criate a dataframe with X and Y coordinates(tmpS) <- ~x+y #criate coordinates with X and Y class(tmpS) proj4string(tmpS) <- CRS("+proj=utm +zone=22 +south +datum=WGS84 +units=m +ellps=WGS84") # curret projetion tmpS2 <- spTransform(tmpS,CRS("+proj=longlat +datum=WGS84 +zone=22 +south")) #convert coordinates to longlat r1 <- raster::extract(rast_MapBiomas_new,tmpS2, buffer=buffer, fun = fun, na.rm=T) #extract to the raster r2<- as.factor(r1) margay_data.ss[[i]][, cov.name] <- r2 #print like a new column } margay_data.ss } margay_data.ss plot(margay_data.ss,margay_data.sf) # Promata # 3 9 12 15 21 catg3_PM<- subset(margay_data.ss[[2]], MapBiomas==3 ) catg9_PM<- subset(margay_data.ss[[2]], MapBiomas==9 ) catg12_PM<- subset(margay_data.ss[[2]], MapBiomas==12 ) catg15_PM<- subset(margay_data.ss[[2]], MapBiomas==15 ) catg21_PM<- subset(margay_data.ss[[2]], MapBiomas==21 ) margay_data.ss_PM<- rbind(catg3_PM, catg9_PM,catg12_PM, catg15_PM,catg21_PM) # Teutonia # Levels: 3 9 15 19 21 24 25 catg3_TT<- subset(margay_data.ss[[3]], MapBiomas==3 ) catg9_TT<- subset(margay_data.ss[[3]], MapBiomas==9 ) catg15_TT<- subset(margay_data.ss[[3]], MapBiomas==15 ) catg19_TT<- subset(margay_data.ss[[3]], MapBiomas==19 ) catg21_TT<- subset(margay_data.ss[[3]], MapBiomas==21 ) catg24_TT<- subset(margay_data.ss[[3]], MapBiomas==24 ) catg25_TT<- subset(margay_data.ss[[3]], MapBiomas==25 ) margay_data.ss_TT<- rbind(catg3_TT,catg9_TT,catg15_TT,catg19_TT,catg21_TT,catg24_TT,catg25_TT) # BPWR # c(3,9,11,12,19,21,24)) catg3_RF<- subset(margay_data.ss[[1]], MapBiomas==3 ) catg9_RF<- subset(margay_data.ss[[1]], MapBiomas==9 ) catg11_RF<- subset(margay_data.ss[[1]], MapBiomas==11 ) catg12_RF<- subset(margay_data.ss[[1]], MapBiomas==12 ) catg19_RF<- subset(margay_data.ss[[1]], MapBiomas==19 ) catg21_RF<- subset(margay_data.ss[[1]], MapBiomas==21 ) catg24_RF<- subset(margay_data.ss[[1]], MapBiomas==24 ) margay_data.ss_BPWR<- rbind(catg3_RF,catg9_RF,catg11_RF,catg12_RF,catg19_RF, catg21_RF,catg24_RF) # Passo Fundo - PFNF # 3 9 12 15 19 21 24 25 catg3_PF <- subset(margay_data.ss[[4]], MapBiomas==3 ) catg9_PF <- subset(margay_data.ss[[4]], MapBiomas==9 ) catg12_PF <- subset(margay_data.ss[[4]], MapBiomas==12 ) catg15_PF <- subset(margay_data.ss[[4]], MapBiomas==15 ) catg19_PF <- subset(margay_data.ss[[4]], MapBiomas==19 ) catg21_PF <- subset(margay_data.ss[[4]], MapBiomas==21 ) catg24_PF <- subset(margay_data.ss[[4]], MapBiomas==24 ) catg25_PF <- subset(margay_data.ss[[4]], MapBiomas==25 ) margay_data.ss_PFNF<- rbind(catg3_PF,catg9_PF, catg12_PF,catg15_PF,catg19_PF,catg21_PF,catg24_PF,catg25_PF) #### NEW ssDF #### margay_data.ss_new <-list(margay_data.ss_BPWR, margay_data.ss_PM, margay_data.ss_TT, margay_data.ss_PFNF) #testing plots par(mfrow=c(1,1)) plot(margay_data.ss_new[[4]]$X, margay_data.ss_new[[4]]$Y, pch=20, cex=2.6, col="darkgrey", asp = TRUE, axes=FALSE, xlab="", ylab="") points(margay_junto.sf$traps[[4]], pch=15, cex=3, col="black") ########### ADD covariates in the ssDF #### ####rast1 = distance river rast1 <-raster(file.choose()) plot(rast1) rast1 rast1_s<-scale(rast1) rast1_s plot(rast1_s) cov.name = "dis_river" #1km extract.rast(margay_data.ss_new, rast1_s, mult = 1, cov.name = "dis_river", func = mean) { for(i in 1:length(margay_data.ss_new)) { tmpS <- margay_data.ss_new[[i]][, c("X", "Y")]*mult #separate the X and Y tmpS <- data.frame(x=tmpS$X,y=tmpS$Y) #criate a dataframe with X and Y coordinates(tmpS) <- ~x+y #criate coordinates with X and Y class(tmpS) proj4string(tmpS) <- CRS("+proj=utm +zone=22 +south +datum=WGS84 +units=m +ellps=WGS84") # curret projetion tmpS2 <- spTransform(tmpS,CRS("+proj=merc +a=6378137 +b=6378137 +lat_ts=0.0 +lon_0=0.0 +x_0=0.0 +y_0=0 +k=1.0 +units=m +nadgrids=@null +no_defs")) #convert coordinates to longlat r2 <- raster::extract(rast1_s,tmpS2, fun = fun, na.rm=T) #extract to the raster margay_data.ss_new[[i]][, cov.name] <- r1 #print like a new column } margay_data.ss_new } margay_data.ss_new #### rast2 = ndvi Vegetation cover #### rast2 <-raster(file.choose()) plot(rast2) rast2_s<-scale(all_rast5) all_rast5_s cov.name ="ndvi" #1km *0.004 não escalado extract.rast(margay_data.ss_new,rast2, mult = 1, cov.name = "ndvi", func = mean) { for(i in 1:length(margay_data.ss_new)) { tmpS <- margay_data.ss_new[[i]][, c("X", "Y")]*mult #separate the X and Y tmpS <- data.frame(x=tmpS$X,y=tmpS$Y) #criate a dataframe with X and Y coordinates(tmpS) <- ~x+y #criate coordinates with X and Y class(tmpS) proj4string(tmpS) <- CRS("+proj=utm +zone=22 +south +datum=WGS84 +units=m +ellps=WGS84") # curret projetion tmpS2 <- spTransform(tmpS,CRS("+proj=longlat +datum=WGS84 +zone=22 +south")) #convert coordinates to longlat r1_ndvi<- raster::extract(all_rast5,tmpS2, buffer=buffer, fun = fun, na.rm=T) #extract to the raster r2<- r1_ndvi*0.004 #r3<- scale(r2) margay_data.ss_new[[i]][, cov.name] <- r2 #print like a new column } margay_data.ss_new } margay_data.ss_new #### rast3 = humans_dens rast3 <-raster(file.choose()) rast3<-crop(rast3, extent(rast_MapBiomas)) #crop the world raster to the area plot(rast3) rast3_s<-scale(rast3) plot(rast3_s) #function to extract raster to ssDF columm cov.name = "humans_dens" extract.rast(margay_data.ss_new,rast3_s, mult = 1, cov.name = "humans_dens", func = mean) { for(i in 1:length(margay_junto.ss)) { tmpS <- margay_data.ss_new[[i]][, c("X", "Y")]*mult #separate the X and Y tmpS <- data.frame(x=tmpS$X,y=tmpS$Y) #criate a dataframe with X and Y coordinates(tmpS) <- ~x+y #criate coordinates with X and Y class(tmpS) proj4string(tmpS) <- CRS("+proj=utm +zone=22 +south +datum=WGS84 +units=m +ellps=WGS84") # curret projetion tmpS2 <- spTransform(tmpS,CRS("+proj=longlat +datum=WGS84 +zone=22 +south")) #convert coordinates to longlat r1 <- raster::extract(all_rast6_s,tmpS2, buffer=buffer, fun = fun, na.rm=T) #extract to the raster margay_data.ss_new[[i]][, cov.name] <- r1 #print like a new column } margay_data.ss_new } margay_data.ss_new #rast4 = dist_roads rast4 <-raster(file.choose()) plot(rast4) rast4_s<-scale(rast4) plot(rast4_s) #function to extract raster to ssDF columm cov.name ="dist_road" extract.rast(margay_data.ss_new, rast4_s, mult = 1, cov.name = "dist_road", func = mean) { for(i in 1:length(margay_data.ss_new)) { tmpS <- margay_data.ss_new[[i]][, c("X", "Y")]*mult #separate the X and Y tmpS <- data.frame(x=tmpS$X,y=tmpS$Y) #criate a dataframe with X and Y coordinates(tmpS) <- ~x+y #criate coordinates with X and Y class(tmpS) proj4string(tmpS) <- CRS("+proj=utm +zone=22 +south +datum=WGS84 +units=m +ellps=WGS84") # curret projetion tmpS2 <- spTransform(tmpS,CRS("+proj=merc +a=6378137 +b=6378137 +lat_ts=0.0 +lon_0=0.0 +x_0=0.0 +y_0=0 +k=1.0 +units=m +nadgrids=@null +no_defs")) #convert coordinates to longlat r1 <- raster::extract(rast4_s,tmpS2, buffer=buffer, fun = fun, na.rm=T) #extract to the raster margay_data.ss_new[[i]][, cov.name] <- r1 #print like a new column } margay_data.ss_new } margay_data.ss_new #### Correlation TEST ##### library(Hmisc) r1<-rcorr(as.matrix(margay_data.ss_new[[1]][3:5]),type="pearson") r1 r2<-rcorr(as.matrix(margay_data.ss_new[[2]][3:5]),type="pearson") r2 r3<-rcorr(as.matrix(margay_data.ss_new[[3]][3:5]),type="pearson") r3 r4<-rcorr(as.matrix(margay_data.ss_new[[4]][3:5]),type="pearson") r4 ############################### CRIATE MODELS #################################### ######## NULL MODEL #### null<- oSCR.fit(list(D ~ 1, p0 ~ 1 , sig ~ 1),margay_data.sf,margay_data.ss_new,encmod="P", se=T) ######## MODELS SIG ##### sex_sig <- oSCR.fit(list(D ~ 1, p0 ~ 1,sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P", se=T) session_sig <- oSCR.fit(list(D ~ 1, p0 ~ 1,sig ~ session),margay_data.sf,margay_data.ss_new,encmod="P", se=T) session_sex_sig <- oSCR.fit(list(D ~ 1, p0 ~ 1,sig ~ sex+session),margay_data.sf,margay_data.ss_new,encmod="P", se=T) ######## MODELS SIG SELECTION #### list_sig_models <- fitList.oSCR(list( sex_sig, session_sig, session_sex_sig, null),rename=TRUE) list_sig_models order_sig_models <-modSel.oSCR(list_sig_models) order_sig_models$aic.tab[,c(2,3,4,5,6,7)] ########## MODELS DETECTION ####### birds_p<- oSCR.fit(list(D ~ 1, p0 ~ birds , sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P",se=T) small_mammals_p<- oSCR.fit(list(D ~ 1, p0 ~ small_mammals , sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P",se=T) dogs_p <- oSCR.fit(list(D ~ 1, p0 ~ dogs , sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P",se=T) cows_p <- oSCR.fit(list(D ~ 1, p0 ~ cows , sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P",se=T) ocelot_p <- oSCR.fit(list(D ~ 1, p0 ~ ocelot , sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P",se=T) dist_sensor_p <- oSCR.fit(list(D ~ 1, p0 ~ dist_sensor , sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P",se=T) sensor_p <- oSCR.fit(list(D ~ 1, p0 ~ sensor , sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P",se=T) humans_p <- oSCR.fit(list(D ~ 1, p0 ~ humans , sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P",se=T) session_p <- oSCR.fit(list(D ~ 1, p0 ~ session , sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P",se=T) cats_p <- oSCR.fit(list(D ~ 1, p0 ~ cats , sig ~ 1),margay_data.sf,margay_data.ss_new,encmod="P",se=T) mix_p<- oSCR.fit(list(D ~ 1, p0 ~ sensor+dist_sensor , sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P",se=T) mix_p2 <- oSCR.fit(list(D ~ 1, p0 ~ birds+small_mammals, sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P",se=T) mix_p3 <- oSCR.fit(list(D ~ 1, p0 ~ dogs+humans+cows, sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P",se=T) sex_p<-oSCR.fit(list(D ~ 1, p0 ~ sex, sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P",se=T) mix_p5<- oSCR.fit(list(D ~ 1, p0 ~ dogs+humans , sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P",se=T) mix_p6 <- oSCR.fit(list(D ~ 1, p0 ~ cats+dogs, sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P",se=T) ####### DETECTION MODELS SELECTION #### list_detection_models <- fitList.oSCR(list( birds_p, dogs_p, small_mammals_p, cows_p, ocelot_p, dist_sensor_p, sensor_p, cats_p, session_p, mix_p, mix_p2, sex_sig, null ),rename=TRUE) order_detection_models <-modSel.oSCR(list_detection_models) order_detection_models$aic.tab mavg2<- ma.coef(order_detection_models) order_detection_models2 <-modSel.oSCR(list_detection_models2) ################# MODELS DENSITY ########### session_d2 <- oSCR.fit(list(D ~ session, p0 ~ birds+small_mammals , sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P",se=T) ndvi_d <- oSCR.fit(list(D ~ ndvi, p0 ~ birds+small_mammals , sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P", se=T) population_d <- oSCR.fit(list(D ~ humans_dens, p0 ~ birds+small_mammals ,sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P", se=T) dis_river_d <-oSCR.fit(list(D ~ dis_river, p0 ~ birds+small_mammals ,sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P", se=T) dis_road_d <-oSCR.fit(list(D ~ dist_road, p0 ~ birds+small_mammals ,sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P", se=T) ndvi_dis_river_d2 <- oSCR.fit(list(D ~ ndvi13+dis_river , p0 ~ b+birds+small_mammals , sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P", se=T) population_dist_road_d <- oSCR.fit(list(D ~ humans_dens+dist_road , p0 ~ birds+small_mammals , sig ~ sex),margay_data.sf,margay_data.ss_new,encmod="P", se=T) #### MODELS DENSITY SELECTION #### list_density_models <- fitList.oSCR(list(session_d2, ndvi_d, population_d, dis_river_d, dis_road_d, ndvi_dis_river_d, population_dist_road_d, null),rename=TRUE) list_density_models order_density_models_2 <-modSel.oSCR(list_density_models) order_density_models_2$aic.tab ######### PREDICAO DOS MELHORES MODELOS ##### par(mfrow=c(2,2)) pred <- predict.oSCR(ndvi_d,margay_data.sf, margay_data.ss_new) map_BPWR <- pred$r[[1]] plot(map_BPWR) plot(map_BPWR) proj4string(map_BPWR) <- CRS("+proj=utm +zone=22 +south +datum=WGS84 +units=m +ellps=WGS84") # curret projetion map_BPWR_new <- writeRaster(map_BPWR, filename="density realized refugio2_new.tif", format="GTiff", overwrite=TRUE) plot(pred$r[[2]]) proj4string(pred$r[[2]]) <- CRS("+proj=utm +zone=22 +south +datum=WGS84 +units=m +ellps=WGS84") # curret projetion map_PM <- writeRaster(pred$r[[2]], filename="density realized promata.tif", format="GTiff", overwrite=TRUE) map_PM2.new <- writeRaster(map_PM, filename="density realized promata_new.tif", format="GTiff", overwrite=TRUE) plot(pred$r[[3]]) proj4string(pred$r[[3]]) <- CRS("+proj=utm +zone=22 +south +datum=WGS84 +units=m +ellps=WGS84") # curret projetion map_TT <- writeRaster(pred$r[[3]], filename="density realized teutonia_new.tif", format="GTiff", overwrite=TRUE) plot(pred$r[[4]]) proj4string(pred$r[[4]]) <- CRS("+proj=utm +zone=22 +south +datum=WGS84 +units=m +ellps=WGS84") # curret projetion map_PF <- writeRaster(pred$r[[4]], filename="density realized passofundo_new.tif", format="GTiff", overwrite=TRUE) dens_model_ndvi <- get.real(model = ndvi_d, type = "dens", d.factor = 16) write.csv(ddens_model_ndvi[[1]],file="Teste densidades") ###### Detection #### pred.det <- data.frame(session = c(margay_data.sf[[1]], margay_data.sf[[2]], margay_data.sf[[3]], margay_data.sf[[4]])) as.matrix(pred.det) pred.df <- data.frame(Session = factor(1)) det_model_ndvi<-get.real(model = ndvi_d, type = "det",pred.det) det_model_ndvi[[2]][[55]] det_r <- rasterFromXYZ(det_model_ndvi) plot(rs) plot(margay_data.sf) all_ndvi_d13_2$coef.mle pred.sig<-data.frame(session=rep(factor(1:4),2), sex=factor((1:0),each=4)) sig_model_ndvi<-get.real(model = ndvi_d, type = "sig") sig_model_ndvi<-get.real(model = ndvi_d, type = "sig", newdata = data.frame(sex=factor(1:0))) ggplot(dens_model_ndvi,aes(ndvi,estimate, group=Session, color=Session))+ geom_point() + geom_smooth(color="#69b3a2") refugio_ndvi<- subset(dens_model_ndvi, Session==1) # per sex RF ED.f_better_session1 <-dens_model_ndvi[[1]][which(dens_model_ndv[[1]]$sex=="f"),] ED.m_better_session1 <-dens_model_ndvi[[1]][which(dens_model_ndvi[[1]]$sex=="m"),] # per sex PM ED.f_better_session2 <- dens_model_ndvi[[2]][which(dens_model_ndvi[[2]]$sex=="f"),] ED.m_better_session2 <- dens_model_ndvi[[2]][which(dens_model_ndvi[[2]]$sex=="m"),] #per sex TT ED.f_better_session3 <- dens_model_ndvi[[3]][which(dens_model_ndvi[[3]]$sex=="f"),] ED.m_better_session3 <- dens_model_ndvi[[3]][which(dens_model_ndvi[[3]]$sex=="m"),] #per sex PF ED.f_better_session4 <- dens_model_ndvi[[4]][which(dens_model_ndvi[[4]]$sex=="f"),] ED.m_better_session4 <- dens_model_ndvi[[4]][which(dens_model_ndvi[[4]]$sex=="m"),] #Total Refúgio ED.tot.better_session1<- data.frame(ED.f_better_session1[,c("X","Y")], estimate = ED.f_better_session1$estimate +ED.m_better_session1$estimate, se=ED.f_better_session1$se + ED.m_better_session1$se, lwr=ED.f_better_session1$lwr + ED.m_better_session1$lwr, upr=ED.f_better_session1$upr + ED.m_better_session1$upr) #ndvi= (ED.f_better_session1[,"ndvi"])) #Session= ED.f_better_session1[,"Session"]) #ndvi6=ED.f_better_session1[,"ndvi6"]) n2<-mean(ED.tot.better_session1$estimate) #Total BPWR RF.fe <- data.frame(ED.f_better_session1[,c("X","Y")],estimate = ED.f_better_session1$estimate) ED.r.better_session1 <- rasterFromXYZ(ED.tot.better_session1[1][,1:3]) plot(ED.r.better_session1);#points(tdf_refugio[,2:3],pch=20) proj4string(ED.r.better_session1) <- CRS("+proj=utm +zone=22 +south +datum=WGS84 +units=m +ellps=WGS84") # curret projetion rf <- writeRaster(ED.r.better_session1, filename="densityrefugio4.tif", format="GTiff", overwrite=TRUE) par(mfrow=c(1,1)) #Total Pro-mata ED.tot.better_session2 <- data.frame(ED.f_better_session2[,c("X","Y")], estimate = ED.f_better_session2$estimate+ED.m_better_session2$estimate, se=ED.f_better_session2$se+ED.m_better_session2$se, lwr=ED.f_better_session2$lwr+ED.m_better_session2$lwr, upr=ED.f_better_session2$upr+ED.m_better_session2$upr) #ndvi= (ED.f_better_session2[,"ndvi"])) total_pm <-mean(ED.tot.better_session2$estimate) PM.fe <- data.frame(ED.f_better_session2[,c("X","Y")],estimate = ED.f_better_session2$estimate) ED.r.better_session2 <- rasterFromXYZ(ED.tot.better_session2[,1:3]) plot(ED.r.better_session2) plot(ED.r.better_session2);points(tdf_promata[,2:3],pch=20) proj4string(ED.r.better_session2) <- CRS("+proj=utm +zone=22 +south +datum=WGS84 +units=m +ellps=WGS84") # curret projetion rasterteuto <- writeRaster(ED.r.better_session2, filename="densityteuto.tif", format="GTiff", overwrite=TRUE) #Total Teutônia ED.tot.better_session3 <- data.frame(ED.f_better_session3[,c("X","Y")], estimate = ED.f_better_session3$estimate+ED.m_better_session3$estimate, se=ED.f_better_session3$se+ED.m_better_session3$se, lwr=ED.f_better_session3$lwr+ED.m_better_session3$lwr, upr=ED.f_better_session3$upr+ED.m_better_session3$upr ) total_TT <- mean(ED.tot.better_session3$estimate) ED.r.better_session3 <- rasterFromXYZ(ED.tot.better_session3[,1:3]) plot(ED.r.better_session3) plot(ED.r.better_session3);points(tdf_teutonia[,2:3],pch=20) proj4string(ED.r.better_session3) <- CRS("+proj=utm +zone=22 +south +datum=WGS84 +units=m +ellps=WGS84") # curret projetion rasterrefugio <- writeRaster(ED.r.better_session3, filename="densityrefugio.tif", format="GTiff", overwrite=TRUE) #Total Passo Fundo ED.tot.better_session4 <- data.frame(ED.f_better_session4[,c("X","Y")], estimate = ED.f_better_session4$estimate+ED.f_better_session4$estimate, se=ED.f_better_session4$se+ED.f_better_session4$se, lwr=ED.f_better_session4$lwr+ED.f_better_session4$lwr, upr=ED.f_better_session4$upr+ED.f_better_session4$upr ) Total_PF<-mean(ED.tot.better_session4$estimate) ED.r.better_session4 <- rasterFromXYZ(ED.tot.better_session4[,1:3]) plot(ED.r.better_session4) plot(ED.r.better_session4);points(tdf_passofundo[,2:3],pch=20) proj4string(ED.r.better_session4) <- CRS("+proj=utm +zone=22 +south +datum=WGS84 +units=m +ellps=WGS84") # curret projetion rasterpassofundo<- writeRaster(ED.r.better_session4, filename="densitypassofundo.tif", format="GTiff", overwrite=TRUE) ###### DENSIDADES #### Dens_RF<-data.frame(estimate=mean(ED.tot.better_session1$estimate), se=mean(ED.tot.better_session1$se), lwr=mean(ED.tot.better_session1$lwr), upr=mean(ED.tot.better_session1$upr)) Dens_RF Dens_PM<-data.frame(estimate=mean(ED.tot.better_session2$estimate), se=mean(ED.tot.better_session2$se), lwr=mean(ED.tot.better_session2$lwr), upr=mean(ED.tot.better_session2$upr)) Dens_TT<-data.frame(estimate=mean(ED.tot.better_session3$estimate), se=mean(ED.tot.better_session3$se), lwr=mean(ED.tot.better_session3$lwr), upr=mean(ED.tot.better_session3$upr)) Dens_PF<-data.frame(estimate=mean(ED.tot.better_session4$estimate), se=mean(ED.tot.better_session4$se), lwr=mean(ED.tot.better_session4$lwr), upr=mean(ED.tot.better_session4$upr)) ###### Figures ##### data_figure6 <- data.frame(Session = factor(c(1,2,3,4)), estimate=c(Dens_RF$estimate, Dens_PM$estimate, Dens_TT$estimate, Dens_PF$estimate), lwr=c(Dens_RF$estimate-Dens_RF$se, Dens_PM$estimate-Dens_PM$se, Dens_TT$estimate-Dens_TT$se, Dens_PF$estimate-Dens_PF$se), upr=c(Dens_RF$estimate+Dens_RF$se, Dens_PM$estimate+Dens_PM$se, Dens_TT$estimate+Dens_TT$se, Dens_PF$estimate+Dens_PF$se), se=c(Dens_RF$se, Dens_PM$se, Dens_TT$se, Dens_PF$se)) data_figure6 ggplot(data_figure6,aes(x=Session, y=estimate, group=Session, fill=Session)) + geom_boxplot() + scale_fill_viridis(discrete = F) + theme_minimal() + ggtitle("") + xlab("Session")+ ylab("Density( 100 km²)")+ theme( legend.position="none") data_figure5 <- data.frame(Session = dens_model_ndvi$Session, estimate= dens_model_ndvi$estimate, ndvi=dens_model_ndvi$ndvi) test.f<- dens_model_ndvi [which(dens_model_ndvi$sex=="f"),] test.m<- dens_model_ndvi [which(dens_model_ndvi$sex=="m"),] data_figure5<- data.frame(test.f[,c("X","Y","ndvi","Session","lwr","upr","se")], estimate = test.f$estimate+test.m$estimate) ggplot(data_figure5, aes(x=ndvi, y=estimate)) + geom_line(size=1.5)+ geom_ribbon(data=test.fm, aes(x=ndvi,ymin=lwr,ymax=upr),col="grey72",alpha=0.2,inherit.aes=F)+ theme_ipsum()+ geom_line(size=1.5) ##### Beta estimate Grafic - Figure 4 #### ndvi_d$outStats beta_estimate <- data.frame(ndvi_d$outStats) class(beta_estimate) beta_estimate2<- beta_estimate[-c(3,4,5,7,8,9),] #detection beta_estimate3<- beta_estimate[-c(1,2,5,6,7,8,9),] #sigma beta_estimate4<- beta_estimate[-c(1,2,3,4,5,6,9),] #density beta_estimate5 <- beta_estimate[-c(1,2,6,8),] beta_estimate6 <- beta_estimate5[c("4","5","3","7"),] test<- as.data.frame(beta_estimate6) beta_estimate6$ymin <- (beta_estimate6$mle-(1.88*beta_estimate6$std.er)) beta_estimate6$ymax <- (beta_estimate6$mle+(1.88*beta_estimate6$std.er)) ggplot(beta_estimate6, fill=parameters)+ ylim(-5,13)+ geom_point(aes(x=parameters, y=mle, ),width = 0.5, fill= "gray26",size=2, ) + geom_errorbar(aes(x=parameters, ymin=ymin, ymax=ymax), width=0.1, colour="gray40", alpha=0.9, size=0.5)+ scale_fill_viridis(discrete=F)+ xlab("")+ ylab("Beta estimate")+ geom_hline(yintercept = c(0), colour = "black", size = 0.5,linetype=1)+ theme(panel.background = element_rect(colour="white",fill="white"))+ coord_flip() #### CIRCULAR ANALYSES #### #Instale-o: #install.packages("circular") #Chame o pacote library(circular) #chamando os dados: margay_activty = read.csv(file.choose(), header=TRUE, sep=";") margay_activty #olhar o inicio dos dados head(margay_activty) #Para verificar se o arquivo foi lido corretamente, podemos usar o comando: str(margay_activty)#Se o R ler o mes como factor (palavra) e os angulo e as frequencias como int (numeros inteiros), esta tudo certo. #Nossa planilha deve ter uma coluna para os angulos e duas colunas para frequencias. #Para fazer as analises no R, precisamos apenas dos valores dos angulos e das frequencias. L1 <- rep(margay_activty$angulo, margay_activty$frequencia_refugio) L1 L2 <- rep(margay_activty$angulo , margay_activty$frequencia_promata) L2 L3 <- rep(margay_activty$angulo , margay_activty$frequencia_teutonia) L3 L4 <- rep(margay_activty$angulo , margay_activty$frequencia_passofundo) L4 L5 <- rep(margay_activty$angulo , margay_activty$frequencia_total) L5 #Para trabalharmos com os dados corretamente, temos que transforma-los em radianos: L1.rad <- rad(L1) L2.rad <- rad(L2) L3.rad <- rad(L3) L4.rad <- rad(L4) L5.rad <- rad(L5) #O proximo passo e transformar os dados em circular: L1.circ <- as.circular(L1.rad) L1.circ L2.circ <- as.circular(L2.rad) L3.circ <- as.circular(L3.rad) L4.circ <- as.circular(L4.rad) L5.circ <- as.circular(L5.rad) #Dessa maneira, criamos dois vetores, cujos valores dos angulos se repetem. Podemos, dessa maneira, comecar a visualizar os dados. Para isso, vamos colocar os dados no grafico: plot.circular(cbind(L1.circ, L2.circ), rotation = "clock", units = "rads") #Podemos tambem visualizar a frequencia dos angulos: rose.diag(cbind(L1.circ, L2.circ)) rose.diag(L1.circ) #Visao geral dos dados summary(L1.circ) #Graficos plot.circular(L1.circ, rotation = "clock", bins = 80, zero = pi/2,stack = T, units = "rads", axes = F, col = "blue", ticks = F) plot.circular(L2.circ, rotation = "clock", bins = 80, zero = pi/2,stack = T, units = "rads", axes = F, col = "green", ticks = F) plot.circular(cbind(L1.circ, L2.circ), rotation = "clock", bins = 80, zero = pi/2,stack = T, units = "rads", axes = F, col = c("blue", "green"), ticks = F) #Para colocar as horas: axis.circular(at=circular(sort(seq(0, 11/6*pi, pi/6), decreasing = T)), c(labels = c("5", "6","7", "8", "9", "10", "11", "12", "1", "2", "3","4"))) #Para colocar os vetores medios: arrows.circular(mean(L1.circ), rho.circular(L1.circ), zero = pi/2, rotation = "clock",col = "blue") arrows.circular(mean(L2.circ), rho.circular(L2.circ), zero = pi/2, rotation = "clock",col = "green") #Distribuicao dos dados #Uniformidade dos dados ###### Rayleigh Test #### rayleigh.test(L1.circ) rayleigh.test(L2.circ) rayleigh.test(L3.circ) rayleigh.test(L4.circ) rayleigh.test(L5.circ) ##### Raos spacing test ##### #Segundo BERGIN (1991)rao.spacing.test is more flexible; able to handle more types of circular data with fewer limiting assumptions; and is more powerful with small sample sizes. rao.spacing.test(L1.circ, alpha=0)#numeric value specifying the significance level of the test. The default value is 0, in which case, a range for the p-value will be returned. Valid significance levels are 0.10, 0.05, 0.01 and 0.001. rao.spacing.test(L2.circ, alpha=0) rao.spacing.test(L3.circ, alpha=0) rao.spacing.test(L4.circ, alpha=0) rao.spacing.test(L5.circ, alpha=0) #como se apresentam os dados em um estudo cientifico? Volta apresentacao #Existem outros testes no pacote. #Comparando amostras #Duas amostras:dados parametricos(teste chamado Watson) watson.two.test(L3.circ, L4.circ) #Duas amostras:dados nao-parametricos(teste chamado Watson-Wheeler) watson.wheeler.test(list(L1.circ, L4.circ)) #Obs: nao se esqueca de listar suas variaveis (list()), caso contrario o teste nao funciona para amostras de tamanhos diferentes. #Mais de duas amostras #O teste de Watson-Williams pode ser usado tambem para mais de duas amostras,quando elas possuem distribuicao de von Mises. Nao esqueca de listar(list(x, y, z)), os vetores: watson.williams.test(list(rad(L1), rad(L2), rad(L3), rad(L4))) #O teste de Watson-Wheeler, quando as amostras nao provem de populacao com distribuicao de von Mises, tambem pode ser usado para mais de duas variaveis: watson.wheeler.test(list(L1.circ,L4.circ)) ### Birds Activity pattern dados_birds = read.csv(file.choose(), header=TRUE, sep=";") dados_birds #olhar o inicio dos dados head(dados_birds) #Para verificar se o arquivo foi lido corretamente, podemos usar o comando: str(dados_birds)#Se o R ler o mes como factor (palavra) e os angulo e as frequencias como int (numeros inteiros), esta tudo certo. #Nossa planilha deve ter uma coluna para os angulos e duas colunas para frequencias. #Para fazer as analises no R, precisamos apenas dos valores dos angulos e das frequencias. B1 <- rep(dados_birds$angulo, dados_birds$birds_refugio) B1 B2 <- rep(dados_birds$angulo , dados_birds$birds_promata) B2 B3 <- rep(dados_birds$angulo , dados_birds$birds_teutonia) B3 B4 <- rep(dados_birds$angulo , dados_birds$birds_passofundo) B4 B5 <- rep(dados_birds$angulo , dados_birds$birds_total) B5 #Para trabalharmos com os dados corretamente, temos que transforma-los em radianos: B1_rad <- rad(B1) B2_rad <- rad(B2) B3_rad <- rad(B3) B4_rad <- rad(B4) B5_rad <- rad(B5) #O proximo passo e transformar os dados em circular: B1.circ <- as.circular(B1_rad) B1.circ B2.circ <- as.circular(B2_rad) B3.circ <- as.circular(B3_rad) B4.circ <- as.circular(B4_rad) B5.circ <- as.circular(B5_rad) #Dessa maneira, criamos dois vetores, cujos valores dos angulos se repetem. Podemos, dessa maneira, comecar a visualizar os dados. Para isso, vamos colocar os dados no grafico: plot.circular(cbind(B1.circ, B2.circ), rotation = "clock", units = "rads") #Distribuicao dos dados #Uniformidade dos dados ###### Rayleigh Test #### rayleigh.test(B1.circ) rayleigh.test(B2.circ) rayleigh.test(B3.circ) rayleigh.test(B4.circ) rayleigh.test(B5.circ) ##### Raos spacing test ##### #Segundo BERGIN (1991)rao.spacing.test is more flexible; able to handle more types of circular data with fewer limiting assumptions; and is more powerful with small sample sizes. rao.spacing.test(B1.circ, alpha=0)#numeric value specifying the significance level of the test. The default value is 0, in which case, a range for the p-value will be returned. Valid significance levels are 0.10, 0.05, 0.01 and 0.001. rao.spacing.test(B2.circ, alpha=0) rao.spacing.test(B3.circ, alpha=0) rao.spacing.test(B4.circ, alpha=0) rao.spacing.test(B5.circ, alpha=0) #como se apresentam os dados em um estudo cientifico? Volta apresentacao #Existem outros testes no pacote. #Comparando amostras #Duas amostras:dados parametricos(teste chamado Watson) watson.two.test(B3.circ, B4.circ) #Duas amostras:dados nao-parametricos(teste chamado Watson-Wheeler) watson.wheeler.test(list(B1.circ, B4.circ)) #Obs: nao se esqueca de listar suas variaveis (list()), caso contrario o teste nao funciona para amostras de tamanhos diferentes. #Mais de duas amostras #O teste de Watson-Williams pode ser usado tambem para mais de duas amostras,quando elas possuem distribuicao de von Mises. Nao esqueca de listar(list(x, y, z)), os vetores: watson.williams.test(list(rad(B1), rad(B2), rad(L3), rad(B4))) #O teste de Watson-Wheeler, quando as amostras nao provem de populacao com distribuicao de von Mises, tambem pode ser usado para mais de duas variaveis: watson.wheeler.test(list(B1.circ,B3.circ)) #Small_mammals activiy pattern #chamando os dados: dados_rodents = read.csv(file.choose(), header=TRUE, sep=";") dados_rodents #olhar o inicio dos dados head(dados_rodents) #Para verificar se o arquivo foi lido corretamente, podemos usar o comando: str(dados_rodents)#Se o R ler o mes como factor (palavra) e os angulo e as frequencias como int (numeros inteiros), esta tudo certo. #Nossa planilha deve ter uma coluna para os angulos e duas colunas para frequencias. #Para fazer as analises no R, precisamos apenas dos valores dos angulos e das frequencias. R1 <- rep(dados_rodents$angulo, dados_rodents$rodent_refugio) R1 R2 <- rep(dados_rodents$angulo , dados_rodents$rodent_promata) R2 R3 <- rep(dados_rodents$angulo , dados_rodents$rodent_teutonia) R3 R4 <- rep(dados_rodents$angulo , dados_rodents$rodent_passofundo) R4 R5 <- rep(dados_rodents$angulo , dados_rodents$rodent_total) R5 #Para trabalharmos com os dados corretamente, temos que transforma-los em radianos: R1_rad <- rad(R1) R2_rad <- rad(R2) R3_rad <- rad(R3) R4_rad <- rad(R4) R5_rad <- rad(R5) #O proximo passo e transformar os dados em circular: R1.circ <- as.circular(R1_rad) R1.circ R2.circ <- as.circular(R2_rad) R3.circ <- as.circular(R3_rad) R4.circ <- as.circular(R4_rad) R5.circ <- as.circular(R5_rad) #Dessa maneira, criamos dois vetores, cujos valores dos angulos se repetem. Podemos, dessa maneira, comecar a visualizar os dados. Para isso, vamos colocar os dados no grafico: plot.circular(cbind(R1.circ, R2.circ), rotation = "clock", units = "rads") #Podemos tambem visualizar a frequencia dos angulos: rose.diag(cbind(R1.circ, R2.circ)) rose.diag(R1.circ) #Visao geral dos dados summary(R1.circ) #Distribuicao dos dados #Uniformidade dos dados ###### Rayleigh Test #### rayleigh.test(R1.circ) rayleigh.test(R2.circ) rayleigh.test(R3.circ) rayleigh.test(R4.circ) rayleigh.test(R5.circ) ##### Raos spacing test ##### #Segundo BERGIN (1991)rao.spacing.test is more flexible; able to handle more types of circular data with fewer limiting assumptions; and is more powerful with small sample sizes. rao.spacing.test(R1.circ, alpha=0)#numeric value specifying the significance level of the test. The default value is 0, in which case, a range for the p-value will be returned. Valid significance levels are 0.10, 0.05, 0.01 and 0.001. rao.spacing.test(R2.circ, alpha=0) rao.spacing.test(R3.circ, alpha=0) rao.spacing.test(R4.circ, alpha=0) rao.spacing.test(R5.circ, alpha=0) #como se apresentam os dados em um estudo cientifico? Volta apresentacao #Existem outros testes no pacote. #Comparando amostras #Duas amostras:dados parametricos(teste chamado Watson) watson.two.test(R4.circ, R3.circ) #Duas amostras:dados nao-parametricos(teste chamado Watson-Wheeler) watson.wheeler.test(list(R1.circ, R4.circ)) #Obs: nao se esqueca de listar suas variaveis (list()), caso contrario o teste nao funciona para amostras de tamanhos diferentes. #Mais de duas amostras #O teste de Watson-Williams pode ser usado tambem para mais de duas amostras,quando elas possuem distribuicao de von Mises. Nao esqueca de listar(list(x, y, z)), os vetores: watson.williams.test(list(rad(R1), rad(R2), rad(L3), rad(R4))) #O teste de Watson-Wheeler, quando as amostras nao provem de populacao com distribuicao de von Mises, tambem pode ser usado para mais de duas variaveis: watson.wheeler.test(list(R1.circ,R3.circ)) #### Grafic Clocks #### #Open your data #meia noite deve entrar como 24h nao como 00h data <-read.csv(file.choose(), header=TRUE, sep=";") #verificando o arquivo head(data) #Criando a coluna de per?odo noite/dia #essa coluna ? importante para pintar as barras de cores referentes ao periodo data$noitedia <- ifelse(data$eventhour %in% seq(6, 18),'Day','Night') head(data) #Make the plot ggplot(data, aes(x = eventhour,fill=noitedia)) +#funcao aes indica os dados que serao plotados(x), fill=preenche as barras de acordo com suas divis?es NOITE/DIA geom_hline(yintercept = c(0,2,4,6,8), colour = "gray19", size = 1.5,linetype=3) + geom_hline(yintercept = c(9), colour = "black", size = 1.5,linetype=1) + geom_vline(xintercept = c(0,6,12,18) , colour = "black", size = 1.5, lty= 1) + coord_polar(theta = 'x', start = 0, direction = 1) +#funcao para tornar o grafico circular labs(x = '', y = '') + scale_x_continuous(limits = c(0,24), breaks = c(0,6,12,18)) + scale_y_continuous(limits = c(0,10), breaks = c(0,2,4,6,8)) + theme_bw() + theme(panel.border = element_blank(), legend.key = element_blank(), axis.ticks = element_blank(), axis.text.x = element_text(face="plain", color="black", size=16,vjust=4), axis.text.y = element_blank(), panel.grid = element_blank())+ geom_histogram(breaks = seq(0,24), colour = "black")+#geom_bar= grafico de barras, colour=cor das bordas #DESIGN ggtitle("") + #title scale_fill_manual(values=c("gray30","gray30"))+#altera as corres das barras labs(fill="")+#titulo da legenda ylab("") + # axis_y title lims()+ annotate("text", label = c("","","2","","4","","6","","8"), x = 12, y=c(0,1,2,3,4,5,6,7,8), color="black", size=13, fontface="plain", hjust=1.3, vjust=1.3)#legenda das densidades ############### BIRDS #Open your data #meia noite deve entrar como 24h nao como 00h data_birds <-read.csv(file.choose(), header=TRUE, sep=";") #verificando o arquivo head(data_birds) #Criando a coluna de per?odo noite/dia #essa coluna ? importante para pintar as barras de cores referentes ao periodo data_birds$noitedia <- ifelse(data_birds$eventhour %in% seq(6, 18),'Day','Night') head(data_birds) #Make the plot ggplot(data_birds, aes(x = eventhour,fill=noitedia)) +#funcao aes indica os dados que serao plotados(x), fill=preenche as barras de acordo com suas divis?es NOITE/DIA geom_hline(yintercept = c(0,75,150,225,300), colour = "gray19", size = 1.5,linetype=3) + geom_hline(yintercept = c(350), colour = "black", size = 1.5,linetype=1) + geom_vline(xintercept = c(0,6,12,18) , colour = "black", size = 1.5, lty= 1) + coord_polar(theta = 'x', start = 0, direction = 1) +#funcao para tornar o grafico circular labs(x = '', y = '') + scale_x_continuous(limits = c(0,24), breaks = c(0,6,12,18)) + scale_y_continuous(limits = c(0,400), breaks = c(0,75,150,225,300)) + theme_bw() + theme(panel.border = element_blank(), legend.key = element_blank(), axis.ticks = element_blank(), axis.text.x = element_text(face="plain", color="black", size=16,vjust=4), axis.text.y = element_blank(), panel.grid = element_blank())+ geom_histogram(breaks = seq(0,24), colour = "black")+#geom_bar= grafico de barras, colour=cor das bordas #DESIGN ggtitle("") + #title scale_fill_manual(values=c("gray30","gray30"))+#altera as corres das barras labs(fill="")+#titulo da legenda ylab("") + # axis_y title lims()+ annotate("text", label = c("","75","150","225","300"), x = 0, y=c(0,75,150,225,300), color="black", size=13, fontface="plain", hjust=1.1, vjust=1.3)#legenda das densidades ############### small Mammals #Open your data #meia noite deve entrar como 24h nao como 00h data_small <-read.csv(file.choose(), header=TRUE, sep=";") #verificando o arquivo head(data_small) #Criando a coluna de per?odo noite/dia #essa coluna ? importante para pintar as barras de cores referentes ao periodo data_small$noitedia <- ifelse(data_small$eventhour %in% seq(6, 18),'Day','Night') head(data_small) #Make the plot ggplot(data_small, aes(x = eventhour,fill=noitedia)) +#funcao aes indica os dados que serao plotados(x), fill=preenche as barras de acordo com suas divis?es NOITE/DIA geom_hline(yintercept = c(0,10,20,30,40), colour = "gray19", size = 1.5,linetype=3) + geom_hline(yintercept = c(45), colour = "black", size = 1.5,linetype=1) + geom_vline(xintercept = c(0,6,12,18) , colour = "black", size = 1.5, lty= 1) + coord_polar(theta = 'x', start = 0, direction = 1) +#funcao para tornar o grafico circular labs(x = '', y = '') + scale_x_continuous(limits = c(0,24), breaks = c(0,6,12,18)) + scale_y_continuous(limits = c(0,50), breaks = c(0,10,20,30,40)) + theme_bw() + theme(panel.border = element_blank(), legend.key = element_blank(), axis.ticks = element_blank(), axis.text.x = element_text(face="plain", color="black", size=16,vjust=4), axis.text.y = element_blank(), panel.grid = element_blank())+ geom_histogram(breaks = seq(0,24), colour = "black")+#geom_bar= grafico de barras, colour=cor das bordas #DESIGN ggtitle("") + #title scale_fill_manual(values=c("gray30","gray30"))+#altera as corres das barras labs(fill="")+#titulo da legenda ylab("") + # axis_y title lims()+ annotate("text", label = c("","10","20","30","40"), x = 12, y=c(0,10,20,30,40), color="black", size=13, fontface="plain", hjust=1.1, vjust=1.3)#legenda das densidades #Para imagens com melhor qualidade salve em SVG e depois fa?a um print. #### OVERLAP ANALYSES #### library(overlap) #lendo dados par(mfrow=c(2,2)) margay_refugio=read.csv(file.choose(), header=TRUE, sep=";") pets_RF=read.csv(file.choose(), header=TRUE, sep=";") humans_RF=read.csv(file.choose(), header=TRUE, sep=";") cows_RF=read.csv(file.choose(), header=TRUE, sep=";") chicken_RF=read.csv(file.choose(), header=TRUE, sep=";") dogs_RF=read.csv(file.choose(), header=TRUE, sep=";") cats_RF=read.csv(file.choose(), header=TRUE, sep=";") birds_RF=read.csv(file.choose(), header=TRUE, sep=";") #Para trabalharmos com os dados corretamente, temos que transforma-los em radianos: margay_refugioRad <- margay_refugio$refugio * 2 * pi humans_RF_Rad <- humans_RF$humans * 2 * pi humans_RF_Rad pets_RF_Rad <-pets_RF$pets * 2 * pi pets_RF_Rad cows_RF_Rad <- cows_RF$cows * 2 * pi cows_RF_Rad chicken_RF_Rad <- chicken_RF$chicken * 2 * pi chicken_RF_Rad dogs_RF_Rad <- dogs_RF$dogs * 2 * pi dogs_RF_Rad cats_RF_Rad <- cats_RF$cats * 2 * pi cats_RF_Rad birds_RF_Rad <- birds_RF$birds * 2 * pi birds_RF_Rad #plotando gra?fico de densidade de kernel margay_RF_plot <- densityPlot( margay_refugioRad, rug=TRUE, main="Activity Pattern of margay in Atlantic Forest") abline(v=c(6+26/60, 19+49/60), lty=3, col= 1, lwd=1) pets_RF_plot <- densityPlot(pets_RF_Rad, rug=TRUE, col=1, lwd=2, lty= 1, col.axis="red", col.lab="red", col.main="blue")#add=TRUE adiciona o grafico por cima do overlap #sabendo o comprimento minimo dentre as amostras min(length(humans_RF_Rad), length(margay_refugioRad)) #bootstraps refugio_boots<- resample(margay_refugioRad, 1000) pets_RF_boots <- resample(pets_RF_Rad, 1000) humans_RF_boots <- resample(humans_RF_Rad, 1000) cows_RF_boots<- resample(cows_RF_Rad, 1000) chicken_RF_boots<- resample(chicken_RF_Rad, 1000) dogs_RF_boots<- resample(dogs_RF_Rad, 1000) cats_RF_boots<- resample(cats_RF_Rad, 1000) birds_RF_boots<- resample(birds_RF_Rad, 1000) ### Margay vs. Pets #### ov_margay_pets_RF <- overlapEst(margay_refugioRad, pets_RF_Rad) ov_margay_pets_RF # Analyse usando dados de boostraps: bsOut_RF_pets <- bootEst(refugio_boots, pets_RF_boots) # Demora um pouquinho bsOut_RF_pets colMeans(bsOut_RF_pets) # Convertendo columna 1 para um vector: bsOut_RF_pets_vec <- as.vector(bsOut_RF_pets[, 1]) # Observando o intervalo de confianca: CI_RF_pets<- bootCI(bsOut_RF_pets[1], bsOut_RF_pets_vec)['norm0', ] CI_RF_pets overlapTrue(refugio_boots, pets_RF_boots) par() #plotando grafico de sobreposicao overlapPlot(pets_RF_Rad, margay_refugioRad, linewidth = c(3,3), linet = c(10,1), linec = c("lightseagreen", "black"), xcenter="noon", rug=TRUE, main = "", olapcol= "grey63", font.main=2, #font.main=muda a fonte do t?tulo n.grid =100, xscale = 24, fg=1, bg=1, xlab=c("Time")) #olapcol=muda a cor da ?rea de sobreposi??o, abline(v=c(6+20/60,19+35/60), lty=2, col= "black", lwd=1) #inseri linha de nascer e p?r-do-sol #legenda legend("top", #inset=0,01, c("Pets","Margay"), lty=c(10,1), #tipo de linha da legenda col=c("lightseagreen", "black" ), #cores lwd = c(3,3), #tamanho text.font=1, # text.font=fonte 3= it?lico # cex=posi??o legenda horiz=FALSE, # horiz=legenda horizontal bg="white",bty='n', ) ##### Margay vs humans #### #análise de sobreposicao de atividade (usando densidade de kernel) ov_margay_humans_RF <- overlapEst(margay_refugioRad, humans_RF_Rad) #resultado sobreposicao ov_margay_humans_RF bsOut_RF_humans <- bootEst(refugio_boots, humans_RF_boots) # Demora um pouquinho bsOut_RF_humans colMeans(bsOut_RF_humans) # Convertendo columna 1 para um vector: bsOut_RF_humans_vec <- as.vector(bsOut_RF_humans[, 1]) # Observando o intervalo de confianca: CI_RF_humans<- bootCI(bsOut_RF_humans[1], bsOut_RF_humans_vec)['norm0', ] CI_RF_humans #plotando grafico de sobreposicao overlapPlot( humans_RF_Rad, margay_refugioRad, linewidth = c(3,3), linet = c(10,1), linec = c("#F2A428", "black"), xcenter="noon", rug=TRUE, main = "Overlap Activity Pattern", olapcol= "gray63", font.main=2, #font.main=muda a fonte do t?tulo n.grid =100, xscale = 24, fg=1, bg=1) #olapcol=muda a cor da ?rea de sobreposi??o, abline(v=c(6+20/60,19+35/60), lty=2, col= "black", lwd=1) #inseri linha de nascer e p?r-do-sol #legenda legend("top", #inset=0,01, c("Humans","Margay"), lty=c(10,1), #tipo de linha da legenda col=c("#F2A428", "black"), #cores lwd = c(3,3), #tamanho text.font=1, # text.font=fonte 3= it?lico # cex=posi??o legenda horiz=FALSE, # horiz=legenda horizontal bg="white",bty='n', ) #### Margay vs. Cows #### ov_margay_cows_RF <- overlapEst(margay_refugioRad, cows_RF_Rad) ov_margay_cows_RF bsOut_RF_cows <- bootEst(refugio_boots, cows_RF_boots) # Demora um pouquinho bsOut_RF_cows colMeans(bsOut_RF_cows) # Convertendo columna 1 para um vector: bsOut_RF_cows_vec <- as.vector(bsOut_RF_cows[, 1]) # Observando o intervalo de confianca: CI_RF_cows<- bootCI(bsOut_RF_cows[1], bsOut_RF_cows_vec)['norm0', ] CI_RF_cows #plotando grafico de sobreposicao overlapPlot(cows_RF_Rad, margay_refugioRad, linewidth = c(3,3), linet = c(10,1), linec = c("#07B56A", "black"), xcenter="noon", rug=TRUE, main = "", olapcol= "gray63", font.main=2, #font.main=muda a fonte do t?tulo n.grid =100, xscale = 24, fg=1, bg=1) #olapcol=muda a cor da ?rea de sobreposi??o, abline(v=c(6+20/60,19+35/60), lty=2, col= "black", lwd=1) #inseri linha de nascer e p?r-do-sol #legenda legend("top", #inset=0,01, c("Cows","Margay"), lty=c(10,1), #tipo de linha da legenda col=c("#07B56A", "black"), #cores lwd = c(3,3), #tamanho text.font=1, # text.font=fonte 3= it?lico # cex=posi??o legenda horiz=FALSE, # horiz=legenda horizontal bg="white",bty='n', ) #### Margay vs. Chicken #### ov_margay_chicken_RF <- overlapEst(margay_refugioRad, chicken_RF_Rad) ov_margay_chicken_RF bsOut_RF_chicken <- bootEst(refugio_boots, chicken_RF_boots) # Demora um pouquinho bsOut_RF_chicken colMeans(bsOut_RF_chicken) # Convertendo columna 1 para um vector: bsOut_RF_chicken_vec <- as.vector(bsOut_RF_chicken[, 1]) # Observando o intervalo de confianca: CI_RF_chicken<- bootCI(bsOut_RF_chicken[1], bsOut_RF_chicken_vec)['norm0', ] CI_RF_chicken #plotando grafico de sobreposicao overlapPlot( margay_refugioRad, chicken_RF_Rad, linewidth = c(2,2), linet = c(10,1), linec = c("goldenrod2", "deepskyblue4"), xcenter="midnight", rug=TRUE, main = "Overlap Activity Pattern", olapcol= "lightgrey", font.main=2, #font.main=muda a fonte do t?tulo n.grid =500, xscale = 24, fg=1, bg=1) #olapcol=muda a cor da ?rea de sobreposi??o, par(mfrow=c(1,1),mar=c(6,13,4,2)) #### Margay vs. Dogs #### ov_margay_dogs_RF <- overlapEst(margay_refugioRad, dogs_RF_Rad) ov_margay_dogs_RF bsOut_RF_dogs <- bootEst(refugio_boots, dogs_RF_boots) # Demora um pouquinho bsOut_RF_dogs colMeans(bsOut_RF_dogs) # Convertendo columna 1 para um vector: bsOut_RF_dogs_vec <- as.vector(bsOut_RF_dogs[, 1]) # Observando o intervalo de confianca: CI_RF_dogs<- bootCI(bsOut_RF_dogs[1], bsOut_RF_dogs_vec)['norm0', ] CI_RF_dogs #plotando grafico de sobreposicao overlapPlot(dogs_RF_Rad, margay_refugioRad, linewidth = c(3,3), linet = c(10,1), linec = c("lightseagreen", "black"), xcenter="noon", rug=TRUE, main = "", olapcol= "gray63", font.main=2, #font.main=muda a fonte do t?tulo n.grid =100, xscale = 24, fg=1, bg=1) #olapcol=muda a cor da ?rea de sobreposi??o, abline(v=c(6+20/60,19+35/60), lty=2, col= "black", lwd=1) #inseri linha de nascer e p?r-do-sol #legenda legend("top", #inset=0,01, c("Dogs","Margay"), lty=c(10,1), #tipo de linha da legenda col=c("lightseagreen", "black"), #cores lwd = c(3,3), #tamanho text.font=1, # text.font=fonte 3= it?lico # cex=posi??o legenda horiz=FALSE, # horiz=legenda horizontal bg="white",bty='n', ) #### Margay vs. Cats #### ov_margay_cats_RF <- overlapEst(margay_refugioRad, cats_RF_Rad) ov_margay_cats_RF bsOut_RF_cats <- bootEst(refugio_boots, cats_RF_boots) # Demora um pouquinho bsOut_RF_cats colMeans(bsOut_RF_cats) # Convertendo columna 1 para um vector: bsOut_RF_cats_vec <- as.vector(bsOut_RF_cats[, 1]) # Observando o intervalo de confianca: CI_RF_cats<- bootCI(bsOut_RF_cats[1], bsOut_RF_cats_vec)['norm0', ] CI_RF_cats #plotando grafico de sobreposicao overlapPlot(cats_RF_Rad, margay_refugioRad, linewidth = c(3,3), linet = c(10,1), linec = c("#4605E8", "black"), xcenter="noon", rug=TRUE, main = "", olapcol= "gray63", font.main=2, #font.main=muda a fonte do t?tulo n.grid =100, xscale = 24, fg=1, bg=1) #olapcol=muda a cor da ?rea de sobreposi??o, abline(v=c(6+20/60,19+35/60), lty=2, col= "black", lwd=1) #inseri linha de nascer e p?r-do-sol #legenda legend("top", #inset=0,01, c("Cats","Margay"), lty=c(10,1), #tipo de linha da legenda col=c("#4605E8", "black"), #cores lwd = c(3,3), #tamanho text.font=1, # text.font=fonte 3= it?lico # cex=posi??o legenda horiz=FALSE, # horiz=legenda horizontal bg="white",bty='n', ) #### Margay vs. birds #### ov_margay_birds_RF <- overlapEst(margay_refugioRad, birds_RF_Rad) ov_margay_birds_RF bsOut_RF_birds <- bootEst(refugio_boots, birds_RF_boots) # Demora um pouquinho bsOut_RF_birds colMeans(bsOut_RF_birds) # Convertendo columna 1 para um vector: bsOut_RF_birds_vec <- as.vector(bsOut_RF_birds[, 1]) # Observando o intervalo de confianca: CI_RF_birds<- bootCI(bsOut_RF_birds[1], bsOut_RF_birds_vec)['norm0', ] CI_RF_birds par(mfrow=c(1,1),mar=c(6,14,4,2)) # salvar em tamanho 600x427 #plotando grafico de sobreposicao overlapPlot(birds_RF_Rad, margay_refugioRad, linewidth = c(3,3), linet = c(10,1), linec = c("#D94929", "black"), #F24405,#4605E8 xcenter="noon", rug=TRUE, main = "", olapcol= "gray63", font.main=2, #font.main=muda a fonte do t?tulo n.grid =100, xscale = 24, fg=1, bg=1) #olapcol=muda a cor da ?rea de sobreposi??o, abline(v=c(6+20/60,19+35/60), lty=2, col= "black", lwd=1) #inseri linha de nascer e p?r-do-sol #legenda legend("top", #inset=0,01, c("Small birds","Margay"), lty=c(10,1), #tipo de linha da legenda col=c("#D94929", "black"), #cores lwd = c(3,3), #tamanho text.font=1, # text.font=fonte 3= it?lico # cex=posi??o legenda horiz=FALSE, # horiz=legenda horizontal bg="white",bty='n', ) par(mfrow=c(2,2)) margay_teutonia=read.csv(file.choose(), header=TRUE, sep=";") pets_TT=read.csv(file.choose(), header=TRUE, sep=";") humans_TT=read.csv(file.choose(), header=TRUE, sep=";") cows_TT=read.csv(file.choose(), header=TRUE, sep=";") chicken_TT=read.csv(file.choose(), header=TRUE, sep=";") dogs_TT=read.csv(file.choose(), header=TRUE, sep=";") cats_TT=read.csv(file.choose(), header=TRUE, sep=";") birds_TT=read.csv(file.choose(), header=TRUE, sep=";") #Para trabalharmos com os dados corretamente, temos que transforma-los em radianos: margay_teutoniaRad <- margay_teutonia$teutonia * 2 * pi humans_TT_Rad <- humans_TT$humans * 2 * pi humans_TT_Rad pets_TT_Rad <-pets_TT$pets * 2 * pi pets_TT_Rad cows_TT_Rad <- cows_TT$cows * 2 * pi cows_TT_Rad chicken_TT_Rad <- chicken_TT$chicken * 2 * pi chicken_TT_Rad dogs_TT_Rad <- dogs_TT$dogs * 2 * pi dogs_TT_Rad cats_TT_Rad <- cats_TT$cats * 2 * pi cats_TT_Rad birds_TT_Rad <- birds_TT$birds * 2 * pi birds_TT_Rad #plotando gra?fico de densidade de kernel margay_TT_plot <- densityPlot( margay_teutoniaRad, rug=TRUE, main="Activity Pattern of margay in Atlantic Forest") abline(v=c(6+26/60, 19+49/60), lty=3, col= 1, lwd=1) pets_TT_plot <- densityPlot(pets_TT_Rad, rug=TRUE, col=1, lwd=2, lty= 1, col.axis="red", col.lab="red", col.main="blue")#add=TRUE adiciona o grafico por cima do overlap #sabendo o comprimento minimo dentre as amostras min(length(humans_TT_Rad), length(margay_teutoniaRad)) #bootstraps teutonia_boots<- resample(margay_teutoniaRad, 1000) pets_TT_boots <- resample(pets_TT_Rad, 1000) humans_TT_boots <- resample(humans_TT_Rad, 1000) cows_TT_boots<- resample(cows_TT_Rad, 1000) chicken_TT_boots<- resample(chicken_TT_Rad, 1000) dogs_TT_boots<- resample(dogs_TT_Rad, 1000) cats_TT_boots<- resample(cats_TT_Rad, 1000) birds_TT_boots<- resample(birds_TT_Rad, 1000) ### Margay vs. Pets #### ov_margay_pets_TT <- overlapEst(margay_teutoniaRad, pets_TT_Rad) ov_margay_pets_TT # Analyse usando dados de boostraps: bsOut_TT_pets <- bootEst(teutonia_boots, pets_TT_boots) # Demora um pouquinho bsOut_TT_pets colMeans(bsOut_TT_pets) # Convertendo columna 1 para um vector: bsOut_TT_pets_vec <- as.vector(bsOut_TT_pets[, 1]) # Observando o intervalo de confianca: CI_TT_pets<- bootCI(bsOut_TT_pets[1], bsOut_TT_pets_vec)['norm0', ] CI_TT_pets overlapTrue(teutonia_boots, pets_TT_boots) par() #plotando grafico de sobreposicao overlapPlot(pets_TT_Rad, margay_teutoniaRad, linewidth = c(3,3), linet = c(10,1), linec = c("lightseagreen", "black"), xcenter="noon", rug=TRUE, main = "", olapcol= "gray63", font.main=2, #font.main=muda a fonte do t?tulo n.grid =100, xscale = 24, fg=1, bg=1, xlab=c("Time")) #olapcol=muda a cor da ?rea de sobreposi??o, par(mfrow=c(1,1),mar=c(6,10,4,2)) abline(v=c(6+06/60,19+26/60), lty=2, col= "black", lwd=1) #inseri linha de nascer e p?r-do-sol #legenda legend("top", #inset=0,01, c("Pets","Margay"), lty=c(10,1), #tipo de linha da legenda col=c("lightseagreen", "black"), #cores lwd = c(3,3), #tamanho text.font=1, # text.font=fonte 3= it?lico # cex=posi??o legenda horiz=FALSE, # horiz=legenda horizontal bg="white",bty='n', ) ##### Margay vs humans #### #análise de sobreposicao de atividade (usando densidade de kernel) ov_margay_humans_TT <- overlapEst(margay_teutoniaRad, humans_TT_Rad) #resultado sobreposicao ov_margay_humans_TT bsOut_TT_humans <- bootEst(teutonia_boots, humans_TT_boots) # Demora um pouquinho bsOut_TT_humans colMeans(bsOut_TT_humans) # Convertendo columna 1 para um vector: bsOut_TT_humans_vec <- as.vector(bsOut_TT_humans[, 1]) # Observando o intervalo de confianca: CI_TT_humans<- bootCI(bsOut_TT_humans[1], bsOut_TT_humans_vec)['norm0', ] CI_TT_humans #plotando grafico de sobreposicao overlapPlot(humans_TT_Rad, margay_teutoniaRad, linewidth = c(3,3), linet = c(10,1), linec = c("white", "black"), xcenter="noon", rug=TRUE, main = "Overlap Activity Pattern", olapcol= "grey63", font.main=2, #font.main=muda a fonte do t?tulo n.grid =100, xscale = 24, fg=1, bg=1) #olapcol=muda a cor da ?rea de sobreposi??o, par(mfrow=c(1,1),mar=c(6,10,4,2)) abline(v=c(6+06/60,19+26/60), lty=2, col= "black", lwd=1) #inseri linha de nascer e p?r-do-sol #legenda legend("top", #inset=0,01, c("Humans","Margay"), lty=c(10,1), #tipo de linha da legenda col=c("#F2A428", "black"), #cores lwd = c(3,3), #tamanho text.font=1, # text.font=fonte 3= it?lico # cex=posi??o legenda horiz=FALSE, # horiz=legenda horizontal bg="white",bty='n', ) #### Margay vs. Cows #### ov_margay_cows_TT <- overlapEst(margay_teutoniaRad, cows_TT_Rad) ov_margay_cows_TT bsOut_TT_cows <- bootEst(teutonia_boots, cows_TT_boots) # Demora um pouquinho bsOut_TT_cows colMeans(bsOut_TT_cows) # Convertendo columna 1 para um vector: bsOut_TT_cows_vec <- as.vector(bsOut_TT_cows[, 1]) # Observando o intervalo de confianca: CI_TT_cows<- bootCI(bsOut_TT_cows[1], bsOut_TT_cows_vec)['norm0', ] CI_TT_cows #plotando grafico de sobreposicao overlapPlot( cows_TT_Rad, margay_teutoniaRad, linewidth = c(3,3), linet = c(10,1), linec = c("white", "black"), xcenter="noon", rug=TRUE, main = "", olapcol= "grey63", font.main=2, #font.main=muda a fonte do t?tulo n.grid =100, xscale = 24, fg=1, bg=1) #olapcol=muda a cor da ?rea de sobreposi??o, abline(v=c(6+06/60,19+26/60), lty=2, col= "black", lwd=1) #inseri linha de nascer e p?r-do-sol #legenda legend("top", #inset=0,01, c("Cows","Margay"), lty=c(10,1), #tipo de linha da legenda col=c("#07B56A", "black"), #cores lwd = c(3,3), #tamanho text.font=1, # text.font=fonte 3= it?lico # cex=posi??o legenda horiz=FALSE, # horiz=legenda horizontal bg="white",bty='n', ) #### Margay vs. Chicken #### ov_margay_chicken_TT <- overlapEst(margay_teutoniaRad, chicken_TT_Rad) ov_margay_chicken_TT bsOut_TT_chicken <- bootEst(teutonia_boots, chicken_TT) # Demora um pouquinho bsOut_RF_chicken colMeans(bsOut_RF_chicken) # Convertendo columna 1 para um vector: bsOut_TT_chicken_vec <- as.vector(bsOut_RF_chicken[, 1]) # Observando o intervalo de confianca: CI_TT_chicken<- bootCI(bsOut_RF_chicken[1], bsOut_RF_chicken_vec)['norm0', ] CI_TT_chicken #plotando grafico de sobreposicao overlapPlot( margay_teutoniaRad, chicken_TT_Rad, linewidth = c(2,2), linet = c(10,1), linec = c("goldenrod2", "deepskyblue4"), xcenter="noon", rug=TRUE, main = "Overlap Activity Pattern", olapcol= "lightgrey", font.main=2, #font.main=muda a fonte do t?tulo n.grid =500, xscale = 24, fg=1, bg=1) #olapcol=muda a cor da ?rea de sobreposi??o, abline(v=c(6+06/60,19+26/60), lty=2, col= "black", lwd=1) #inseri linha de nascer e p?r-do-sol #legenda legend("top", #inset=0,01, c("margay","chicken"), lty=c(10,1), #tipo de linha da legenda col=c("goldenrod2","deepskyblue4"), #cores lwd = c(2,2), #tamanho text.font=3, # text.font=fonte 3= it?lico # cex=posi??o legenda horiz=FALSE, # horiz=legenda horizontal bg="white",bty='n', ) #### Margay vs. Dogs #### ov_margay_dogs_TT <- overlapEst(margay_teutoniaRad, dogs_TT_Rad) ov_margay_dogs_TT bsOut_TT_dogs <- bootEst(teutonia_boots, dogs_TT_boots) # Demora um pouquinho bsOut_TT_dogs colMeans(bsOut_TT_dogs) # Convertendo columna 1 para um vector: bsOut_TT_dogs_vec <- as.vector(bsOut_TT_dogs[, 1]) # Observando o intervalo de confianca: CI_TT_dogs<- bootCI(bsOut_TT_dogs[1], bsOut_TT_dogs_vec)['norm0', ] CI_TT_dogs #plotando grafico de sobreposicao overlapPlot(dogs_TT_Rad, margay_teutoniaRad, linewidth = c(3,3), linet = c(10,1), linec = c("lightseagreen", "black"), xcenter="noon", rug=TRUE, main = "", olapcol= "gray63", font.main=2, #font.main=muda a fonte do t?tulo n.grid =100, xscale = 24, fg=1, bg=1) #olapcol=muda a cor da ?rea de sobreposi??o, abline(v=c(6+06/60,19+26/60), lty=2, col= "black", lwd=1) #inseri linha de nascer e p?r-do-sol #legenda legend("topright", #inset=0,01, c("Dogs","Margay"), lty=c(10,1), #tipo de linha da legenda col=c("#07B56A", "black"), #cores lwd = c(3,3), #tamanho text.font=1, # text.font=fonte 3= it?lico # cex=posi??o legenda horiz=FALSE, # horiz=legenda horizontal bg="white",bty='n', ) #### Margay vs. Cats #### ov_margay_cats_TT <- overlapEst(margay_teutoniaRad, cats_TT_Rad) ov_margay_cats_TT bsOut_TT_cats <- bootEst(teutonia_boots, cats_TT_boots) # Demora um pouquinho bsOut_TT_cats colMeans(bsOut_TT_cats) # Convertendo columna 1 para um vector: bsOut_TT_cats_vec <- as.vector(bsOut_TT_cats[, 1]) # Observando o intervalo de confianca: CI_TT_cats<- bootCI(bsOut_TT_cats[1], bsOut_TT_cats_vec)['norm0', ] CI_TT_cats #plotando grafico de sobreposicao overlapPlot(cats_TT_Rad, margay_teutoniaRad, linewidth = c(3,3), linet = c(10,1), linec = c("#4605E8", "black"), xcenter="noon", rug=TRUE, main = "", olapcol= "gray63", font.main=2, #font.main=muda a fonte do t?tulo n.grid =100, xscale = 24, fg=1, bg=1) #olapcol=muda a cor da ?rea de sobreposi??o, abline(v=c(6+06/60,19+26/60), lty=2, col= "black", lwd=1) #inseri linha de nascer e p?r-do-sol #legenda legend("top", #inset=0,01, c("Cats","Margay"), lty=c(10,1), #tipo de linha da legenda col=c("#4605E8", "black"), #cores lwd = c(3,3), #tamanho text.font=1, # text.font=fonte 3= it?lico # cex=posi??o legenda horiz=FALSE, # horiz=legenda horizontal bg="white",bty='n', ) #### Margay vs. birds #### ov_margay_birds_TT <- overlapEst(margay_teutoniaRad, birds_TT_Rad) ov_margay_birds_TT bsOut_TT_birds <- bootEst(teutonia_boots, birds_TT_boots) # Demora um pouquinho bsOut_TT_birds colMeans(bsOut_TT_birds) # Convertendo columna 1 para um vector: bsOut_TT_birds_vec <- as.vector(bsOut_TT_birds[, 1]) # Observando o intervalo de confianca: CI_TT_birds<- bootCI(bsOut_TT_birds[1], bsOut_TT_birds_vec)['norm0', ] CI_TT_birds par(mfrow=c(1,1),mar=c(6,14,4,2)) # salvar em tamanho 600x427 #plotando grafico de sobreposicao overlapPlot(birds_TT_Rad, margay_teutoniaRad, linewidth = c(3,3), linet = c(10,1), linec = c("#D94929", "black"), xcenter="noon", rug=TRUE, main = "", olapcol= "gray63", font.main=2, #font.main=muda a fonte do t?tulo n.grid =100, xscale = 24, fg=1, bg=1) #olapcol=muda a cor da ?rea de sobreposi??o, abline(v=c(6+06/60,19+26/60), lty=2, col= "black", lwd=1) #inseri linha de nascer e p?r-do-sol #legenda legend("topright", #inset=0,01, c("Small birds","Margay"), lty=c(10,1), #tipo de linha da legenda col=c("#D94929", "black"), #cores lwd = c(3,3), #tamanho text.font=1, # text.font=fonte 3= it?lico # cex=posi??o legenda horiz=FALSE, # horiz=legenda horizontal bg="white",bty='n', ) ## Make the same with the others areas