Course Notes: Spatial Statistics in R

PLAN

1 Total - map

2a Trees per km2 - map and table

2b Most common trees - chart

3 Top 10 trees vs nta - chart

4 Curb location

5 Health - All trees

6 Health - 10 most common trees

8 Health - map of nta (add legend)

9 Trunk

10 Recommended trees - table with ratings

11 Recommended trees - map

12 Recommended trees - details table

12 Appendix - dead trees

#install.packages('ggmap') #install.packages('sf') #install.packages('cowplot') #install.packages('jpeg') #install.packages('hrbrthemes') #install.packages('formattable') #webshot::install_phantomjs() #install.packages('kableExtra') library(kableExtra) library(ggmap) library(tidyverse) library(sf) library(cowplot) library(gridExtra) library(grid) library(RColorBrewer) library(jpeg) library(hrbrthemes) library(formattable) library(webshot)

Which tree species should the city plant?

Background

You work for a nonprofit organization advising the planning department on ways to

improve the quantity and quality of trees in New York City.

The urban design team believes tree size (using trunk diameter as a proxy for size)

and health are the most desirable characteristics of city trees.

The city would like to learn more about which tree species are

the best choice to plant on the streets of Manhattan.

Tree Census

- "tree_id" - Unique id of each tree.

- "tree_dbh" - The diameter of the tree in inches measured at 54 inches above the ground.

- "curb_loc" - Location of the tree bed in relation to the curb. Either along the curb (OnCurb) or offset from the curb (OffsetFromCurb).

- "spc_common" - Common name for the species.

- "status" - Indicates whether the tree is alive or standing dead.

- "health" - Indication of the tree's health (Good, Fair, and Poor).

- "root_stone" - Indicates the presence of a root problem caused by paving stones in the tree bed.

- "root_grate" - Indicates the presence of a root problem caused by metal grates in the tree bed.

- "root_other" - Indicates the presence of other root problems.

- "trunk_wire" - Indicates the presence of a trunk problem caused by wires or rope wrapped around the trunk.

- "trnk_light" - Indicates the presence of a trunk problem caused by lighting installed on the tree.

- "trnk_other" - Indicates the presence of other trunk problems.

- "brch_light" - Indicates the presence of a branch problem caused by lights or wires in the branches.

- "brch_shoe" - Indicates the presence of a branch problem caused by shoes in the branches.

- "brch_other" - Indicates the presence of other branch problems.

- "postcode" - Five-digit zip code where the tree is located.

- "nta" - Neighborhood Tabulation Area (NTA) code from the 2010 US Census for the tree.

- "nta_name" - Neighborhood name.

- "latitude" - Latitude of the tree, in decimal degrees.

- "longitude" - Longitude of the tree, in decimal degrees.

Neighborhoods' geographical information

- "ntacode" - NTA code (matches Tree Census information).

- "ntaname" - Neighborhood name (matches Tree Census information).

- "geometry" - Polygon that defines the neighborhood.

Challenge

Create a report that covers the following:

* What are the most common tree species in Manhattan?

* Which are the neighborhoods with the most trees?

* A visualization of Manhattan's neighborhoods and tree locations.

* What ten tree species would you recommend the city plant in the future?

0.) DATA

trees <- as.data.frame(read_csv('trees1.csv')) trees <- trees %>% mutate(spc_common = str_to_title(spc_common))

str_to_title() converts to title case, where only the first letter of each word is capitalized.

neighborhoods <- st_read("nta.geojson")

1.) Number of trees - map + background

1a.) Boroughs

neighborhoodsBoroughs <- neighborhoods %>% group_by(boroname) %>% summarise(abc = sum(as.numeric(shape_leng)))

neighborhoodsManhattan <- neighborhoods %>% filter(boroname == 'Manhattan')

centroids = st_centroid(neighborhoodsBoroughs) #st_centroid() is used to extract the centroid coordinates of each polygon feature and #returns a point geometry. Note that the centroid is not always inside of the polygon that #it is the center of (for example, a “C-shaped” island or a doughnut). st_point_on_surface() #provides an alternative in which a point within the polygon will be returned.

centroids <- centroids %>% mutate(geometry1 = gsub('[()°c\]', '', geometry)) %>% separate(col = geometry1, into = c('Latitude', 'Longitude'), sep = '\, ') %>% mutate(Latitude = as.numeric(Latitude), Longitude = as.numeric(Longitude)) %>% as.data.frame()

transformsa point geometry into 2 separate columns latitude and longitude

plotBoroughs <- ggplot() + geom_sf(data = neighborhoodsBoroughs, #aes(fill = as.numeric(abc)), -> fills polygons with colors depending on abc #lwd = 0, colour = "white" ) + geom_sf(data = subset(neighborhoodsBoroughs, boroname == 'Manhattan'), fill = brewer.pal(n = 9, name = 'Greens')[5]) + geom_sf(data = subset(neighborhoodsBoroughs, boroname != 'Manhattan'), fill = "#FCB97D") + theme_void() + theme(legend.position = "none") + #scale_fill_gradientn(

colors = c("lightgray")) +

#geom_point(data = centroids, aes(x = Latitude, y = Longitude)) + geom_text(label = centroidsLatitude, y = centroids$Longitude), size = 3) + geom_rect(mapping = aes( xmin = st_bbox(neighborhoodsManhattan)[1] - 0.01, ymin = st_bbox(neighborhoodsManhattan)[2] - 0.01, xmax = st_bbox(neighborhoodsManhattan)[3] + 0.01, ymax = st_bbox(neighborhoodsManhattan)[4] + 0.01), fill = NA, colour = "black", size = 0.9 )

plotBoroughs

1b.) Manhattan

neighborhoodsManhattan <- neighborhoods %>% filter(boroname == 'Manhattan')

neighborhoodsManhattanExclParks <- neighborhoodsManhattan %>% filter(ntacode != 'MN99')

centroidsManhattan = st_centroid(neighborhoodsManhattanExclParks)

centroidsManhattan <- centroidsManhattan %>% mutate(geometry1 = gsub('[()°c\]', '', geometry)) %>% separate(col = geometry1, into = c('Latitude', 'Longitude'), sep = '\, ') %>% mutate(Latitude = as.numeric(Latitude), Longitude = as.numeric(Longitude)) %>% as.data.frame()

centroidsManhattanntaname) centroidsManhattanname)

plotManhattan <- ggplot() + geom_sf(data = neighborhoodsManhattanExclParks, aes(fill = as.numeric(shape_area)), colour = "white" ) + theme_void() + theme(legend.position = "none") + scale_fill_gradientn( colors = c("#FF5E5B")) + #geom_point(data = centroidsManhattan, aes(x = Latitude, y = Longitude)) + geom_text(label = centroidsManhattanLatitude, y = centroidsManhattan$Longitude), size = 1.8)

#plotManhattan

1c.) Manhattan trees density

#neighborhoodsManhattan <- neighborhoods %>%

filter(boroname == 'Manhattan')

#neighborhoodsManhattanExclParks <- neighborhoodsManhattan %>%

filter(ntacode != 'MN99')

centroidsManhattan = st_centroid(neighborhoodsManhattanExclParks)

#centroidsManhattan <- centroidsManhattan %>%

mutate(geometry1 = gsub('[()°c\]', '', geometry)) %>%

separate(col = geometry1, into = c('Latitude', 'Longitude'), sep = '\, ') %>%

mutate(Latitude = as.numeric(Latitude), Longitude = as.numeric(Longitude)) %>%

as.data.frame()

#centroidsManhattanntaname) #centroidsManhattanname)

treesSP <- st_as_sf(trees, coords = c('longitude','latitude'), crs = st_crs(neighborhoodsManhattan))

st_as_sf - convert foreign object to an sf object reference system specified by st_crs(...)

#treesSPTest <- head(treesSP, n = 100, crs = st_crs(neighborhoodsManhattan))

st_crs(neighborhoodsManhattan) == st_crs(treesSP)

neighborhoodsManhattanExclParks$area_sqkm <- st_area(neighborhoodsManhattanExclParks)

neighborhoodsManhattanAreas <- neighborhoodsManhattanExclParks%>% select(ntacode, area_sqkm) %>% st_drop_geometry() %>% mutate(area_sqkm = as.numeric(area_sqkm)/10^6)

manhattanTrees <- trees %>% select(nta,nta_name) %>% group_by(nta, nta_name) %>% summarise(n = n()) %>% ungroup() %>% left_join(neighborhoodsManhattanAreas, by = c('nta' = 'ntacode')) %>% mutate(treesPerAreaUnit = n / area_sqkm) %>% as.data.frame()

customGreen = brewer.pal(n = 9 , name = "Greens")[1] customGreen0 = brewer.pal(n = 9 , name = "Greens")[8] customBlue = brewer.pal(n = 9 , name = "Blues")[1] customBlue0 = brewer.pal(n = 9 , name = "Blues")[6]

neighborhoodsManhattanExclParksTrees <- neighborhoodsManhattanExclParks %>% left_join(manhattanTrees, by = c('ntacode'='nta'))

neighborhoodsManhattanExclParksTreesPerKmPlot <- ggplot() + geom_sf(data = neighborhoodsManhattanExclParksTrees, aes(fill = n), colour = "darkgrey" ) + theme_void() + theme(legend.position = "none") + theme(#legend.position="left", legend.key.size = unit(0.5, 'cm'), legend.position = c(-0.05, 0.3), legend.title = element_text(size=7), legend.text = element_text(size=7)) + guides(fill=guide_legend(title="number of trees")) + scale_fill_gradientn(colors = brewer.pal(n = 9 , name = "Greens")[1:8]) + geom_text(label = centroidsManhattanLatitude, y = centroidsManhattan$Longitude), size = 1.8)

neighborhoodsManhattanExclParksTreesPerKmPlot

1d.) GRID

img1 <- readJPEG("pic2.JPG") img1 = abind::abind(img1, img1[,,1]) img1[,,4] = 0.15

h<-dim(img1)[1] # image height w<-dim(img1)[2] # image width

plotBackground <- plot(ggplot() + annotation_custom(grid::rasterGrob(img1, width=unit(1,"npc"), height=unit(1,"npc")), 0, w, 0, -h) + # The minus is needed to get the y scale reversed scale_x_continuous(expand=c(0,0),limits=c(0,w)) + scale_y_reverse(expand=c(0,0),limits=c(h,0)) + # The y scale is reversed because in image the vertical positive direction is typically downward coord_equal() + # To keep the aspect ratio of the image. stat_bin2d(binwidth=2,aes(fill = ..density..))+ theme_void() )

plotBackgroundWithMap <- ggdraw() + draw_plot(plotBackground) + draw_plot(neighborhoodsManhattanExclParksTreesPerKmPlot, hjust =0.075, width = 1, height = 1) + #draw_plot(plotManhattan, hjust =0.1, width = 1, height = 1) + draw_plot(plotBoroughs, hjust = -0.7, vjust = -0.2, width = 0.5, height = 0.5)

plotBackgroundWithMap

##################################################

2) Trees per km2

2a.) Trees per km2 - table

improvement_formatter <- formatter("span", style = x ~ style(font.weight = "bold", display = 'block', width = '100px', color = ifelse(x > 0, 'Green', ifelse(x < 0, 'Red',"black"))), x ~ icontext(ifelse(x>0, "arrow-up", "arrow-down"), x))

customGreen = brewer.pal(n = 9 , name = "Greens")[1] customGreen0 = brewer.pal(n = 9 , name = "Greens")[8] customBlue = brewer.pal(n = 9 , name = "Blues")[1] customBlue0 = brewer.pal(n = 9 , name = "Blues")[6]

costum_format1 <- formatter(.tag = "span", style = function(x) style( display = "block", padding = "0 4px", width = '150px', 'border-radius' = "4px", 'background-color' = csscolor(gradient(as.numeric(x),customGreen, customGreen0)), color = ifelse(x >= 3, "black","black")))

costum_format2 <- formatter(.tag = "span", style = function(x) style( display = "block", width = "50px"))

costum_format3 <- formatter(.tag = "span", style = function(x) style( display = "block", padding = "0 4px", width = '50px', 'border-radius' = "4px"))

manhattanTreesTable <- manhattanTrees %>% mutate(treesPerKm2 = round(treesPerAreaUnit)) %>% select(nta, nta_name, treesPerKm2) %>% arrange(desc(treesPerKm2))

manhattanTreesFormattable = formattable(manhattanTreesTable, align =c("l","l","l"), list('nta' = costum_format3, 'ntaname' = costum_format2, treesPerKm2 = costum_format1 ))

export_formattable <- function(f, file, width = "60%", height = NULL, background = "white", delay = 0.2) { w <- as.htmlwidget(f, width = width, height = height) path <- htmltools::html_print(w, background = background, viewer = NULL) url <- paste0("file:///", gsub("\\", "/", normalizePath(path))) webshot(url, file = file, selector = ".formattable_widget", delay = delay) }

export_formattable(manhattanTreesFormattable, 'manhattanTreesFormattable.jpg')

img2 <- readJPEG('manhattanTreesFormattable.jpg') img2 = abind::abind(img2, img2[,,1]) img2[,,4] = 1

h <-dim(img2)[1] # image height w <-dim(img2)[2] # image width

manhattanTreesFormattablePlot <- ggplot() + annotation_custom(grid::rasterGrob(img2, width=unit(1,"npc"), height=unit(1,"npc")), 0, w, 0, -h) + scale_x_continuous(expand=c(0,0),limits=c(0,w)) + scale_y_reverse(expand=c(0,0),limits=c(h,0)) + coord_equal() + stat_bin2d(binwidth=2,aes(fill = ..density..)) + theme_void()

#manhattanTreesFormattablePlot

2b.) Trees per km2 - plot

neighborhoodsManhattan <- neighborhoods %>% filter(boroname == 'Manhattan')

neighborhoodsManhattanExclParks <- neighborhoodsManhattan %>% filter(ntacode != 'MN99')

centroidsManhattan = st_centroid(neighborhoodsManhattanExclParks)

centroidsManhattan <- centroidsManhattan %>% mutate(geometry1 = gsub('[()°c\]', '', geometry)) %>% separate(col = geometry1, into = c('Latitude', 'Longitude'), sep = '\, ') %>% mutate(Latitude = as.numeric(Latitude), Longitude = as.numeric(Longitude)) %>% as.data.frame()

centroidsManhattanntaname) centroidsManhattanname)

treesSP <- st_as_sf(trees, coords = c('longitude','latitude'), crs = st_crs(neighborhoodsManhattan)) treesSPTest <- head(treesSP, n = 100, crs = st_crs(neighborhoodsManhattan))

st_crs(neighborhoodsManhattan) == st_crs(treesSP)

neighborhoodsManhattanExclParks$area_sqkm <- st_area(neighborhoodsManhattanExclParks)

neighborhoodsManhattanAreas <- neighborhoodsManhattanExclParks%>% select(ntacode, area_sqkm) %>% st_drop_geometry() %>% mutate(area_sqkm = as.numeric(area_sqkm)/10^6)

manhattanTrees <- trees %>% select(nta,nta_name) %>% group_by(nta, nta_name) %>% summarise(n = n()) %>% ungroup() %>% left_join(neighborhoodsManhattanAreas, by = c('nta' = 'ntacode')) %>% mutate(treesPerAreaUnit = n / area_sqkm) %>% as.data.frame()

customGreen = brewer.pal(n = 9 , name = "Greens")[1] customGreen0 = brewer.pal(n = 9 , name = "Greens")[8] customBlue = brewer.pal(n = 9 , name = "Blues")[1] customBlue0 = brewer.pal(n = 9 , name = "Blues")[6]

neighborhoodsManhattanExclParksTrees <- neighborhoodsManhattanExclParks %>% left_join(manhattanTrees, by = c('ntacode'='nta'))

neighborhoodsManhattanExclParksTreesPerKmPlot <- ggplot() + geom_sf(data = neighborhoodsManhattanExclParksTrees, aes(fill = treesPerAreaUnit), colour = "darkgrey" ) + theme_void() + theme(legend.position = "none") + theme(#legend.position="left", legend.key.size = unit(0.5, 'cm'), legend.position = c(-0.05, 0.3), legend.title = element_text(size=7), legend.text = element_text(size=7)) + guides(fill=guide_legend(title="trees per km2")) + scale_fill_gradientn(colors = brewer.pal(n = 9 , name = "Greens")[1:8]) + geom_text(label = centroidsManhattanLatitude, y = centroidsManhattan$Longitude), size = 1.8)

2c.) Trees per km2 - table + plot

grid.arrange(neighborhoodsManhattanExclParksTreesPerKmPlot, manhattanTreesFormattablePlot, ncol = 2)

##################################################

3.) Most common trees

3a.) Most common trees - chart

treesTop20 <- trees %>% group_by(spc_common) %>% summarise(n = n()) %>% arrange(desc(n)) %>% mutate(nPerc = round(n/sum(n)*100,2), cumSum = cumsum(n), cumSumPerc = round(cumSum/sum(n)*100,2)) %>% ungroup() %>% head(n = 21) %>% drop_na() %>% as.data.frame()

treesTop20 <- treesTop20 %>% ggplot() + geom_col(aes(x = reorder(spc_common,-nPerc), y = nPerc), fill="#69b3a2", alpha= .4) + geom_point(aes(x = reorder(spc_common,-nPerc), y = cumSumPerc/3.03), size = 2, color = rgb(0.2, 0.6, 0.9, 1)) + scale_y_continuous(limits = c(0,30), sec.axis = sec_axis(~.*3.3, name="cum % of total trees")) + theme_ipsum() + ylab('% of total trees') + ggtitle('Fig_2: Most common Manhattan trees species') + theme( plot.title=element_text(size = 10, face = 'plain', hjust=0.5, color = '#576161'), axis.text.x = element_text(color = '#576161', angle = 90, vjust = 0.5, hjust=1, size = 10), #axis.title.x = element_text(color = "#576161"), axis.title.x = element_blank(), axis.title.y = element_text(color = "#69b3a2", size=12), axis.text.y = element_text(color = "#69b3a2"), panel.grid.major = element_line(colour = "aliceblue"), panel.grid.minor = element_line(colour = "aliceblue"), axis.title.y.right = element_text(color = rgb(0.2, 0.6, 0.9, 1), size=12), axis.text.y.right = element_text(color = rgb(0.2, 0.6, 0.9, 1)) ) + annotate("curve", x = 15, y = 12.6, xend = 8, yend = 24, arrow = arrow(length = unit(0.2, "cm"), type = "closed"), color = rgb(0.2, 0.6, 0.9, 1)) + annotate("text", x = 15, y = 12.6, label = "10 most common trees make \n up 80% of all trees ", vjust = 1, size = 3, color = rgb(0.2, 0.6, 0.9, 1)) + annotate("curve", x = 3, y = 24, xend = 1, yend = 19, arrow = arrow(length = unit(0.2, "cm"), type = "closed"), color = '#69b3a2') + annotate("text", x = 4, y = 28, label = "Honeylocust being the most common tree \n makes up 21% of all trees", vjust = 1, size = 3, color = '#69b3a2')

#treesTop20

3b.) Most common trees - distribution

treesTop10SpeciesVector <- trees %>% group_by(spc_common) %>% summarise(n = n()) %>% arrange(desc(n)) %>% mutate(nPerc = round(n/sum(n)*100,2), cumSum = cumsum(n), cumSumPerc = round(cumSum/sum(n)*100,2)) %>% ungroup() %>% head(n = 11) %>% drop_na() %>% as.data.frame()

treesTop10Species <- trees %>% filter(spc_common %in% c(treesTop10SpeciesVector$spc_common)) %>% left_join(neighborhoods, by = c('nta' = 'ntacode')) %>% left_join(treesTop10SpeciesVector, by = c('spc_common' = 'spc_common')) %>% select(spc_common, ntaname, boroname, n) %>% filter(boroname == 'Manhattan') %>% #head() %>% ggplot(aes(x = reorder(spc_common,n), y = ntaname)) + geom_jitter(alpha=0.1, size = 0.5, width = 0.10, height = 0.20) + theme_minimal() + theme(axis.text.x = element_text(angle = 90, vjust = 1.1, hjust=1, size = 8), axis.text.y = element_text(size = 8) )

mostCommonTreesDist <- trees %>% filter(spc_common %in% c(treesTop10SpeciesVector$spc_common)) %>% left_join(neighborhoods, by = c('nta' = 'ntacode')) %>% select(spc_common, ntaname, boroname) %>% group_by(spc_common, ntaname) %>% summarize(n = n()) %>% mutate(pct = round(n / sum(n),2) * 100) %>% ungroup %>% left_join(treesTop10SpeciesVector, by = c('spc_common' = 'spc_common')) %>% ggplot(aes(reorder(x = spc_common,-nPerc), y = ntaname, size = pct, group = spc_common, color = spc_common, label = paste0(pct,'%') )) + #scale_size(range = c(1, 10)) + geom_point(position = position_dodge(width = 0.5), alpha = 0.7 ) + geom_text(size = 3, position = position_dodge(width = 0.5), show.legend = FALSE, size = 4, hjust = -1) + theme_minimal() + theme(axis.text.x = element_text(angle = 90, vjust = 1.1, hjust=1, size = 8), axis.title = element_blank(), axis.text.y = element_text(size = 8)) + guides(size = FALSE) + theme(legend.position = "none") + scale_colour_manual(values=c("Honeylocust" = "brown", "Callery Pear" = "darkgreen", "Ginkgo" = "brown",
"Pin Oak" = "darkgreen",
"Sophora" = "brown", "London Planetree" = "darkgreen", "Japanese Zelkova" = "brown", "Littleleaf Linden" = "darkgreen", "American Elm" = "brown", "American Linden" = "darkgreen"))

mostCommonTreesDist

##################################################

4.) Curb location - ON / OFF CURB

mycols <- c("#E0E0E0", "#E71313")

treesCurb <- trees %>% select(curb_loc) %>% group_by(curb_loc) %>% summarize(n=n()) %>% mutate(nPerc = round(n/sum(n),3)) %>% mutate(lab.ypos = case_when(curb_loc == 'OffsetFromCurb' ~ cumsum(nPerc) +13.5nPerc, curb_loc == 'OnCurb' ~ cumsum(nPerc) - 0.95nPerc)) %>% ungroup %>% ggplot(aes(x = 3, y = nPerc, fill = curb_loc)) + geom_bar(width = 1, stat = "identity", color = "white") + coord_polar(thet = "y", start = 0.1) + geom_text(aes(y = lab.ypos, label = nPerc*100), color = "black", size = 3)+ scale_fill_manual(values = mycols) + theme_void() + xlim(0.4, 4.5) + ggtitle('Fig_3: Trees location in relation to the curb [%]') + theme(legend.position = "none") + theme(plot.title=element_text(size = 10, face = 'plain', hjust=0.5, color = '#576161'), legend.key.size = unit(0.5, 'cm'), legend.position = c(0.51, 0.5), legend.title = element_blank(), legend.text = element_text(size=8))

#treesCurb

treesCurbMapData <- trees %>% select(nta, curb_loc) %>% group_by(nta, curb_loc) %>% summarize(n = n()) %>% mutate(pct = round(n / sum(n),2)*100) %>% ungroup() %>% filter(curb_loc == 'OnCurb')

neighborhoodsManhattan <- neighborhoods %>% filter(boroname == 'Manhattan')

neighborhoodsManhattanExclParks <- neighborhoodsManhattan %>% filter(ntacode != 'MN99')

neighborhoodsManhattanExclParks <- neighborhoodsManhattanExclParks %>% left_join(treesCurbMapData, by = c('ntacode' = 'nta'))

centroids = st_centroid(neighborhoodsManhattanExclParks)

centroids <- centroids %>% mutate(geometry1 = gsub('[()°c\]', '', geometry)) %>% separate(col = geometry1, into = c('Latitude', 'Longitude'), sep = '\, ') %>% mutate(Latitude = as.numeric(Latitude), Longitude = as.numeric(Longitude)) %>% as.data.frame()

onCurbTreesMapPlot <- ggplot() + geom_sf(data = neighborhoodsManhattanExclParks, aes(fill = pct), colour = "grey" ) + theme_void() + theme(legend.position = "none") + ggtitle('Fig_4: OnCurb trees as % of total trees') + theme(plot.title=element_text(size = 10, face = 'plain', hjust=0.5, color = '#576161')) + scale_fill_gradientn(colors = brewer.pal(n = 9, name = 'YlOrRd')[1:9]) + geom_text(label = neighborhoodsManhattanExclParksLatitude, y = centroids$Longitude), size = 3, color = 'lightgrey')

#onCurbTreesMapPlot

grid.arrange(treesCurb, onCurbTreesMapPlot, ncol = 2)

##################################################

5.) Health - All trees

treesTop20 <- trees %>% group_by(spc_common) %>% summarise(n = n()) %>% arrange(desc(n)) %>% mutate(nPerc = round(n/sum(n)*100,2), cumSum = cumsum(n), cumSumPerc = round(cumSum/sum(n)*100,2)) %>% ungroup() %>% head(n = 21) %>% drop_na() %>% as.data.frame()

treesAll <- trees %>% mutate(spc_common = 'All trees')

treesHealthData <- rbind(trees, treesAll)

treesHealtPlot <- treesHealthData %>% filter(spc_common %in% c(treesTop20$spc_common, 'All trees')) %>% filter(status =='Alive') %>% group_by(spc_common, curb_loc, health) %>% summarize(n = n()) %>% mutate(pct = round(n / sum(n),2)*100) %>% mutate(health = factor(health,levels = c("Good","Fair","Poor"))) %>% ggplot(aes(y = spc_common, x = curb_loc, size = pct, group = health,
label = pct )) + scale_size(range = c(2, 5)) + geom_point(aes(color = health), position = position_dodge(width = 0.5), alpha = 0.5) + geom_text(size = 3, position = position_dodge(width = 0.5),show.legend = FALSE, size = 4, hjust = -0.8) + theme_minimal() + theme(axis.title = element_blank()) + guides(size = FALSE) + scale_color_manual(values = c("Good" = "green", "Fair" = "orange", 'Poor' = 'red'))

treesHealtPlot

##################################################

6.) Health - Map of ntas

treesHealth <- trees %>% filter(status =='Alive') %>% select(nta,health) %>% group_by(nta,health) %>% summarize(n = n()) %>% mutate(pct = round(n / sum(n),2)*100) %>% ungroup() %>% spread(key = health, value = pct) %>% select(-n) %>% group_by(nta) %>% replace(is.na(.), 0) %>% summarise(Fair = sum(Fair), Good = sum(Good), Poor = sum(Poor))

neighborhoodsManhattan <- neighborhoods %>% filter(boroname == 'Manhattan')

neighborhoodsManhattanExclParks <- neighborhoodsManhattan %>% filter(ntacode != 'MN99')

neighborhoodsManhattanExclParks <- neighborhoodsManhattanExclParks %>% left_join(treesHealth, by = c('ntacode' = 'nta'))

centroids = st_centroid(neighborhoodsManhattanExclParks)

centroids <- centroids %>% mutate(geometry1 = gsub('[()°c\]', '', geometry)) %>% separate(col = geometry1, into = c('Latitude', 'Longitude'), sep = '\, ') %>% mutate(Latitude = as.numeric(Latitude), Longitude = as.numeric(Longitude)) %>% as.data.frame()

plotManhattanFun <- function(health, colors, labelsTest, plotTitle) { ggplot() + geom_sf(data = neighborhoodsManhattanExclParks, aes_string(fill = health), colour = "grey" ) + ggtitle(plotTitle) + theme_void() + theme(plot.title=element_text(size = 12, face = 'plain', hjust=0.5, color = '#576161'), legend.position = "none") + # theme(legend.position="bottom") + scale_fill_gradientn(colors = brewer.pal(n = 9, name = colors)[2:9]) + geom_text(label = labelsTest, aes(x = centroidsLongitude), size = 3) }

neighborhoodsManhattanDict <- ggplot() + geom_sf(data = neighborhoodsManhattanExclParksTrees, colour = "darkgrey" ) + theme_void() + ggtitle("NTA names") + theme(plot.title=element_text(size = 12, face = 'plain', hjust=0.5, color = '#576161'), legend.position = "none") + guides(fill=guide_legend(title="number of trees")) + geom_text(label = centroidsManhattanLatitude, y = centroidsManhattan$Longitude), size = 1.8)

grid.arrange(neighborhoodsManhattanDict, plotManhattanFun("Good", "Greens", neighborhoodsManhattanExclParksFair, '% of trees with fair health'), plotManhattanFun("Poor", "Reds", neighborhoodsManhattanExclParks$Poor, '% of trees with poor health'), ncol = 4)

##################################################

7.) Trunk

treesTop10 <- trees %>% group_by(spc_common) %>% summarise(n = n()) %>% arrange(desc(n)) %>% mutate(nPerc = round(n/sum(n)*100,2), cumSum = cumsum(n), cumSumPerc = round(cumSum/sum(n)*100,2)) %>% ungroup() %>% head(n = 11) %>% drop_na() %>% as.data.frame()

outliers

trees %>% filter(spc_common %in% (treesTop10$spc_common)) %>% select(spc_common, tree_dbh) %>% group_by(spc_common) %>% summarise(IQRRange = IQR(tree_dbh), median = median(tree_dbh), Q3 = quantile(tree_dbh, probs=c(.25, .75))[2], outlier = 1.5 * IQRRange + Q3)

#boxplot treesTop10BoxPlot <- trees %>% filter(spc_common %in% (treesTop10$spc_common)) %>% select(spc_common, tree_dbh) %>% ggplot(aes(x = tree_dbh, y = reorder(spc_common,tree_dbh,median))) + geom_boxplot(color="black", fill='#4DAF4A', alpha= 0.5, outlier.size = 2, outlier.color = 'grey') + scale_x_continuous(limits=c(0,50)) + theme_minimal() + xlab("medianDiameter") + theme(axis.title.y = element_blank())

#barChart treesTop10Median <- trees %>% filter(spc_common %in% (treesTop10$spc_common)) %>% group_by(spc_common, curb_loc) %>% summarize(medianDiameter = median(tree_dbh)) %>% ggplot(aes(y = reorder(spc_common,medianDiameter,median), x = medianDiameter, fill = curb_loc, label = paste0(medianDiameter))) + geom_col(width = 0.5, position = position_dodge(0.7), alpha = 0.7) + geom_text(width = 0.5, position = position_dodge(0.7), hjust = -0.2, vjust = 0.4, size = 3) + theme_minimal() + scale_x_continuous(limits = c(0,20)) + scale_fill_brewer(palette="Set1") + theme(axis.title.y = element_blank())

grid.arrange(treesTop10BoxPlot, treesTop10Median, ncol = 2)

##################################################

8.) Ratings - table

TOP TEN BEST TREES

trees <- as.data.frame(read_csv('trees1.csv')) trees <- trees %>% mutate(spc_common = str_to_title(spc_common))

treesTop20Names <- trees %>% group_by(spc_common) %>% summarise(n = n()) %>% arrange(desc(n)) %>% mutate(nPerc = round(n/sum(n)*100,2), cumSum = cumsum(n), cumSumPerc = round(cumSum/sum(n)*100,2)) %>% ungroup() %>% head(n = 21) %>% drop_na() %>% select(spc_common)

treesTop10Health <- trees %>% filter(spc_common %in% c(treesTop20Names$spc_common)) %>% select(spc_common,health) %>% group_by(spc_common,health) %>% summarize(n = n()) %>% mutate(pct = round(n / sum(n),2)*100) %>% filter(health == 'Good') %>% arrange((pct)) %>% ungroup() %>% mutate(pointsHealth = row_number())

treesTop10Trunk <- trees %>% filter(spc_common %in% c(treesTop20Names$spc_common)) %>% select(spc_common,tree_dbh) %>% group_by(spc_common) %>% summarize(medianDiameter = median((tree_dbh))) %>% arrange((medianDiameter)) %>% mutate(pointsDiameter = row_number())

treesTop10Ranking <- treesTop10Health %>% left_join(treesTop10Trunk, by = 'spc_common') %>% select(spc_common, pointsHealth, pointsDiameter) %>% mutate(pointsTotal = pointsHealth + pointsDiameter) %>% arrange(desc(pointsTotal)) %>% mutate(positionTotal = row_number()) %>% mutate(price = case_when(positionTotal == 1 ~'1st', positionTotal == 2 ~'2nd', positionTotal == 3 ~'3rd', positionTotal == 4 ~'4th', positionTotal == 5 ~'5th', positionTotal == 6 ~'6th', positionTotal == 7 ~'7th', positionTotal == 8 ~'8th', positionTotal == 9 ~'9th', positionTotal == 10 ~'10th' )) %>% mutate(position = price) %>% head(n = 10) %>% select(spc_common, price,position, pointsTotal, pointsHealth, pointsDiameter)

#install.packages('base64enc') library(base64enc)

gold <- sprintf("data:image/png;base64,%s", base64encode("gold_medal_1.png")) silver <- sprintf("data:image/png;base64,%s", base64encode("silver_medal_1.png")) bronze <- sprintf("data:image/png;base64,%s", base64encode("bronze_medal_1.png")) flowers <- sprintf("data:image/png;base64,%s", base64encode("flowers_medal_1.png"))

image_tile <- formatter("img", src = position ~ ifelse(position == "1st", gold, ifelse(position == "2nd", silver, ifelse(position == "3rd", bronze, flowers))), # Control height and width, either directly - width = 25, # Or via a formula height = 20, NA)

treesTop10RankingFormattable <- formattable( treesTop10Ranking, align =c("l","c"), list(price = image_tile ))

export_formattable <- function(f, file, width = "60%", height = NULL, background = "white", delay = 0.2) { w <- as.htmlwidget(f, width = width, height = height) path <- htmltools::html_print(w, background = background, viewer = NULL) url <- paste0("file:///", gsub("\\", "/", normalizePath(path))) webshot(url, file = file, selector = ".formattable_widget", delay = delay) }

export_formattable(treesTop10RankingFormattable, 'treesTop10RankingFormattable.jpg')

img2 <- readJPEG('treesTop10RankingFormattable.jpg') img2 = abind::abind(img2, img2[,,1]) img2[,,4] = 1

h <-dim(img2)[1] # image height w <-dim(img2)[2] # image width

treesTop10RankingFormattablePlot <- ggplot() + annotation_custom(grid::rasterGrob(img2, width=unit(1,"npc"), height=unit(1,"npc")), 0, w, 0, -h) + scale_x_continuous(expand=c(0,0),limits=c(0,w)) + scale_y_reverse(expand=c(0,0),limits=c(h,0)) + coord_equal() + stat_bin2d(binwidth=2,aes(fill = ..density..)) + theme_void()

##################################################

9.) Appendix -dead trees

trees <- as.data.frame(read_csv('trees1.csv')) trees <- trees %>% mutate(spc_common = str_to_title(spc_common))

neighborhoods <- st_read("nta.geojson")

deadTreesPlotFunction <- function(curb, title) { treesDead <- trees %>% filter(curb_loc %in% curb) %>% select(nta, status) %>% group_by(nta, status) %>% summarize(n = n()) %>% mutate(pct = round(n / sum(n),3)*100) %>% ungroup() %>% complete(nta, status, fill = list(n = 0)) %>% replace(is.na(.), 0) %>% filter(status == 'Dead')

neighborhoodsManhattan <- neighborhoods %>% filter(boroname == 'Manhattan')

neighborhoodsManhattanExclParks <- neighborhoodsManhattan %>% filter(ntacode != 'MN99')

neighborhoodsManhattanExclParks <- neighborhoodsManhattanExclParks %>% left_join(treesDead, by = c('ntacode' = 'nta'))

centroids = st_centroid(neighborhoodsManhattanExclParks)

centroids <- centroids %>% mutate(geometry1 = gsub('[()°c\]', '', geometry)) %>% separate(col = geometry1, into = c('Latitude', 'Longitude'), sep = '\, ') %>% mutate(Latitude = as.numeric(Latitude), Longitude = as.numeric(Longitude)) %>% as.data.frame()

ggplot() + geom_sf(data = neighborhoodsManhattanExclParks, aes(fill = pct), colour = "grey" ) + theme_void() + ggtitle(title) + theme(legend.position = 'none', plot.title=element_text(size = 10, face = 'plain', hjust=0.5, color = '#576161')) + scale_fill_gradientn(colors = brewer.pal(n = 9, name = 'Reds')[2:8]) + geom_text(label = neighborhoodsManhattanExclParksLatitude, y = centroids$Longitude), size = 3) }

#options(repr.plot.width=10, repr.plot.height=7)

plot1 <- deadTreesPlotFunction(c('OffsetFromCurb','OnCurb'), 'All trees') plot2 <- deadTreesPlotFunction(c('OnCurb'), 'OnCurb') plot3 <- deadTreesPlotFunction(c('OffsetFromCurb'), 'OffsetFromCurb')

grid.arrange(plot1, plot2, plot3, ncol = 3)

grid.arrange(arrangeGrob(plot1, plot2, plot3, ncol = 3), heights=c(20, 1),top = textGrob("Fig_6: Percentage of dead trees",gp=gpar(col = '#576161',fontsize=10,font=3)))

##################################################

head(n = 10) %>% select(spc_common, pointsHealth, pointsDiameter, positionTotal) %>% gather(key = "category", value = "points", 2:4) %>% mutate(positionInRanking = case_when(category == 'positionTotal' ~ points*1, category == 'pointsHealth' | category == 'pointsDiameter' ~ 21 - points )) %>% view()

treesTop10RankingSummary <- treesTop10Ranking %>% filter(category == 'positionTotal') %>% select(spc_common,points) %>% mutate(points = points*1) %>% arrange(desc(points))

treesTop10Ranking %>% ggplot(aes(y = factor(spc_common,levels = c(treesTop10RankingSummary$spc_common)), x = factor(category, levels = c('positionTotal', 'pointsDiameter', 'pointsHealth')), size = points, group = category, label = positionInRanking)) + scale_size(range = c(2, 5)) + geom_point(aes(color = category), position = position_dodge(width = 0.5), alpha = 0.5) + geom_text(size = 3, position = position_dodge(width = 0.5),show.legend = FALSE, size = 4, hjust = -0.8) + theme_minimal() + theme(axis.title = element_blank()) + guides(size = FALSE) + scale_color_manual(values = c("pointsHealth" = "green", "pointsDiameter" = "orange", 'positionTotal' = 'red'))

treesTop10RankingMapData <- trees %>% left_join(treesTop10Ranking, by = 'spc_common') %>% replace(is.na(.), 'oth') %>% select(nta, nta_name, topTen) %>% group_by(nta, nta_name, topTen) %>% summarise(n = n()) %>% mutate(pct = round(n / sum(n),2)*100) %>% filter(topTen == 'topTen') %>% arrange(desc(pct))

neighborhoodsManhattan <- neighborhoods %>% filter(boroname == 'Manhattan')

neighborhoodsManhattanExclParks <- neighborhoodsManhattan %>% filter(ntacode != 'MN99')

centroidsManhattan = st_centroid(neighborhoodsManhattanExclParks)

centroidsManhattan <- centroidsManhattan %>% mutate(geometry1 = gsub('[()°c\]', '', geometry)) %>% separate(col = geometry1, into = c('Latitude', 'Longitude'), sep = '\, ') %>% mutate(Latitude = as.numeric(Latitude), Longitude = as.numeric(Longitude)) %>% as.data.frame()

neighborhoodsManhattanExclParksTopTenRanking <- neighborhoodsManhattanExclParks %>% left_join(treesTop10RankingMapData, by = c('ntacode' = 'nta'))

neighborhoodsManhattanExclParksTopTenRankingPlot <- neighborhoodsManhattanExclParksTopTenRanking %>% ggplot() + geom_sf(data = neighborhoodsManhattanExclParksTopTenRanking, aes(fill = pct), colour = "grey") + theme_void() + theme(legend.position = "none") + theme(legend.position="bottom") + theme(legend.position = "none") + theme(#legend.position="left", legend.key.size = unit(0.5, 'cm'), legend.position = c(-0.05, 0.3), legend.title = element_text(size=7), legend.text = element_text(size=7)) + guides(fill=guide_legend(title="top 10 trees [%]")) + scale_fill_gradientn(colors = brewer.pal(n = 9, name = 'Greens')[1:8]) + geom_text(label = centroidsManhattanLatitude, y = centroidsManhattan$Longitude), size = 2)

improvement_formatter <- formatter("span", style = x ~ style(font.weight = "bold", display = 'block', width = '100px', color = ifelse(x > 0, 'Green', ifelse(x < 0, 'Red',"black"))), x ~ icontext(ifelse(x>0, "arrow-up", "arrow-down"), x))

customGreen = brewer.pal(n = 9 , name = "Greens")[1] customGreen0 = brewer.pal(n = 9 , name = "Greens")[8] customBlue = brewer.pal(n = 9 , name = "Blues")[1] customBlue0 = brewer.pal(n = 9 , name = "Blues")[6]

costum_format1 <- formatter(.tag = "span", style = function(x) style( display = "block", padding = "0 4px", width = '150px', 'border-radius' = "4px", 'background-color' = csscolor(gradient(as.numeric(x),customGreen, customGreen0)), color = ifelse(x >= 3, "black","black")))

costum_format2 <- formatter(.tag = "span", style = function(x) style( display = "block", width = "50px"))

costum_format3 <- formatter(.tag = "span", style = function(x) style( display = "block", padding = "0 4px", width = '50px', 'border-radius' = "4px"))

treesTop10RankingMapTable <- treesTop10RankingMapData %>% select(nta, nta_name, pct)

treesTop10RankingMapTable

testFormattable = formattable(treesTop10RankingMapTable, align =c("l","l","l"), list('nta' = costum_format3, 'ntaname' = costum_format2, pct = costum_format1 ))

testFormattable %>%

kable() %>%

kable_styling(bootstrap_options = "bordered",

full_width = FALSE) %>%

add_header_above(c("", "Category1" = 2)) %>%

collapse_rows(columns = 1,

valign = "middle")

webshot::install_phantomjs()

export_formattable <- function(f, file, width = "60%", height = NULL, background = "white", delay = 0.2) { w <- as.htmlwidget(f, width = width, height = height) path <- htmltools::html_print(w, background = background, viewer = NULL) url <- paste0("file:///", gsub("\\", "/", normalizePath(path))) webshot(url, file = file, selector = ".formattable_widget", delay = delay) }

export_formattable(testFormattable, 'testFormattable.jpg')

img2 <- readJPEG('testFormattable.jpg') img2 = abind::abind(img2, img2[,,1]) img2[,,4] = 1

h <-dim(img2)[1] # image height w <-dim(img2)[2] # image width

plotTableTopTen <- ggplot() + annotation_custom(grid::rasterGrob(img2, width=unit(1,"npc"), height=unit(1,"npc")), 0, w, 0, -h) + scale_x_continuous(expand=c(0,0),limits=c(0,w)) + scale_y_reverse(expand=c(0,0),limits=c(h,0)) + coord_equal() + stat_bin2d(binwidth=2,aes(fill = ..density..)) + theme_void()

plotTableTopTen neighborhoodsManhattanExclParksTopTenRankingPlot

grid.arrange(neighborhoodsManhattanExclParksTopTenRankingPlot, plotTableTopTen, ncol = 2)

DEAD TREES

single trees

neighborhoodsManhattan <- neighborhoods %>% filter(boroname == 'Manhattan')

neighborhoodsManhattanExclParks <- neighborhoodsManhattan %>% filter(ntacode != 'MN99')

treesSP <- st_as_sf(trees, coords = c('longitude','latitude'), crs = st_crs(neighborhoodsManhattan)) treesSPTest <- head(treesSP, n = 100, crs = st_crs(neighborhoodsManhattan))

ggplot() + #geom_sf(data = subset(treesSP, status == 'Alive'), size = 0.05, alpha = 0.5, color = "green") + geom_sf(data = subset(treesSP, status == 'Dead'), size = 0.05, color = "red") + geom_sf(data = neighborhoodsManhattanExclParks, lwd = 0, alpha = 0.3) + theme_void()

DEAD TREES

single trees

neighborhoodsManhattan <- neighborhoods %>% filter(boroname == 'Manhattan')

neighborhoodsManhattanExclParks <- neighborhoodsManhattan %>% filter(ntacode != 'MN99')

treesSP <- st_as_sf(trees, coords = c('longitude','latitude'), crs = st_crs(neighborhoodsManhattan)) treesSPTest <- head(treesSP, n = 100, crs = st_crs(neighborhoodsManhattan))

neighborhoodsNoGeometry <- st_set_geometry(neighborhoods, NULL)

neighborhoodsNoGeometryManhattan <- neighborhoodsNoGeometry %>% filter(boroname == 'Manhattan') %>% select(ntacode)

mycols <- c(brewer.pal(n = 9 , name = "Greens")[4], "#E71313")

deadTreesPlot <- trees %>% #filter(nta %in% c(neighborhoodsNoGeometryManhattan$ntacode)) %>% select(status) %>% group_by(status) %>% summarize(n=n()) %>% mutate(nPerc = round(n/sum(n),3)) %>% #mutate(lab.ypos = cumsum(nPerc) - 0.48nPerc) %>% mutate(lab.ypos = case_when(status == 'Dead' ~ cumsum(nPerc) - 35.15nPerc, status == 'Alive' ~ cumsum(nPerc) - 0.48nPerc)) %>% ungroup %>% ggplot(aes(x = 3, y = nPerc, fill = status)) + geom_bar(width = 1, stat = "identity", color = 'white') + coord_polar(thet = "y", start = 0) + geom_text(aes(y = lab.ypos, label = nPerc100), color = "black", size = 3.2)+ scale_fill_manual(values = mycols) + theme_void() + ggtitle('All trees') + theme(legend.position = "none", plot.title=element_text(size = 10, face = 'plain', hjust=0.5, color = '#576161')) + xlim(0.5, 4.5)

deadTreesPlotOffCurb <- trees %>% filter(curb_loc == 'OffsetFromCurb') %>% select(status) %>% group_by(status) %>% summarize(n=n()) %>% mutate(nPerc = round(n/sum(n),3)) %>% #mutate(lab.ypos = cumsum(nPerc) - 0.48nPerc) %>% mutate(lab.ypos = case_when(status == 'Dead' ~ cumsum(nPerc) - 52.15nPerc, status == 'Alive' ~ cumsum(nPerc) - 0.48nPerc)) %>% ungroup %>% ggplot(aes(x = 3, y = nPerc, fill = status)) + geom_bar(width = 1, stat = "identity", color = "white") + coord_polar(thet = "y", start = 0) + geom_text(aes(y = lab.ypos, label = nPerc100), color = "black", size = 3.2)+ scale_fill_manual(values = mycols) + theme_void() + ggtitle('OffsetFromCurb') + theme(legend.position = "none", plot.title=element_text(size = 10, face = 'plain', hjust=0.5, color = '#576161')) + xlim(0.5, 4.5)

deadTreesPlotOnCurb <- trees %>% filter(curb_loc == 'OnCurb') %>% select(status) %>% group_by(status) %>% summarize(n=n()) %>% mutate(nPerc = round(n/sum(n),3)) %>% #mutate(lab.ypos = cumsum(nPerc) - 0.48nPerc) %>% mutate(lab.ypos = case_when(status == 'Dead' ~ cumsum(nPerc) - 33.9nPerc, status == 'Alive' ~ cumsum(nPerc) - 0.48nPerc)) %>% ungroup %>% ggplot(aes(x = 3, y = nPerc, fill = status)) + geom_bar(width = 1, stat = "identity", color = "white") + coord_polar(thet = "y", start = 0) + geom_text(aes(y = lab.ypos, label = nPerc100), color = "black", size = 3.2)+ scale_fill_manual(values = mycols) + theme_void() + ggtitle('OnCurb') + theme(legend.position = "bottom", plot.title=element_text(size = 10, face = 'plain', hjust=0.5, color = '#576161')) + xlim(0.5, 4.5)

g_legend<-function(a.gplot){ tmp <- ggplot_gtable(ggplot_build(a.gplot)) leg <- which(sapply(tmpname) == "guide-box") legend <- tmp$grobs[[leg]] return(legend)}

mylegend<-g_legend(deadTreesPlotOnCurb)

grid.arrange(arrangeGrob(deadTreesPlot, deadTreesPlotOnCurb + theme(legend.position="none"), deadTreesPlotOffCurb, ncol = 3), mylegend ,heights=c(20, 1),top = textGrob("Fig_5: Percentage of dead trees",gp=gpar(col = '#576161',fontsize=10,font=3)))

percentages - map

treesDead <- trees %>% #filter(status =='Dead') %>% select(nta, status) %>% group_by(nta, status) %>% summarize(n = n()) %>% mutate(pct = round(n / sum(n),3)*100) %>% ungroup() %>% filter(status == 'Dead')

neighborhoodsManhattan <- neighborhoods %>% filter(boroname == 'Manhattan')

neighborhoodsManhattanExclParks <- neighborhoodsManhattan %>% filter(ntacode != 'MN99')

neighborhoodsManhattanExclParks <- neighborhoodsManhattanExclParks %>% left_join(treesDead, by = c('ntacode' = 'nta'))

centroids = st_centroid(neighborhoodsManhattanExclParks)

centroids <- centroids %>% mutate(geometry1 = gsub('[()°c\]', '', geometry)) %>% separate(col = geometry1, into = c('Latitude', 'Longitude'), sep = '\, ') %>% mutate(Latitude = as.numeric(Latitude), Longitude = as.numeric(Longitude)) %>% as.data.frame()

deadTreesMapPlot <- ggplot() + geom_sf(data = neighborhoodsManhattanExclParks, aes(fill = pct), colour = "grey" ) + theme_void() + theme(legend.position = "none") + scale_fill_gradientn(colors = brewer.pal(n = 9, name = 'Reds')[1:9]) + geom_text(label = neighborhoodsManhattanExclParksLatitude, y = centroids$Longitude), size = 3)

dead trees summary

deadTreesPlot deadTreesPlotOnCurb deadTreesPlotOffCurb deadTreesMapPlot

grid.arrange(arrangeGrob(deadTreesPlot,arrangeGrob( deadTreesPlotOnCurb, deadTreesPlotOffCurb,ncol = 2), ncol = 1), deadTreesMapPlot, ncol = 2)

ALIVE TREES

head(trees) glimpse(trees)

trees %>% group_by(spc_common) %>% summarise(n = n()) %>% arrange(desc(n)) %>% mutate(nPerc = round(n/sum(n)*100,1), cumSum = cumsum(n), cumSumPerc = round(cumSum/sum(n)*100,1)) %>% view()

treesTop10SpeciesVector <- trees %>% group_by(spc_common) %>% summarise(n = n()) %>% arrange(desc(n)) %>% mutate(nPerc = round(n/sum(n)*100,2), cumSum = cumsum(n), cumSumPerc = round(cumSum/sum(n)*100,2)) %>% ungroup() %>% head(n = 11) %>% drop_na() %>% as.data.frame()

dfTest <- trees %>% filter(spc_common %in% c(treesTop10SpeciesVector$spc_common)) %>% left_join(neighborhoods, by = c('nta' = 'ntacode')) %>% select(ntaname) %>% arrange(desc(ntaname)) %>% unique() %>% mutate(lp = row_number())

dfTest$lp <- rep(c('brown', 'darkgreen'), times = 28/2)

trees %>% filter(spc_common %in% c(treesTop10SpeciesVector$spc_common)) %>% left_join(neighborhoods, by = c('nta' = 'ntacode')) %>% select(ntaname,spc_common, boroname) %>% group_by(ntaname,spc_common) %>% summarize(n = n()) %>% mutate(pct = round(n / sum(n),2) * 100) %>% ungroup %>% mutate(ntaname = factor(ntaname, unique(ntaname)), spc_common = factor(spc_common, unique(spc_common))) %>% mutate(xstart2 = row_number() - 0.5, xend2 = row_number()+0.5) %>% #mutate(ystart2 = col_number() - 0.5, yend2 = col_number()+0.5) %>% left_join(treesTop10SpeciesVector, by = c('spc_common' = 'spc_common')) %>% left_join(dfTest, by = c('ntaname' = 'ntaname')) %>% ggplot(aes(x = reorder(spc_common,-nPerc), y = ntaname, size = pct, group = spc_common,
color = ntaname, label = paste0(pct,'%'), )) + #scale_size(range = c(1, 10)) + geom_point(position = position_dodge(width = 0.5), alpha = 0.5) + geom_text(size = 3, position = position_dodge(width = 0.5),show.legend = FALSE, size = 4, hjust = -1, vjust= 0.3) + scale_colour_manual(values = rep(c('black'), times = 28)) + geom_rect(aes(xmin = 1-0.5, xmax = 10+0.5, ymin = xstart2/10+0.5, ymax = xend2/10+0.5, fill = lp), alpha = 0.20, col = NA) + #scale_fill_manual(values = alpha(c("red", "green"))) + theme_minimal() + theme(axis.text.x = element_text(angle = 90, vjust = 1.1, hjust=1, size = 8), axis.text.y = element_text(size = 8)) + guides(size = FALSE) + theme(legend.position = "none") + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank())

geom_rect( data = rect_data, # or rect_data for actual season end boundary mapping = aes( xmin = xstart2, xmax = xend2, ymin = 0, ymax = 10, fill = spc_common, alpha = 0.5 ))

rect_data <- treesTop10SpeciesVector %>% select(spc_common) %>% mutate(xstart2 = row_number() - 0.5, xend2 = row_number()+0.5) %>% distinct()

ggplot() + geom_rect( data = rect_data, # or rect_data for actual season end boundary mapping = aes( xmin = xstart2, xmax = xend2, ymin = 0, ymax = 10, fill = spc_common, alpha = 0.5 ))

TREES - HEALTH

TREES - HEALTH - GENERAL

trees %>% filter(status =='Alive') %>% mutate(spc_common = 'All trees') %>% group_by(spc_common, curb_loc, health) %>% summarize(n = n()) %>% mutate(pct = round(n / sum(n),2)*100) %>% mutate(health = factor(health,levels = c("Good","Fair","Poor"))) %>% ggplot(aes(y = spc_common, x = curb_loc, size = pct, group = health,
label = pct )) + scale_size(range = c(2, 5)) + geom_point(aes(color = health), position = position_dodge(width = 0.5), alpha = 0.5) + geom_text(size = 3, position = position_dodge(width = 0.5),show.legend = FALSE, size = 4, hjust = -0.8) + theme_minimal()+ guides(size = FALSE) + scale_color_manual(values = c("Good" = "green", "Fair" = "orange", 'Poor' = 'red'))

TREES - HEALTH - DETAILES

treesTop20 <- trees %>% group_by(spc_common) %>% summarise(n = n()) %>% arrange(desc(n)) %>% mutate(nPerc = round(n/sum(n)*100,2), cumSum = cumsum(n), cumSumPerc = round(cumSum/sum(n)*100,2)) %>% ungroup() %>% head(n = 21) %>% drop_na() %>% as.data.frame()

treesAll <- trees %>% mutate(spc_common = 'All trees')

treesHealthData <- rbind(trees, treesAll)

treesHealthData %>% filter(spc_common %in% c(treesTop10Names$spc_common, 'All trees')) %>% filter(status =='Alive') %>% group_by(spc_common, curb_loc, health) %>% summarize(n = n()) %>% mutate(pct = round(n / sum(n),2)*100) %>% mutate(health = factor(health,levels = c("Good","Fair","Poor"))) %>% ggplot(aes(y = spc_common, x = curb_loc, size = pct, group = health,
label = pct )) + scale_size(range = c(2, 5)) + geom_point(aes(color = health), position = position_dodge(width = 0.5), alpha = 0.5) + geom_text(size = 3, position = position_dodge(width = 0.5),show.legend = FALSE, size = 4, hjust = -0.8) + theme_minimal()+ guides(size = FALSE) + scale_color_manual(values = c("Good" = "green", "Fair" = "orange", 'Poor' = 'red'))

TREES - HEALTH - MAPS

treesHealth <- trees %>% filter(status =='Alive') %>% select(nta,health) %>% group_by(nta,health) %>% summarize(n = n()) %>% mutate(pct = round(n / sum(n),2)*100) %>% ungroup() %>% spread(key = health, value = pct) %>% select(-n) %>% group_by(nta) %>% replace(is.na(.), 0) %>% summarise(Fair = sum(Fair), Good = sum(Good), Poor = sum(Poor))

neighborhoodsManhattan <- neighborhoods %>% filter(boroname == 'Manhattan')

neighborhoodsManhattanExclParks <- neighborhoodsManhattan %>% filter(ntacode != 'MN99')

neighborhoodsManhattanExclParks <- neighborhoodsManhattanExclParks %>% left_join(treesHealth, by = c('ntacode' = 'nta'))

centroids = st_centroid(neighborhoodsManhattanExclParks)

centroids <- centroids %>% mutate(geometry1 = gsub('[()°c\]', '', geometry)) %>% separate(col = geometry1, into = c('Latitude', 'Longitude'), sep = '\, ') %>% mutate(Latitude = as.numeric(Latitude), Longitude = as.numeric(Longitude)) %>% as.data.frame()

plotManhattanFun <- function(health, colors, labelsTest) { ggplot() + geom_sf(data = neighborhoodsManhattanExclParks, aes_string(fill = health), colour = "grey" ) + theme_void() + theme(legend.position = "none") +

theme(legend.position="bottom") +

scale_fill_gradientn(colors = brewer.pal(n = 9, name = colors)[2:9]) + geom_text(label = labelsTest, aes(x = centroidsLongitude), size = 3) }

grid.arrange(plotManhattanFun("Good", "Greens", neighborhoodsManhattanExclParksFair), plotManhattanFun("Poor", "Reds", neighborhoodsManhattanExclParks$Poor), ncol = 3)

TREES - SIZE

treesTop10 <- trees %>% group_by(spc_common) %>% summarise(n = n()) %>% arrange(desc(n)) %>% mutate(nPerc = round(n/sum(n)*100,2), cumSum = cumsum(n), cumSumPerc = round(cumSum/sum(n)*100,2)) %>% ungroup() %>% head(n = 11) %>% drop_na() %>% as.data.frame()

outliers

trees %>% filter(spc_common %in% (treesTop10$spc_common)) %>% select(spc_common, tree_dbh) %>% group_by(spc_common) %>% summarise(IQRRange = IQR(tree_dbh), median = median(tree_dbh), Q3 = quantile(tree_dbh, probs=c(.25, .75))[2], outlier = 1.5 * IQRRange + Q3)

#boxplot treesTop10BoxPlot <- trees %>% filter(spc_common %in% (treesTop10$spc_common)) %>% select(spc_common, tree_dbh) %>% ggplot(aes(x = tree_dbh, y = reorder(spc_common,tree_dbh,median))) + geom_boxplot(color="black", fill='#4DAF4A', alpha= 0.5, outlier.size = 2, outlier.color = 'grey') + scale_x_continuous(limits=c(0,50)) + theme_minimal()

#barChart treesTop10Median <- trees %>% filter(spc_common %in% (treesTop10$spc_common)) %>% group_by(spc_common, curb_loc) %>% summarize(medianDiameter = median(tree_dbh)) %>% ggplot(aes(y = reorder(spc_common,medianDiameter,median), x = medianDiameter, fill = curb_loc, label = paste0(medianDiameter))) + geom_col(width = 0.5, position = position_dodge(0.7), alpha = 0.7) + geom_text(width = 0.5, position = position_dodge(0.7), hjust = -0.2, vjust = 0.4, size = 3) + theme_minimal() + scale_x_continuous(limits = c(0,20)) + scale_fill_brewer(palette="Set1") + theme(#axis.text.x = element_blank(), #axis.ticks.x = element_blank() )

grid.arrange(treesTop10BoxPlot, treesTop10Median, ncol = 2)

brewer.pal(n = 9, name = 'Set1')

theme(axis.text.x = element_text(angle = 90, vjust = 0.5, hjust=1, size = 10))

TOP TEN BEST TREES

trees %>% select(spc_common, health) %>% head()

trees <- as.data.frame(read_csv('Databases/trees1.csv')) trees <- trees %>% mutate(spc_common = str_to_title(spc_common))

treesTop20Names <- trees %>% group_by(spc_common) %>% summarise(n = n()) %>% arrange(desc(n)) %>% mutate(nPerc = round(n/sum(n)*100,2), cumSum = cumsum(n), cumSumPerc = round(cumSum/sum(n)*100,2)) %>% ungroup() %>% head(n = 21) %>% drop_na() %>% select(spc_common)

treesTop10Health <- trees %>% filter(spc_common %in% c(treesTop20Names$spc_common)) %>% select(spc_common,health) %>% group_by(spc_common,health) %>% summarize(n = n()) %>% mutate(pct = round(n / sum(n),2)*100) %>% filter(health == 'Good') %>% arrange((pct)) %>% ungroup() %>% mutate(pointsHealth = row_number())

treesTop10Trunk <- trees %>% filter(spc_common %in% c(treesTop20Names$spc_common)) %>% select(spc_common,tree_dbh) %>% group_by(spc_common) %>% summarize(medianDiameter = median((tree_dbh))) %>% arrange((medianDiameter)) %>% mutate(pointsDiameter = row_number())

treesTop10Ranking <- treesTop10Health %>% left_join(treesTop10Trunk, by = 'spc_common') %>% select(spc_common, pointsHealth, pointsDiameter) %>% mutate(pointsTotal = pointsHealth + pointsDiameter) %>% arrange(desc(pointsTotal)) %>% head(n = 10) %>% mutate(topTen = 'topTen')

treesTop10RankingMapData <- trees %>% left_join(treesTop10Ranking, by = 'spc_common') %>% replace(is.na(.), 'oth') %>% select(nta, nta_name, topTen) %>% group_by(nta, nta_name, topTen) %>% summarise(n = n()) %>% mutate(pct = round(n / sum(n),2)*100) %>% filter(topTen == 'topTen') %>% arrange(desc(pct))

neighborhoodsManhattan <- neighborhoods %>% filter(boroname == 'Manhattan')

neighborhoodsManhattanExclParks <- neighborhoodsManhattan %>% filter(ntacode != 'MN99')

centroidsManhattan = st_centroid(neighborhoodsManhattanExclParks)

centroidsManhattan <- centroidsManhattan %>% mutate(geometry1 = gsub('[()°c\]', '', geometry)) %>% separate(col = geometry1, into = c('Latitude', 'Longitude'), sep = '\, ') %>% mutate(Latitude = as.numeric(Latitude), Longitude = as.numeric(Longitude)) %>% as.data.frame()

neighborhoodsManhattanExclParksTopTenRanking <- neighborhoodsManhattanExclParks %>% left_join(treesTop10RankingMapData, by = c('ntacode' = 'nta'))

neighborhoodsManhattanExclParksTopTenRankingPlot <- neighborhoodsManhattanExclParksTopTenRanking %>% ggplot() + geom_sf(data = neighborhoodsManhattanExclParksTopTenRanking, aes(fill = pct), colour = "grey") + theme_void() + theme(legend.position = "none") + theme(legend.position="bottom") + theme(legend.position = "none") + theme(#legend.position="left", legend.key.size = unit(0.5, 'cm'), legend.position = c(-0.05, 0.3), legend.title = element_text(size=7), legend.text = element_text(size=7)) + guides(fill=guide_legend(title="top 10 trees [%]")) + scale_fill_gradientn(colors = brewer.pal(n = 9, name = 'Greens')[1:8]) + geom_text(label = centroidsManhattanLatitude, y = centroidsManhattan$Longitude), size = 2)

improvement_formatter <- formatter("span", style = x ~ style(font.weight = "bold", display = 'block', width = '100px', color = ifelse(x > 0, 'Green', ifelse(x < 0, 'Red',"black"))), x ~ icontext(ifelse(x>0, "arrow-up", "arrow-down"), x))

customGreen = brewer.pal(n = 9 , name = "Greens")[1] customGreen0 = brewer.pal(n = 9 , name = "Greens")[8] customBlue = brewer.pal(n = 9 , name = "Blues")[1] customBlue0 = brewer.pal(n = 9 , name = "Blues")[6]

costum_format1 <- formatter(.tag = "span", style = function(x) style( display = "block", padding = "0 4px", width = '150px', 'border-radius' = "4px", 'background-color' = csscolor(gradient(as.numeric(x),customGreen, customGreen0)), color = ifelse(x >= 3, "black","black")))

costum_format2 <- formatter(.tag = "span", style = function(x) style( display = "block", width = "50px"))

costum_format3 <- formatter(.tag = "span", style = function(x) style( display = "block", padding = "0 4px", width = '50px', 'border-radius' = "4px"))

treesTop10RankingMapTable <- treesTop10RankingMapData %>% select(nta, nta_name, pct)

treesTop10RankingMapTable

testFormattable = formattable(treesTop10RankingMapTable, align =c("l","l","l"), list('nta' = costum_format3, 'ntaname' = costum_format2, pct = costum_format1 ))

testFormattable %>%

kable() %>%

kable_styling(bootstrap_options = "bordered",

full_width = FALSE) %>%

add_header_above(c("", "Category1" = 2)) %>%

collapse_rows(columns = 1,

valign = "middle")

webshot::install_phantomjs()

export_formattable <- function(f, file, width = "60%", height = NULL, background = "white", delay = 0.2) { w <- as.htmlwidget(f, width = width, height = height) path <- htmltools::html_print(w, background = background, viewer = NULL) url <- paste0("file:///", gsub("\\", "/", normalizePath(path))) webshot(url, file = file, selector = ".formattable_widget", delay = delay) }

export_formattable(testFormattable, 'testFormattable.jpg')

img2 <- readJPEG('testFormattable.jpg') img2 = abind::abind(img2, img2[,,1]) img2[,,4] = 1

h <-dim(img2)[1] # image height w <-dim(img2)[2] # image width

plotTableTopTen <- ggplot() + annotation_custom(grid::rasterGrob(img2, width=unit(1,"npc"), height=unit(1,"npc")), 0, w, 0, -h) + scale_x_continuous(expand=c(0,0),limits=c(0,w)) + scale_y_reverse(expand=c(0,0),limits=c(h,0)) + coord_equal() + stat_bin2d(binwidth=2,aes(fill = ..density..)) + theme_void()

plotTableTopTen neighborhoodsManhattanExclParksTopTenRankingPlot

grid.arrange(neighborhoodsManhattanExclParksTopTenRankingPlot, plotTableTopTen, ncol = 2)

filter(spc_common %in% c(treesTop10SpeciesVector$spc_common)) %>%

treesTop10SpeciesVector <- trees %>% filter(spc_common %in% c(treesTop10Ranking$spc_common)) %>% group_by(spc_common) %>% summarise(n = n()) %>% arrange(desc(n)) %>% mutate(nPerc = round(n/sum(n)*100,2), cumSum = cumsum(n), cumSumPerc = round(cumSum/sum(n)*100,2)) %>% ungroup() %>% head(n = 10) %>% drop_na() %>% as.data.frame()

dfTest <- trees %>% filter(spc_common %in% c(treesTop10SpeciesVector$spc_common)) %>% left_join(neighborhoods, by = c('nta' = 'ntacode')) %>% select(ntaname) %>% arrange(desc(ntaname)) %>% unique() %>% mutate(lp = row_number())

dfTest$lp <- rep(c('brown', 'darkgreen'), times = 28/2)

trees %>% filter(spc_common %in% c(treesTop10SpeciesVector$spc_common)) %>% left_join(neighborhoods, by = c('nta' = 'ntacode')) %>% select(ntaname,spc_common, boroname) %>% group_by(ntaname,spc_common) %>% summarize(n = n()) %>% mutate(pct = round(n / sum(n),2) * 100) %>% ungroup %>% complete(ntaname,spc_common, fill = list(stat = 0)) %>% replace(is.na(.), 0) %>% mutate(ntaname = factor(ntaname, unique(ntaname)), spc_common = factor(spc_common, unique(spc_common))) %>% mutate(xstart2 = row_number() - 0.5, xend2 = row_number()+0.5) %>% #mutate(ystart2 = col_number() - 0.5, yend2 = col_number()+0.5) %>% left_join(treesTop10SpeciesVector, by = c('spc_common' = 'spc_common')) %>% left_join(dfTest, by = c('ntaname' = 'ntaname')) %>% ggplot(aes(x = reorder(spc_common,-nPerc), y = ntaname, size = pct, group = spc_common,
color = ntaname, label = paste0(pct,'%'), )) + #scale_size(range = c(1, 10)) + geom_point(position = position_dodge(width = 0.5), alpha = 0.5) + geom_text(size = 3, position = position_dodge(width = 0.5),show.legend = FALSE, size = 4, hjust = -1, vjust= 0.3) + scale_colour_manual(values = rep(c('black'), times = 28)) + geom_rect(aes(xmin = 1-0.5, xmax = 10+0.5, ymin = xstart2/10+0.5, ymax = xend2/10+0.5, fill = lp), alpha = 0.20, col = NA) + #scale_fill_manual(values = alpha(c("red", "green"))) + theme_minimal() + theme(axis.text.x = element_text(angle = 90, vjust = 1.1, hjust=1, size = 8), axis.text.y = element_text(size = 8)) + guides(size = FALSE) + theme(legend.position = "none") + theme(panel.grid.major = element_blank(), panel.grid.minor = element_blank())

TOP TREES LOCATION vs SPECIES

treesTop10SpeciesVector <- trees %>% filter(spc_common %in% c(treesTop10Ranking$spc_common)) %>% group_by(spc_common) %>% summarise(n = n()) %>% arrange(desc(n)) %>% mutate(nPerc = round(n/sum(n)*100,2), cumSum = cumsum(n), cumSumPerc = round(cumSum/sum(n)*100,2)) %>% ungroup() %>% head(n = 10) %>% drop_na() %>% as.data.frame()

treesTop10Species <- trees %>% filter(spc_common %in% c(treesTop10SpeciesVector$spc_common)) %>% left_join(neighborhoods, by = c('nta' = 'ntacode')) %>% left_join(treesTop10SpeciesVector, by = c('spc_common' = 'spc_common')) %>% select(spc_common, ntaname, boroname, n) %>% filter(boroname == 'Manhattan') %>% #head() %>% ggplot(aes(x = reorder(spc_common,n), y = ntaname)) + geom_jitter(alpha=0.1, size = 0.5, width = 0.10, height = 0.20) + theme_minimal() + theme(axis.text.x = element_text(angle = 90, vjust = 1.1, hjust=1, size = 8), axis.text.y = element_text(size = 8) )

trees %>% filter(spc_common %in% c(treesTop10SpeciesVector$spc_common)) %>% left_join(neighborhoods, by = c('nta' = 'ntacode')) %>% select(spc_common, ntaname, boroname) %>% group_by(spc_common, ntaname) %>% summarize(n = n()) %>% mutate(pct = round(n / sum(n),2) * 100) %>% ungroup %>% complete(ntaname,spc_common, fill = list(stat = 0)) %>% replace(is.na(.), 0) %>% left_join(treesTop10SpeciesVector, by = c('spc_common' = 'spc_common')) %>% ggplot(aes(reorder(x = spc_common,-nPerc), y = ntaname, size = pct, group = spc_common, color = spc_common, label = paste0(pct,'%') )) + #scale_size(range = c(1, 10)) + geom_point(position = position_dodge(width = 0.5), alpha = 0.7 ) + geom_text(size = 3, position = position_dodge(width = 0.5), show.legend = FALSE, size = 4, hjust = -1) + theme_minimal() + theme(axis.text.x = element_text(angle = 90, vjust = 1.1, hjust=1, size = 8), axis.title = element_blank(), axis.text.y = element_text(size = 8)) + guides(size = FALSE) + theme(legend.position = "none") + scale_colour_manual(values=c("Honeylocust" = "brown", "Ginkgo" = "darkgreen", "Pin Oak" = "brown",
"Sophora" = "darkgreen",
"London Planetree" = "brown", "American Elm" = "darkgreen", "Willow Oak" = "brown", "Green Ash" = "darkgreen", "Crab Apple" = "brown", "Golden Raintree" = "darkgreen"))

TREES

glimpse(trees)

neighborhoodsManhattan <- neighborhoods %>% filter(boroname == 'Manhattan')

neighborhoodsManhattanExclParks <- neighborhoodsManhattan %>% filter(ntacode != 'MN99')

centroidsManhattan = st_centroid(neighborhoodsManhattanExclParks)

treesSP <- st_as_sf(trees, coords = c('longitude','latitude'), crs = st_crs(neighborhoodsManhattan)) treesSPTest <- head(treesSP, n = 100, crs = st_crs(neighborhoodsManhattan))

st_crs(neighborhoodsManhattan) == st_crs(treesSP)

neighborhoodsManhattanExclParksTreesPlot <- ggplot() + geom_sf(data = treesSP, lwd = 0, aes(color = factor(status))) + scale_color_manual(values = c("green", "red")) +

theme(legend.key.size = unit(3,"point")) +

guides(color = guide_legend(override.aes = list(size = 2))) + geom_sf(data = neighborhoodsManhattanExclParks, lwd = 0, alpha = 0.3) + theme_void()

neighborhoodsManhattanExclParks$area_sqkm <- st_area(neighborhoodsManhattanExclParks)

neighborhoodsManhattanAreas <- neighborhoodsManhattanExclParks%>% select(ntacode, area_sqkm) %>% st_drop_geometry() %>% mutate(area_sqkm = as.numeric(area_sqkm)/10^6)

manhattanTrees <- trees %>% select(nta,nta_name) %>% group_by(nta, nta_name) %>% summarise(n = n()) %>% ungroup() %>% left_join(neighborhoodsManhattanAreas, by = c('nta' = 'ntacode')) %>% mutate(treesPerAreaUnit = n / area_sqkm) %>% as.data.frame()

neighborhoodsManhattanExclParksTrees <- neighborhoodsManhattanExclParks %>% left_join(manhattanTrees, by = c('ntacode'='nta'))

neighborhoodsManhattanExclParksTreesPerKmPlot <- ggplot() + geom_sf(data = neighborhoodsManhattanExclParksTrees, aes(fill = treesPerAreaUnit), colour = "darkgrey" ) + theme_void() + theme(legend.position = "none") + theme(legend.position="left", legend.key.size = unit(0.5, 'cm'), legend.title = element_text(size=8)) + scale_fill_gradientn(colors = brewer.pal(n = 9 , name = "Greens"))

grid.arrange(neighborhoodsManhattanExclParksTreesPerKmPlot, neighborhoodsManhattanExclParksTreesPlot, ncol = 2)

DATACAMP

neighborhoodsManhattan <- neighborhoods %>% filter(boroname == 'Manhattan')

neighborhoodsManhattanExclParks <- neighborhoodsManhattan %>% filter(ntacode != 'MN99')

centroidsManhattan = st_centroid(neighborhoodsManhattanExclParks)

Boroughs

neighborhoodsBoroughs <- neighborhoods %>% group_by(boroname) %>% summarise(abc = sum(as.numeric(shape_leng)))

centroids = st_centroid(neighborhoodsBoroughs)

centroids <- centroids %>% mutate(geometry1 = gsub('[()°c\]', '', geometry)) %>% separate(col = geometry1, into = c('Latitude', 'Longitude'), sep = '\, ') %>% mutate(Latitude = as.numeric(Latitude), Longitude = as.numeric(Longitude)) %>% as.data.frame()

"#A84268"

plotBoroughs <- ggplot() + geom_sf(data = neighborhoodsBoroughs, aes(fill = as.numeric(abc)), lwd = 0, colour = "white" ) + theme_void() + theme(legend.position = "none") + scale_fill_gradientn( colors = c("lightgray")) + #geom_point(data = centroids, aes(x = Latitude, y = Longitude)) + geom_text(label = centroidsLatitude, y = centroids$Longitude), size = 3) + geom_rect(mapping= aes( xmin = st_bbox(neighborhoodsManhattan)[1], ymin = st_bbox(neighborhoodsManhattan)[2], xmax = st_bbox(neighborhoodsManhattan)[3], ymax = st_bbox(neighborhoodsManhattan)[4]), fill = NA, colour = "black", size = 0.9 )

plotBoroughs

neighborhoodsManhattan <- neighborhoods %>% filter(boroname == 'Manhattan')

neighborhoodsManhattanExclParks <- neighborhoodsManhattan %>% filter(ntacode != 'MN99')

centroidsManhattan = st_centroid(neighborhoodsManhattanExclParks)

centroidsManhattan <- centroidsManhattan %>% mutate(geometry1 = gsub('[()°c\]', '', geometry)) %>% separate(col = geometry1, into = c('Latitude', 'Longitude'), sep = '\, ') %>% mutate(Latitude = as.numeric(Latitude), Longitude = as.numeric(Longitude)) %>% as.data.frame()

centroidsManhattanntaname) centroidsManhattanname)

"#A84268"

plotManhattan <- ggplot() + geom_sf(data = neighborhoodsManhattanExclParks, aes(fill = as.numeric(shape_area)), colour = "white" ) + theme_void() + theme(legend.position = "none") + scale_fill_gradientn( colors = c("#9DBF9E", "#FCB97D")) + #geom_point(data = centroidsManhattan, aes(x = Latitude, y = Longitude)) + geom_text(label = centroidsManhattanLatitude, y = centroidsManhattan$Longitude), size = 1.8) + geom_rect(mapping= aes( xmin = st_bbox(neighborhoodsManhattan)[1], ymin = st_bbox(neighborhoodsManhattan)[2], xmax = st_bbox(neighborhoodsManhattan)[3], ymax = st_bbox(neighborhoodsManhattan)[4]), fill = NA, colour = "black", size = 0.9 )

plotManhattan

neighborhoods$area_sqkm <- st_area(neighborhoods)

neighborhoodsarea_sqkm)

ggplot() + geom_sf(data = neighborhoodsManhattanExclParks, lwd = 0, alpha = 0.3) + theme_void()+ geom_sf(data = subset(neighborhoods, ntaname == 'East Village'), size = 0.05, color = "green") + geom_sf(data = subset(neighborhoods, ntaname == 'Manhattanville'), size = 0.05, color = "red")

neighborhoods %>% filter(ntaname %in% c('East Village','Manhattanville'))

ggplot() + geom_sf(data = neighborhoodsManhattan, lwd = 0, )

trees %>% filter(status == 'Dead') %>% ggplot() + geom_point(aes(x = latitude, y = longitude, color = status), size = 0.3, alpha= 0.3) + theme_minimal() + geom_sf(data = neighborhoodsManhattan, aes(fill = as.numeric(shape_area)), lwd = 0, colour = "white" ) + theme_void() + theme(legend.position = "none") + scale_fill_gradientn( colors = c("#9DBF9E", "#FCB97D", "#b8b8b8")) + #geom_point(data = centroidsManhattan, aes(x = Latitude, y = Longitude)) + geom_text(label = centroidsManhattanLatitude, y = centroidsManhattan$Longitude), size = 1.8) + geom_rect(mapping= aes( xmin = st_bbox(neighborhoodsManhattan)[1], ymin = st_bbox(neighborhoodsManhattan)[2], xmax = st_bbox(neighborhoodsManhattan)[3], ymax = st_bbox(neighborhoodsManhattan)[4]), fill = NA, colour = "black", size = 0.6 )

install.packages('formattable') library(formattable)

formattable(i1, align =c("l","c","c","c","c", "c", "c", "c", "r"), list(Indicator Name = formatter("span", style = ~ style(color = "grey",font.weight = "bold")), 2011= color_tile(customGreen, customGreen0), 2012= color_tile(customGreen, customGreen0), 2013= color_tile(customGreen, customGreen0), 2014= color_tile(customGreen, customGreen0), 2015= color_tile(customGreen, customGreen0), 2016= color_tile(customGreen, customGreen0), Average = color_bar(customRed) ))

Manhattan trees density

neighborhoodsManhattan <- neighborhoods %>% filter(boroname == 'Manhattan')

neighborhoodsManhattanExclParks <- neighborhoodsManhattan %>% filter(ntacode != 'MN99')

centroidsManhattan = st_centroid(neighborhoodsManhattanExclParks)

centroidsManhattan <- centroidsManhattan %>% mutate(geometry1 = gsub('[()°c\]', '', geometry)) %>% separate(col = geometry1, into = c('Latitude', 'Longitude'), sep = '\, ') %>% mutate(Latitude = as.numeric(Latitude), Longitude = as.numeric(Longitude)) %>% as.data.frame()

centroidsManhattanntaname) centroidsManhattanname)

treesSP <- st_as_sf(trees, coords = c('longitude','latitude'), crs = st_crs(neighborhoodsManhattan)) treesSPTest <- head(treesSP, n = 100, crs = st_crs(neighborhoodsManhattan))

st_crs(neighborhoodsManhattan) == st_crs(treesSP)

neighborhoodsManhattanExclParks$area_sqkm <- st_area(neighborhoodsManhattanExclParks)

neighborhoodsManhattanAreas <- neighborhoodsManhattanExclParks%>% select(ntacode, area_sqkm) %>% st_drop_geometry() %>% mutate(area_sqkm = as.numeric(area_sqkm)/10^6)

manhattanTrees <- trees %>% select(nta,nta_name) %>% group_by(nta, nta_name) %>% summarise(n = n()) %>% ungroup() %>% left_join(neighborhoodsManhattanAreas, by = c('nta' = 'ntacode')) %>% mutate(treesPerAreaUnit = n / area_sqkm) %>% as.data.frame()

customGreen = brewer.pal(n = 9 , name = "Greens")[1] customGreen0 = brewer.pal(n = 9 , name = "Greens")[8] customBlue = brewer.pal(n = 9 , name = "Blues")[1] customBlue0 = brewer.pal(n = 9 , name = "Blues")[6]

neighborhoodsManhattanExclParksTrees <- neighborhoodsManhattanExclParks %>% left_join(manhattanTrees, by = c('ntacode'='nta'))

neighborhoodsManhattanExclParksTreesPerKmPlot <- ggplot() + geom_sf(data = neighborhoodsManhattanExclParksTrees, aes(), colour = "darkgrey") + theme_void() + theme(legend.position = "none") + theme(#legend.position="left", legend.key.size = unit(0.6, 'cm'), legend.position = c(+1.55, 0.4), legend.title = element_text(size=7), legend.text = element_text(size=7)) + guides(fill=guide_legend(title="trees per km^2")) + #scale_fill_gradientn(colors = brewer.pal(n = 9 , name = "Blues")[1:8]) + geom_text(label = centroidsManhattanLatitude, y = centroidsManhattan$Longitude), size = 3)

################### SUMMARY ###################

1.) Number of trees - map + backgroung

plotBackgroundWithMap

2.) Trees per km2 - map and table

grid.arrange(neighborhoodsManhattanExclParksTreesPerKmPlot, manhattanTreesFormattablePlot, ncol = 2)

3 Most common trees

3a.) Most common trees - chart

treesTop20

3b.) Most common trees - top 10 trees vs nta - chart

mostCommonTreesDist

4.) Curb location - ON / OFF CURB

grid.arrange(treesCurb, onCurbTreesMapPlot, ncol = 2)

5.) Health - All trees

treesHealtPlot

6.) Health - Map of ntas

grid.arrange(neighborhoodsManhattanDict, plotManhattanFun("Good", "Greens", neighborhoodsManhattanExclParksFair, '% of trees with fair health'), plotManhattanFun("Poor", "Reds", neighborhoodsManhattanExclParks$Poor, '% of trees with poor health'), ncol = 4)

7.) Trunk

grid.arrange(treesTop10BoxPlot, treesTop10Median, ncol = 2)

8.) Recommended trees - table with ratings

treesTop10RankingFormattable

9.) Appendix -dead trees

grid.arrange(arrangeGrob(plot1, plot2, plot3, ncol = 3), heights=c(20, 1),top = textGrob("Fig_6: Percentage of dead trees",gp=gpar(col = '#576161',fontsize=10,font=3)))

8 Health - map of nta (add legend)

9 Trunk

11 Recommended trees - map

12 Recommended trees - details table

formattable(manhattanTreesTable, align =c("l","l","l"), list('nta' = costum_format3, 'ntaname' = costum_format2, treesPerKm2 = image_tile ))