stats_sf can't find geometry aesthetics

rodri45z · December 11, 2020, 7:03pm

Hi, I'm trying to gganimate a map, I'm able to do a simple map using renewable energy data from the world bank but when in order to animate it I need to add the years variable in the dataframe, first I join the dataframes (geometry and data)with the years as the name for each column with the data and then I make a pivot longer to create one Value for each year and country. But after doing this when I try to plot the map with geom_sf, I get the following error: "stat_sf requires the following missing aesthetics: geometry" . Any idea? what am I doing wrong?

Here is my code :

 library(tidyverse)
library(rio)
library(plotly)
library(janitor)
library(gganimate)
library(viridis)
library("sf")
library(tidyverse)
library(rio)
url <-("https://raw.githubusercontent.com/rodri45l/plasticwaste/main/trabajo_BigData_equipo/datos/renpwr.csv")

df <- import(url)

df_pwr<- df %>% select(-c(`Series Name`,`Series Code`))  %>% rename(sovereignt = `Country Name`,
                                                                                                `1990` = `1990 [YR1990]`,
                                                                                                `1991`=`1991 [YR1991]`,
                                                                                                `1992`=`1992 [YR1992]`,
                                                                                                `1993`=`1993 [YR1993]`,
                                                                                                `1994`=`1994 [YR1994]`,
                                                                                                `1995`=`1995 [YR1995]`,
                                                                                                `1996`=`1996 [YR1996]`,
                                                                                                `1997`=`1997 [YR1997]`,
                                                                                                `1998`=`1998 [YR1998]`,
                                                                                                `1999`=`1999 [YR1999]`,
                                                                                                `2000`=`2000 [YR2000]`,
                                                                                                `2001`=`2001 [YR2001]`,
                                                                                                `2002`=`2002 [YR2002]`,
                                                                                                `2003`=`2003 [YR2003]`,
                                                                                                `2004`=`2004 [YR2004]`,
                                                                                                `2005`=`2005 [YR2005]`,
                                                                                                `2006`=`2006 [YR2006]`,
                                                                                                `2007`=`2007 [YR2007]`,
                                                                                                `2008`=`2008 [YR2008]`,
                                                                                                `2009`=`2009 [YR2009]`,         
                                                                                                `2010`=`2010 [YR2010]`,
                                                                                                `2011`=`2011 [YR2011]`,
                                                                                                `2012`=`2012 [YR2012]`,
                                                                                                `2013`=`2013 [YR2013]`,
                                                                                                `2014`=`2014 [YR2014]` ,  
                                                                                                `2015`=`2015 [YR2015]`
                                                                                                
) 




df_pwr2 <- df_pwr %>%  pivot_longer(cols = 3:28, names_to = "year") %>% transform(year = as.numeric(year),
                                                                                  value = as.numeric(value)
) %>% filter(year == 2015)






world <- rnaturalearth::ne_countries(scale = "medium", returnclass = "sf")
world_points<- st_centroid(world)
world_points <- cbind(world, st_coordinates(st_centroid(world$geometry)))










df_pwr2$sovereignt <- recode(df_pwr2$sovereignt, 
                             "United States"="United States of America", 
                             "Russian Federation" = "Russia",
                             "Venezuela, RB" = "Venezuela",
                             "Congo, Dem. Rep." = "Democratic Republic of the Congo" ,
                             "	Congo, Rep."="Republic of Congo"	 ,
                             "Egypt, Arab Rep." = "Egypt" ,
                             "Tanzania" = "United Republic of Tanzania",
                             "Cote d'Ivoire" = "Ivory Coast",
                             "Yemen, Rep." = "Yemen",
                             "Iran, Islamic Rep." = "Iran",
                             "Syrian Arab Republic" = "Syria",
                             "Slovak Republic" = "Slovakia",
                             "Serbia" = "Republic of Serbia",
                             "North Macedonia" = "Macedonia",
                             "Kyrgyz Republic" = "Kyrgyzstan",
                             "Korea, Dem. People’s Rep" = "North Korea",
                             "Korea, Rep." = "South Korea",
                             "Lao PDR" = "Laos")


df_inner_pwr <- merge(world,df_pwr2,by="sovereignt",all=TRUE)


#First map
p <- ggplot(data = df_inner_pwr) + geom_sf() +
  labs(title = "Gráfico 1: Mapa del mundo energias renovables",
       caption = "Datos provenientes de World Bank")


p+ geom_sf(aes(fill = value, color =value)) + scale_fill_viridis(direction = 1, label = scales::dollar,begin = 0, end = 0.9) + 
  scale_color_viridis(direction = 1, label = scales::dollar,begin = 0, end = 0.9) 


#Trying to gganimate the map



df_pwr$sovereignt <- recode(df_pwr$sovereignt, 
                            "United States"="United States of America", 
                            "Russian Federation" = "Russia",
                            "Venezuela, RB" = "Venezuela",
                            "Congo, Dem. Rep." = "Democratic Republic of the Congo" ,
                            "	Congo, Rep."="Republic of Congo"	 ,
                            "Egypt, Arab Rep." = "Egypt" ,
                            "Tanzania" = "United Republic of Tanzania",
                            "Cote d'Ivoire" = "Ivory Coast",
                            "Yemen, Rep." = "Yemen",
                            "Iran, Islamic Rep." = "Iran",
                            "Syrian Arab Republic" = "Syria",
                            "Slovak Republic" = "Slovakia",
                            "Serbia" = "Republic of Serbia",
                            "North Macedonia" = "Macedonia",
                            "Kyrgyz Republic" = "Kyrgyzstan",
                            "Korea, Dem. People’s Rep" = "North Korea",
                            "Korea, Rep." = "South Korea",
                            "Lao PDR" = "Laos")
df_pwr4 <- merge(world,df_pwr,by="sovereignt",all=TRUE)
df_pwr6 <- df_pwr4 %>%  pivot_longer(cols = 65:90, names_to = "year") %>% transform(year = as.numeric(year),
                                                                                    value = as.numeric(value)) 





o <- ggplot(data = df_pwr6) + geom_sf(data = df_pwr6, aes(fill = value)) +
  labs(title = "Gráfico 1: Mapa del mundo energias renovables",
       caption = "Datos provenientes de World Bank")


o + geom_sf(aes(fill = value, color =value)) + scale_fill_viridis(direction = 1, label = scales::number,begin = 0, end = 0.9) + 
  scale_color_viridis(direction = 1, label = scales::number,begin = 0, end = 0.9) + transition_reveal(year) +
  labs(title = "year: {as.integer(frame_along)}")

jlacko · December 11, 2020, 9:26pm

Your code is complex and not quite easy to troubleshoot; it might be a good idea to reframe the issue you are facing in as simple format as possible. What you share is quite a chunk...

For a simple example of spatial data frame animated via {gganimate} consider this post; it builds on the pressure dataset that is shipped with base R and the North Carolina shapefile that comes with {sf}.

Best packages for making map? leaflet vs ggmap vs sf vs ...? General

I think that ggplot and leaflet will handle most of your needs. A honorable mention goes to tmap package - https://github.com/mtennekes/tmap. It introduces yet another map making syntax, and produces decent maps with very little tweaking. It supports both static and dynamic output. But competing against ggplot in static maps is hard, and in dynamic mapping using leaflet directly gives you more control. If you are familiar with ggplot syntax I would leave it out. And as my curiosity got the better of me I propose a workflow for animated maps, built on ggplot and gganimate. The hardest part was faking the data to be animated; at the end I ended up recycling the well known & loved pressure d…

rodri45z · December 12, 2020, 12:30pm

@jlacko Thanks for answering; my bad, didn't think my code was so big or complex. I guess I'll have to re think the way to do it. Your example helped me a lot. Thanks again.

jlacko · December 12, 2020, 9:11pm

It is not a huge piece of code, but a sizable one; most of seems not directly relevant to your issue (animating a geom_sf). You also import a number of libraries, listing some of them twice. This can be tricky - for example I don't have {rio} installed and so I am unable to run your code as it stands now.

In general it is a good idea when you run into a problem with a piece of code to simplify it into a minimal reproducible example - focusing only on the tricky part, and cutting out the chaff.

I know it is harder than it sounds, but by making it easier for others to help you you are increasing your chances for a quality answer.

jlacko · December 12, 2020, 9:55pm

On second thought - I have managed to install {rio} - it seems that the immediate cause of failure of your geom_sf() call is due to class of your df_pwr6 object being "only" data.frame.

It has lost its special sf sauce. Likely this is a result of your pivot_longer() call, when you select only columns 65 to 90, omitting the special geometry column.

rodri45z · December 13, 2020, 7:56pm

I managed to fix it, the problem was gganimate doesn't recognize geometry when it's not in the last column, so just add geometry = geometry in aes().

geom_sf(data = .... , aes(geometry = geometry, .... ))

system · December 20, 2020, 7:56pm

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.