Taking an example from the network

library (vegan)

veg.data <- read.delim ('https://raw.githubusercontent.com/zdealveindy/anadat-r/master/data/vltava-spe.txt', row.names = 1)
env.data <- read.delim ('https://raw.githubusercontent.com/zdealveindy/anadat-r/master/data/vltava-env.txt')

DCA <- decorana (veg = log1p (veg.data))

ordiplot (DCA, display = 'sites', type = 'n')
points (DCA, 
        pch = env.data$GROUP,
        col = c("orange", "green", "lightblue", "mediumorchid")[env.data$GROUP]
        )

Exit:

Explanation:

We want to configure the colors based on the variable env.data$GROUP , but we do not want any color but certain in particular, this variable has these data:

[1] 1 2 3 4
> length(env.data$GROUP)
[1] 97
> env.data$GROUP
 [1] 1 1 1 1 2 2 3 3 1 1 2 1 3 3 2 3 1 1 1 3 1 2 3 1 1 4 3 3 3 2 2 3 3 3 3 3 2 1 1
[40] 1 1 1 1 2 1 1 1 3 2 3 3 3 3 3 2 4 4 4 4 4 4 4 4 2 4 4 4 4 4 1 1 3 1 1 4 1 1 2
[79] 4 4 4 2 4 4 1 2 1 1 1 3 1 3 3 1 1 1 1

That is, there are 4 groups divided into 97 observations, what we are going to do is give each observation a particular color and we do it with a simple indexing operation, on the vector of colors:

> head(c("orange", "green", "lightblue", "mediumorchid")[env.data$GROUP], 20)
 [1] "orange"    "orange"    "orange"    "orange"    "green"     "green"    
 [7] "lightblue" "lightblue" "orange"    "orange"    "green"     "orange"   
[13] "lightblue" "lightblue" "green"     "lightblue" "orange"    "orange"   
[19] "orange"    "lightblue"

This we can do it this simple because env.data$GROUP is numeric and it is already useful to make the selection, in your case:

> head(raw.site.hab,20)
 [1] "Crop"      "Crop"      "Crop"      "Crop"      "Wasteland" "Wasteland"
 [7] "Wasteland" "Wasteland" "Wasteland" "Wasteland" "Wasteland" "Wasteland"
[13] "Wasteland" "Wasteland" "Wasteland" "Wasteland" "Wasteland" "Wasteland"
[19] "Wasteland" "Wasteland"
> head(c("orange", "green", "lightblue", "mediumorchid")[as.integer(as.factor(raw.site.hab))], 20)
 [1] "orange"       "orange"       "orange"       "orange"       "mediumorchid"
 [6] "mediumorchid" "mediumorchid" "mediumorchid" "mediumorchid" "mediumorchid"
[11] "mediumorchid" "mediumorchid" "mediumorchid" "mediumorchid" "mediumorchid"
[16] "mediumorchid" "mediumorchid" "mediumorchid" "mediumorchid" "mediumorchid"

Values are strings, so we simply convert them to factor and use the numerical index of each value to replace it with the color. What would remain is to pass this vector to points by means of the col parameter.

An alternative to the response of Patricio Moracho is to extract the scores of the object decorana , pass them to data.frame and make the graph with ggplot. The "disadvantage" is that you have to explore the object and find the coefficients there. The advantages are:

Extract the scores of an object decorana

  

I recycle part of Patricio's response.

library (vegan)

veg.data <- read.delim ('https://raw.githubusercontent.com/zdealveindy/anadat-r/master/data/vltava-spe.txt', row.names = 1)
env.data <- read.delim ('https://raw.githubusercontent.com/zdealveindy/anadat-r/master/data/vltava-env.txt')

DCA <- decorana (veg = log1p (veg.data))

With the DCA object ready the simplest way to draw the coordinates is to take them from the summary:

sumario_DCA <- summary(DCA)

#Miras la estructura del objeto para ver dónde está la información que te interesa.
# En Rstudio podrías usar View(DCA)

str(sumario_DCA)

sumario_DCA is a list and at the location sumario_DCA$site.scores are the site coordinates. Now everything goes in the pipeline, although you could do it step by step.

Generate the graphic

sumario_DCA$site.scores %>%               #Esto es una matriz
  as.data.frame() %>%                     #Que aquí paso a data.frame
  mutate(grupo = env.data$GROUP,          #Agrego la información de grupo del data.frame con los datos. En ambos casos el largo es 97 y el orden el mismo.
         grupo = as.factor(grupo)) %>%    #Paso a factor para que ggplot haga una escala discreta.
  ggplot(aes(x = DCA1, 
             y = DCA2, 
             color = grupo, 
             shape = grupo)) + #por si va a imprenta en blanco y negro. Redundante.
  geom_point() +                                   #Agrego los puntos
#Personalizo el gráfico. No es estrictamente necesario.
  theme_minimal() + 
  labs(title = "Detrended Correspondence Analisys", 
       subtitle = "It's just CA for biologists", 
       caption = "Elaboración propia
                  Datos David Zelený")

Result