1.1 Data frames

# 1. Traits data frame
sp_traits<-read.csv("./Data/species_traits.csv")

# 2. Biome distribution data frame
sp_biomes<-read.csv("./Data/species_biomes_distri.csv")

# 3. Growth form information
Growth_form<-read.csv("./Data/GrowthForm.csv")

1.2 Exploring data

# Exploring the dataframe
dim(sp_traits)

## [1] 100   7

dim(sp_biomes)

## [1] 400   2

# Alternative you can use View()
head(sp_traits)

##                 species Wood_density   Leaf_N      SLA   Seed_mass
## 1   Macleania rupestris    0.5391685 11.87024  5.23328    2.053831
## 2  Coleostephus myconis    0.7003753 24.43410 20.92570    0.320000
## 3 Guatteria oliviformis    0.5007774 23.54603 16.30982  127.082572
## 4      Myrsine africana    0.6390519 12.90000 11.62415   11.700000
## 5  Chrysophyllum viride    0.6985341 18.14058 12.88736 2000.000000
## 6    Ouratea hexasperma    0.7447541 15.80000  7.29000  147.924327
##       Height    Leaf_P
## 1  2.4750017 0.7073387
## 2  0.4907508 1.9359780
## 3 16.5535877 1.1148200
## 4 15.2099828 0.8675000
## 5 17.4286478 0.7873025
## 6  7.4563818 0.7630000

head(sp_biomes)

##                 species                Biome
## 1 Acacia_podalyriifolia         Moist_Forest
## 2           Acer_rubrum   Coniferous_Forests
## 3           Acer_rubrum             Savannas
## 4           Acer_rubrum                Taiga
## 5           Acer_rubrum Temperate_Grasslands
## 6           Acer_rubrum      Temperate_Mixed

# How many species are match between the two dataframes
table(sp_traits$species%in%unique(sp_biomes$species))

## 
## FALSE 
##   100

# We first need to make sure the species vector names have the same format
head(sp_traits$species)

## [1] Macleania rupestris   Coleostephus myconis  Guatteria oliviformis
## [4] Myrsine africana      Chrysophyllum viride  Ouratea hexasperma   
## 100 Levels: Acacia podalyriifolia Acer rubrum ... Yucca rostrata

head(sp_biomes$species)

## [1] Acacia_podalyriifolia Acer_rubrum           Acer_rubrum          
## [4] Acer_rubrum           Acer_rubrum           Acer_rubrum          
## 114 Levels: Abarema_curvicarpa Abies_vejarii ... Yucca_rostrata

1.3 Matching variables and formats

# Replacing spaces in species names
sp_traits$species<-gsub(" ","_",sp_traits$species)
head(sp_traits$species)

## [1] "Macleania_rupestris"   "Coleostephus_myconis"  "Guatteria_oliviformis"
## [4] "Myrsine_africana"      "Chrysophyllum_viride"  "Ouratea_hexasperma"

table(sp_traits$species%in%unique(sp_biomes$species))

## 
## FALSE  TRUE 
##     1    99

table(unique(sp_biomes$species)%in%sp_traits$species)

## 
## FALSE  TRUE 
##    15    99

1.4 Merging data frames

# ?merge
# When species columns have the same name
Trait_biomes_df<-merge(x=sp_traits,y=sp_biomes) 

head(Trait_biomes_df)

##                 species Wood_density   Leaf_N       SLA Seed_mass   Height
## 1 Acacia_podalyriifolia         0.68 26.72163  6.344215  32.65056 23.80193
## 2           Acer_rubrum         0.49 16.94147 20.501333  22.76456 36.99420
## 3           Acer_rubrum         0.49 16.94147 20.501333  22.76456 36.99420
## 4           Acer_rubrum         0.49 16.94147 20.501333  22.76456 36.99420
## 5           Acer_rubrum         0.49 16.94147 20.501333  22.76456 36.99420
## 6           Acer_rubrum         0.49 16.94147 20.501333  22.76456 36.99420
##      Leaf_P                Biome
## 1 0.7857099         Moist_Forest
## 2 1.1150000 Temperate_Grasslands
## 3 1.1150000   Coniferous_Forests
## 4 1.1150000                Taiga
## 5 1.1150000      Temperate_Mixed
## 6 1.1150000             Savannas

DF1<-merge(x=sp_traits,y=sp_biomes,
                       by.x="species",by.y="species")

DF2<-merge(x=sp_biomes,y=sp_traits,
                        by.x="species",by.y="species")

Is there a difference between data frame DF1 and DF2?

1.5 Challenge

Include the Growth form information into the Trait_biomes_df data frame. Remember to check first the format and names of the variables you want to match between the two data frames.

2.1 Let’s practice

Extracting the mean values of seed mass per family in Moist forest

Trait_biomes_df %>% 
  filter(Biome=="Moist_Forest") %>% 
  group_by(FAMILY_STD) %>% 
  summarise(Mean_Trait=mean(Seed_mass, na.rm=TRUE))

## # A tibble: 28 x 2
##    FAMILY_STD       Mean_Trait
##    <fct>                 <dbl>
##  1 Amaranthaceae        2.71  
##  2 Anacardiaceae       13.0   
##  3 Annonaceae        2242.    
##  4 Apiaceae             0.731 
##  5 Bignoniaceae        14.1   
##  6 Brassicaceae         0.339 
##  7 Burseraceae        269.    
##  8 Caricaceae           6.00  
##  9 Chrysobalanaceae  2425.    
## 10 Cyperaceae           0.0210
## # ... with 18 more rows

Biome_grasses<-function(biome){

  Biome_df<-
  Trait_biome_GF %>% 
    filter(Biome==____)
  
  if("Poaceae"%in%___){
   
    Biome_df %>% 
      filter(FAMILY_STD=="Poaceae") %>% 
      pull(species)
    
  }else{
    
    print(___)
    
  }  

}