Updated selection, based on sPlot3.0_v1.1

01_Extract_data_Project_31.Rmd

@@ -63,7 +63,7 @@ Extract all plots containing at least one species in the xylem list.
 ```{r, message=F, results=F, warning=F}
 species_list <- xylem_data$Species
 plot.sel <- DT2 %>%
-  filter(DT2$species %in% species_list) %>%
+  filter(DT2$Species %in% species_list) %>%
   dplyr::select(PlotObservationID) %>%
   distinct() %>%
   pull(PlotObservationID)
@@ -76,19 +76,19 @@ header.xylem <- header %>%
 plot.sel <- header.xylem$PlotObservationID
 
 DT.xylem <- DT2 %>% 
-  filter(taxon_group %in% c("Vascular plant", "Unknown")) %>% 
+  filter(Taxon_group %in% c("Vascular plant", "Unknown")) %>% 
   filter(PlotObservationID %in% plot.sel)
 
 ```
 
-Out of the `r length(species_list)` species in the sRoot list, `r sum(unique(DT2$species) %in% species_list)` species are present in sPlot, for a total of `r nrow(DT.xylem %>% filter(species %in% species_list))` records, across `r length(plot.sel)` plots.
+Out of the `r length(species_list)` species in the sRoot list, `r sum(unique(DT2$Species) %in% species_list)` species are present in sPlot, for a total of `r nrow(DT.xylem %>% filter(Species %in% species_list))` records, across `r length(plot.sel)` plots.
 
 # 2 Extract woody species
 This is partial selection, as we don't have information on the growth form of all species in sPlot
 ```{r}
 #Select all woody species and extract relevant traits from TRY
 woody_species_traits <- sPlot.traits %>%
-  dplyr::select(species, GrowthForm, is.tree.or.tall.shrub, n,
+  dplyr::select(Species, GrowthForm, is.tree.or.tall.shrub, n,
                 starts_with("StemDens"),
                 starts_with("Stem.cond.dens"), 
                 starts_with("StemConduitDiameter"),
@@ -97,7 +97,7 @@ woody_species_traits <- sPlot.traits %>%
                 starts_with("PlantHeight"), 
                 starts_with("Wood"), 
                 starts_with("SpecificRootLength_mean")) %>%
-  filter( (species %in% species_list) |
+  filter( (Species %in% species_list) |
           grepl(pattern = "tree|shrub", x = GrowthForm) |
           is.tree.or.tall.shrub==T
               ) %>% 
@@ -133,7 +133,7 @@ Codes correspond to those reported in [TRY](https://www.try-db.org/TryWeb/Home.p
 ```{r}
 #subset DT.xylem to only retain woody species
 DT.xylem <- DT.xylem %>% 
-  filter(species %in% (woody_species_traits$species))
+  filter(Species %in% (woody_species_traits$Species))
 nrow(DT.xylem)
 ```
 
@@ -149,23 +149,23 @@ combine.cover <- function(x){
 }
 
 DT.xylem <- DT.xylem %>%
-  dplyr::select(PlotObservationID, species,Layer, Relative.cover) %>%
+  dplyr::select(PlotObservationID, Species,Layer, Relative_cover) %>%
   # normalize relative cover to 1 for each plot x layer combination
   left_join({.} %>% 
               group_by(PlotObservationID, Layer) %>% 
-              summarize(Tot.Cover=sum(Relative.cover), .groups="drop"), 
+              summarize(Tot.Cover=sum(Relative_cover), .groups="drop"), 
             by=c("PlotObservationID", "Layer")) %>% 
-  mutate(Relative.cover=Relative.cover/Tot.Cover) %>% 
-  group_by(PlotObservationID, species) %>%
+  mutate(Relative_cover=Relative_cover/Tot.Cover) %>% 
+  group_by(PlotObservationID, Species) %>%
   # merge layers together
-  summarize(Relative.cover=combine.cover(Relative.cover), .groups="drop") %>%
+  summarize(Relative_cover=combine.cover(Relative_cover), .groups="drop") %>%
   ungroup() %>% 
   # normalize relative cover to 1 after merging layers together
   left_join({.} %>% 
               group_by(PlotObservationID) %>% 
-              summarize(Tot.Cover=sum(Relative.cover), .groups="drop"), 
+              summarize(Tot.Cover=sum(Relative_cover), .groups="drop"), 
             by="PlotObservationID") %>% 
-  mutate(Relative.cover=Relative.cover/Tot.Cover) 
+  mutate(Relative_cover=Relative_cover/Tot.Cover) 
 
 nrow(DT.xylem)
 ```
@@ -174,7 +174,7 @@ double check that covers are properly standardized
 DT.xylem %>% 
   filter(PlotObservationID %in% sample(header.xylem$PlotObservationID, 10, replace=F)) %>% 
   group_by(PlotObservationID) %>% 
-  summarize(tot.cover=sum(Relative.cover), .groups="drop")
+  summarize(tot.cover=sum(Relative_cover), .groups="drop")
 
 ```
 
@@ -185,24 +185,23 @@ Calculate CWM and trait coverage for each trait and each plot. Select plots havi
 # Merge species data table with traits
 CWM.xylem0 <- DT.xylem %>%
   as_tibble() %>%
-  dplyr::select(PlotObservationID, species, Relative.cover) %>%
+  dplyr::select(PlotObservationID, Species, Relative_cover) %>%
   left_join(xylem_data %>%
-              dplyr::rename(species=Species) %>%
-              dplyr::select(species, P50, Ks), 
-            by="species")
+              dplyr::select(Species, P50, Ks), 
+            by="Species")
 
 # Calculate CWM for each trait in each plot
 CWM.xylem1 <- CWM.xylem0 %>%
   group_by(PlotObservationID) %>%
   summarize_at(.vars= vars(P50:Ks),
-               .funs = list(~weighted.mean(., Relative.cover, na.rm=T)), 
+               .funs = list(~weighted.mean(., Relative_cover, na.rm=T)), 
                .groups="drop") %>%
   dplyr::select(PlotObservationID, order(colnames(.))) %>%
   pivot_longer(-PlotObservationID, names_to="trait", values_to="trait.value")
 
 # Calculate coverage for each trait in each plot
 CWM.xylem2 <- CWM.xylem0 %>%
-  mutate_at(.funs = list(~if_else(is.na(.),0,1) * Relative.cover), 
+  mutate_at(.funs = list(~if_else(is.na(.),0,1) * Relative_cover), 
             .vars = vars(P50:Ks)) %>%
   group_by(PlotObservationID) %>%
   summarize_at(.vars= vars(P50:Ks),
@@ -232,7 +231,7 @@ variance2.fun <- function(trait, abu){
 CWM.xylem3 <- CWM.xylem0 %>%
   group_by(PlotObservationID) %>%
   summarize_at(.vars= vars(P50:Ks),
-               .funs = list(~variance2.fun(., Relative.cover))) %>%
+               .funs = list(~variance2.fun(., Relative_cover))) %>%
   dplyr::select(PlotObservationID, order(colnames(.))) %>%
   pivot_longer(-PlotObservationID, names_to="trait", values_to="trait.variance")