#######################################################################################
#### Script for: A global database of soil seed bank richness, density and abundance ##
#### Alistair Auffret et al. ##########################################################
#######################################################################################

## This script will open the global seed bank database, producing the values and information cited in the text and tables of the data descriptor.

# Bring in data
sb<-read.csv("gsb_db.csv")


############
# Abstract

nrow(sb) # 3096 records
length(unique(paste(sb$Lat_deg, sb$Lon_deg))) # 1929 locations
length(unique(sb$StudyID)) # 1442 studies
range(sb$Year) # 1940 - 2020.

############
# Intro
nrow(sb) # 3096 records
length(unique(sb$StudyID)) # 1442 studies
sum(sb$Total_number_samples, na.rm=TRUE) # 1 081 363
sum(sb$Number_sites, na.rm=TRUE) # 15 899 sites 
length(unique(sb$Country)) # 94 countries


############
# Data entry
sum(sb$Sample_depth_mm>=30, na.rm=TRUE)/sum(!is.na(sb$Sample_depth_mm)) # 94% of records at least 30 mm
mean(sb$Sample_depth_mm, na.rm=TRUE) # mean sample depth 93 mm
median(sb$Sample_depth_mm, na.rm=TRUE) # median sample depth 100 mm


############
# Database structure
dim(sb) # 3096 records, 41 columns

# Table 1 (Database description)
names(sb) # column names

# Table 2 (Biomes) - names and number of records
table(sb$Biome)
table(sb$Biome_broad)
table(sb$Biome_zone)

# Table 3 (Habitats) - names and number of records per category
table(paste(sb$Habitat_current, sb$Habitat_target))


#############
# Figure 1
library(maps) # a package with simple maps 
transp<-function(col,alpha){ # a simple function to allow transparent plottning
	trans.col<-adjustcolor(col,alpha.f = alpha)
	return(trans.col)} 

# plot it
map('world', col="lightgrey", fill=TRUE, lty=0, resolution=0)
points(Lat_deg~Lon_deg, data=sb[sb$Habitat_current=="Arable",], col=transp("gold",0.5), pch=16, cex=0.5,lwd=0.5)
points(Lat_deg~Lon_deg, data=sb[sb$Habitat_current=="Forest",], col=transp("forestgreen",0.5), pch=16, cex=0.5,lwd=0.5)
points(Lat_deg~Lon_deg, data=sb[sb$Habitat_current=="Grassland",], col=transp("darkseagreen1",0.5), pch=16, cex=0.5,lwd=0.5) 
points(Lat_deg~Lon_deg, data=sb[sb$Habitat_current=="Wetland",], col=transp("skyblue3",0.5), pch=16, cex=0.5,lwd=0.5) 
points(Lat_deg~Lon_deg, data=sb[sb$Habitat_current=="Aquatic",],col=transp("navyblue",0.5), pch=16, cex=0.5,lwd=0.5)

legend(-150,-20,c("Arable","Forest","Grassland","Wetland", "Aquatic"),pch=16,cex=0.75,col=c("gold", "forestgreen","darkseagreen1", "skyblue3","navyblue"),bty="n", pt.lwd=0.6)


###############
# Create list of references of studies included in database

sb$studylong<-paste(sb$Authors, sb$Year, sb$Title, sb$Journal, paste("doi:",sb$Doi),sep=", ")
#cat(sort(unique(sb$studylong)),sep="\n\n", file="component_refs.txt") # write to file