APIs, time-series, and weather Data
Climate Metrics: ClimdEX
Indices representing extreme aspects of climate derived from daily data:
Climate Change Research Centre (CCRC) at University of New South Wales (UNSW) (climdex.org).
Weather Data
Climate Data Online
GHCN
NOAA API
National Climatic Data Center application programming interface (API).
Plot all stations:
ggplot(data=st_filtered,aes(y=latitude,x=longitude)) +
facet_grid(element~.)+ #make panels for each variable
geom_sf(data=world,inherit.aes = F,size=.1,fill="grey",colour="black")+
geom_point(size=.1,col="red")+
coord_sf()
It’s hard to see all the points, let’s bin them…
ggplot(st_filtered,aes(y=latitude,x=longitude)) +
geom_sf(data=world,inherit.aes = F,size=.1,fill="grey",colour="black")+
facet_grid(element~.)+
stat_bin2d(bins=100)+
scale_fill_distiller(palette="YlOrRd",trans="log",direction=-1,
breaks = c(1,10,100,1000))+
coord_sf()
Your turn
Produce a binned map (like above) with the following modifications:
- include only stations with a data record that starts before 1950 and ends after 2000 (keeping only complete records during that time).
- include only
tmax
ggplot(filter(st_filtered,
first_year<=1950 &
last_year>=2000 &
element=="TMAX"),
aes(y=latitude,x=longitude)) +
geom_sf(data=world, inherit.aes = F, size=.1, fill="grey", colour="black")+
stat_bin2d(bins=75)+
scale_fill_distiller(palette="YlOrRd",trans="log",direction=-1,
breaks = c(1,10,50))+
coord_sf()
Plot temperatures
d_rollmean%>%
ggplot(aes(ymax=tmax,ymin=tmin,x=date))+
geom_ribbon(fill="grey")+
geom_line(aes(y=(tmin+tmax)/2),col=grey(0.4),size=.5)+
geom_line(aes(y=tmax.60),col="red")+
geom_line(aes(y=tmax.b60),col="darkred")
Your Turn
Plot a 30-day rolling “right” aligned sum of precipitation.
tp=d_filtered_recent %>%
arrange(date) %>%
mutate(prcp.30 = rollsum(x = prcp, 30, align = "right", fill = NA))
ggplot(tp,aes(y=prcp,x=date))+
geom_line(aes(y=prcp.30),col="black")+
geom_line(col="red") ## Warning: Removed 40 rows containing missing values (geom_path).## Warning: Removed 11 rows containing missing values (geom_path).
Temporal autocorrelation
Values are highly correlated!
## Warning: Removed 12 rows containing missing values (geom_point).
Autocorrelation
Partial Autocorrelation
But be careful, there were lots of missing data in the beginning of the record
# which years don't have complete data?
d_filtered2%>%
group_by(year)%>%
summarize(n=n())%>%
filter(n<360)## # A tibble: 2 x 2
## year n
## <dbl> <int>
## 1 1938 245
## 2 2017 304Plot 10-year means (excluding years without complete data):
d_filtered2%>%
filter(year>1938, year<2017)%>%
group_by(dec)%>%
summarize(
n=n(),
tmin=mean(tmin,na.rm=T),
tmax=mean(tmax,na.rm=T),
prcp=mean(prcp,na.rm=T)
)%>%
ggplot(aes(x=dec,y=tmax))+
geom_line(col="grey")
Then plot all years (in grey) and add 2017 in red.
ggplot(df,aes(x=doydate,y=tmax,group=year))+
geom_line(col="grey",alpha=.5)+ # plot each year in grey
stat_smooth(aes(group=1),col="black")+ # Add a smooth GAM to estimate the long-term mean
geom_line(data=filter(df,year>2016),col="red")+ # add 2017 in red
scale_x_date(labels = date_format("%b"),date_breaks = "2 months")## `geom_smooth()` using method = 'gam' and formula 'y ~ s(x, bs = "cs")'
Then ‘zoom’ into just the past few months and add 2017 in red.
ggplot(df,aes(x=doydate,y=tmax,group=year))+
geom_line(col="grey",alpha=.5)+
stat_smooth(aes(group=1),col="black")+
geom_line(data=filter(df,year>2016),col="red")+
scale_x_date(labels = date_format("%b"),date_breaks = "2 months",
lim=c(as.Date("2017-08-01"),as.Date("2017-10-31")))
So there was an unusually warm spell in late September.
ggplot(seasonal,aes(y=tmin,x=year))+
facet_wrap(~season,scales = "free_y")+
stat_smooth(method="lm", se=T)+
geom_line()
Linear regression of maximum temperature in fall
s1=seasonal%>%
filter(season=="Summer")
ggplot(s1,aes(y=tmin,x=year))+
stat_smooth(method="lm", se=T)+
geom_line()
Cumulative dry days
Diurnal Temperature Range
Frost Days
