data(venice)
data(venice90)
Thanks to Paolo Pirazzoli and Alberto Tomasin for the venice90
data.
venice90
, the value 0 corresponds to a fixed
reference point (e.g., the mean sea level in 1897 at an old
palace of Venice). Clearly since the relative (perceived)
mean sea level has been increasing in trend over time (more
than an overall 0.4 m increase by 2010), therefore the value 0 is
(now) a very low and unusual measurement.
For venice
, in 1935 only the top six values were recorded.
For venice90
, this is a subset of a data set provided by
Paolo Pirazzoli consisting of hourly sea levels from 1940 to 2009.
Values greater than 90 cm were extracted, and then declustered
(each cluster provides no more than one value, and
each value is at least 24 hours apart).
Thus the values are more likely to be independent.
Of the original (2009-1940+1)*365.26*24
values
about 7 percent of these comprise venice90
.
Yet to do: check for consistency between the data sets. Some external data sets elsewhere have some extremes recorded at times not exactly on the hour.
guplot
,
gev
,
gpd
.matplot(venice[["year"]], venice[, -1], xlab = "Year",
ylab = "Sea level (cm)", type = "l")
ymat <- as.matrix(venice[, paste("r", 1:10, sep = "")])
fit1 <- vgam(ymat ~ s(year, df = 3), gumbel(R = 365, mpv = TRUE),
data = venice, trace = TRUE, na.action = na.pass)
head(fitted(fit1))
par(mfrow = c(2, 1), xpd = TRUE)
plot(fit1, se = TRUE, lcol = "blue", llwd = 2, slty = "dashed")
par(mfrow = c(1,1), bty = "l", xpd = TRUE, las = 1)
qtplot(fit1, mpv = TRUE, lcol = c(1, 2, 5), tcol = c(1, 2, 5),
llwd = 2, pcol = "blue", tadj = 0.1)
plot(sealevel ~ Year, data = venice90, type = "h", col = "blue")
summary(venice90)
dim(venice90)
round(100 * nrow(venice90) / ((2009 - 1940 + 1) * 365.26 * 24), digits = 3)
Run the code above in your browser using DataLab