Galeras, Volcano and Popocatepetl, Volcano SO2 Emission Trends (2008-2013)
Introduction
SO2 gas emission measurements near volcanos can provide many understandings of volcanic eruptions and hazards. When recorded, the trends of these measurements can hold value in predicting future volcanic eruptions. Two volcanos that have many years of SO2 gas measurements is Galeras,Volcano in Colombia and Popocatepetl, Volcano in Mexico. Galeras and Popocatepetl are two extremely active volcanos that have many data to anaylze to provide much insight into SO2 trends of volcanos around the world.
This code does a series of simple data analyses of the SO2 flux data set from NOVAC (Network for Observation of Volcanic and Atmospheric Change) for both Galeras and Popocatepetl. NOVAC takes daily measurements of SO2 using specometers that measure emmited SO2 wavelengths. This code aims to identify trends from Galeras and Popocatepetl from the years 2008-2013. These two volcanos and years were picked because it has the most data and the largest overlap of time.
Outline:
1: Daily Mean Trends of SO2 for Galeras and Popocatepetl
2. Yearly Trends of SO2 for Galeras and Popocatepetl
3. Total Sum of SO2 of Galeras and Popocatepetl
4. Histogram of Yearly Trends
5. Dicussion
Source: Chacon, Z., Arellano, S., Burbano, V., Garzon, G., Laverde, C., Galle, B., SO2 flux of -GALERAS- volcano, from the NOVAC data-base; 2020; [Data set]; v.001; doi:10.17196/novac.galeras.001
Load Packages
library(ncdf4)
library(ggplot2)
library(gridExtra)
library(maps)Read Data: Yearly SO2 Emissions for Galeras and Popocatepetl
Open the Galeras NetCDF file
ncfile <- nc_open("groundbased_novac.scandoas.so2_sgc013_galeras_20071114t000000z_20140329t000000z_001.nc")
variable_data <- ncvar_get(ncfile, "SO2.FLUX_SCANNING.DOAS_DAILY.MEAN")
DATETIME_data <- ncvar_get(ncfile, "DATETIME")Open the Popcatepetl NetCDF file
ncfile <- nc_open("groundbased_novac.scandoas.so2_unam032_popocatepetl_20070412t000000z_20161228t000000z_001.nc")
variable_data2 <- ncvar_get(ncfile, "SO2.FLUX_SCANNING.DOAS_DAILY.MEAN")
DATETIME_data2 <- ncvar_get(ncfile, "DATETIME")Trim Data: 2008-2013 Yearly SO2 Emissions for Galeras and Popocatepetl
Get Galeras DATETIME variable into years.
divide_array <- DATETIME_data / 365
finaltime_array <- divide_array + 2000Trim Galeras new variable years down to 2008-2013
rows_to_remove3 <- c(698:704)
x3array <- finaltime_array[-rows_to_remove3]
array3 <- variable_data[-rows_to_remove3]
rows_to_remove5 <- c(1:9)
x4array <- x3array[-rows_to_remove5]
array4 <- array3[-rows_to_remove5]Get DATETIME Popcatepetl into Years
divide_array2 <- DATETIME_data2 / 365
finaltime_array2 <- divide_array2 + 2000Trim Popcatepetl new variable years down to 2008-2013
rows_to_remove <- c(1:151)
array <- finaltime_array2[-rows_to_remove]
xarray <- variable_data2[-rows_to_remove]
rows_to_remove2 <- c(882:1057)
array2 <- array[-rows_to_remove2]
x2array <- xarray[-rows_to_remove2]Plot Mean Daily SO2 Emissions Galeras and Popcatepetl
ggplot() +
geom_line(aes(x = (x4array), y = array4)) + labs(title="Galeras Daily SO2 Measurements vs Date(Year)", x = "Date (Year)", y = "SO2 Flux (kg/s)") + scale_x_continuous(breaks= round(seq(min(2007), max(2015),by = 1)))
ggplot() +
geom_line(aes(x = (array2), y = x2array)) + labs(title="Popocatepetl Daily SO2 Measurements vs Date(Year)", x = "Date (Year)", y = "SO2 Flux (kg/s)") + scale_x_continuous(breaks= round(seq(min(2007), max(2014), by = 1)))
Plot Daily SO2 Emmisions Galeras vs Popcatepetl Side-by-side plots
plot1 <- ggplot() +
geom_line(aes(x = (x4array), y = array4)) + labs(title="Galeras Daily SO2 Measurements vs Date(Year)", x = "Date (Year)", y = "SO2 Flux (kg/s)") + scale_x_continuous(breaks= round(seq(min(2007), max(2015),by = 1)))
plot2 <- ggplot() +
geom_line(aes(x = (array2), y = x2array)) + labs(title="Popocatepetl Daily SO2 Measurements vs Date(Year)", x = "Date (Year)", y = "SO2 Flux (kg/s)") + scale_x_continuous(breaks= round(seq(min(2007), max(2014), by = 1)))
grid.arrange(plot1, plot2, ncol=2)
Total Popcatepetl SO2 Emmisions by Year
Categorize Popocatepetl data into year and sum data for each year.
2008
rows_to_removepop2008 <- c(201:881)
arraypop2008 <- x2array[-rows_to_removepop2008]
yarraypop2008 <- array2[-rows_to_removepop2008]
arraysumpop2008 <- sum(yarraypop2008)2009
rows_to_removepop2009 <- c(371:881)
arraypop2009 <- x2array[-rows_to_removepop2009]
yarraypop2009 <- array2[-rows_to_removepop2009]
rows_to_removepop2009p2 <- c(1:211)
arraypop2009pt2 <- arraypop2009[-rows_to_removepop2009p2]
yarraypop2009pt2 <- yarraypop2009[-rows_to_removepop2009p2]
arraysumpop2009 <- sum(yarraypop2009pt2)2010
rows_to_removepop2010 <- c(531:881)
arraypop2010 <- x2array[-rows_to_removepop2010]
yarraypop2010 <- array2[-rows_to_removepop2010]
rows_to_removepop2010p2 <- c(1:381)
arraypop2010pt2 <- arraypop2010[-rows_to_removepop2010p2]
yarraypop2010pt2 <- yarraypop2010[-rows_to_removepop2010p2]
arraysumpop2010 <- sum(yarraypop2010pt2)2011
rows_to_removepop2011 <- c(711:881)
arraypop2011 <- x2array[-rows_to_removepop2011]
yarraypop2011 <- array2[-rows_to_removepop2011]
rows_to_removepop2011p2 <- c(1:531)
arraypop2011pt2 <- arraypop2011[-rows_to_removepop2011p2]
yarraypop2011pt2 <- yarraypop2011[-rows_to_removepop2011p2]
arraysumpop2011 <- sum(yarraypop2011pt2)2012
rows_to_removepop2012 <- c(811:881)
arraypop2012 <- x2array[-rows_to_removepop2012]
yarraypop2012 <- array2[-rows_to_removepop2012]
rows_to_removepop2012p2 <- c(1:711)
arraypop2012pt2 <- arraypop2012[-rows_to_removepop2012p2]
yarraypop2012pt2 <- yarraypop2012[-rows_to_removepop2012p2]
arraysumpop2012 <- sum(yarraypop2012pt2)2013
rows_to_removepop2013 <- c(1:821)
arraypop2013 <- x2array[-rows_to_removepop2013]
yarraypop2013 <- array2[-rows_to_removepop2013]
arraysumpop2013 <- sum(yarraypop2013)Total Galeras SO2 Emmisions by Year
Categorize Galeras data into year and sum data for each year.
2008
rows_to_removeg2008 <- c(191:651)
arrayg2008 <- x2array[-rows_to_removeg2008]
yarrayg2008 <- array2[-rows_to_removeg2008]
arraysumg2008 <- sum(yarrayg2008)2009
rows_to_removeg2009 <- c(361:651)
arrayg2009 <- x2array[-rows_to_removeg2009]
yarrayg2009 <- array2[-rows_to_removeg2009]
rows_to_removeg2009p2 <- c(1:181)
arrayg2009pt2 <- arrayg2009[-rows_to_removeg2009p2]
yarrayg2009pt2 <- yarrayg2009[-rows_to_removeg2009p2]
arraysumg2009 <- sum(yarrayg2009pt2)2010
rows_to_removeg2010 <- c(401:651)
arrayg2010 <- x2array[-rows_to_removeg2010]
yarrayg2010 <- array2[-rows_to_removeg2010]
rows_to_removeg2010p2 <- c(1:351)
arrayg2010pt2 <- arrayg2010[-rows_to_removeg2010p2]
yarrayg2010pt2 <- yarrayg2010[-rows_to_removeg2010p2]
arraysumg2010 <- sum(yarrayg2010pt2)2011
rows_to_removeg2011 <- c(491:881)
arrayg2011 <- x2array[-rows_to_removeg2011]
yarrayg2011 <- array2[-rows_to_removeg2011]
rows_to_removeg2011p2 <- c(1:391)
arrayg2011pt2 <- arrayg2011[-rows_to_removeg2011p2]
yarrayg2011pt2 <- yarrayg2011[-rows_to_removeg2011p2]
arraysumg2011 <- sum(yarrayg2011pt2)2012
rows_to_removeg2012 <- c(631:881)
arrayg2012 <- x2array[-rows_to_removeg2012]
yarrayg2012 <- array2[-rows_to_removeg2012]
rows_to_removeg2012p2 <- c(1:481)
arrayg2012pt2 <- arrayg2012[-rows_to_removeg2012p2]
yarrayg2012pt2 <- yarrayg2012[-rows_to_removeg2012p2]
arraysumg2012 <- sum(yarrayg2012pt2)2013
rows_to_removeg2013 <- c(1:621)
arrayg2013 <- x2array[-rows_to_removeg2013]
yarrayg2013 <- array2[-rows_to_removeg2013]
arraysumg2013 <- sum(yarrayg2013)Sum of Total SO2 Emmisions by Year for Galeras and Popocatepetl
galerastotalemissions <- array(data = c(arraysumg2008,arraysumg2009,arraysumg2010,arraysumg2011,arraysumg2012,arraysumg2013), dim= c(6,1))
popocatepetltotalemissions <- array(data = c(arraysumpop2008,arraysumpop2009,arraysumpop2010,arraysumpop2011,arraysumpop2012,arraysumpop2013), dim= c(6,1))Plot Sum of Total SO2 Emissions by Year for Galeras and Popocatepetl
ggplot() +
geom_point(aes(x = (2008:2013), y = popocatepetltotalemissions)) + labs(title=" Popocatepetl Total SO2 Emissions Per Year", x = "Year", y = "SO2 Flux (kg/s)") + scale_x_continuous(breaks= round(seq(min(2008), max(2013), by = 1))) + theme_bw() + geom_point(size = 7)
ggplot() +
geom_point(aes(x = (2008:2013), y = galerastotalemissions)) + labs(title=" Galeras Total SO2 Emissions Per Year", x = "Year", y = "SO2 Flux (kg/s)") + scale_x_continuous(breaks= round(seq(min(2008), max(2013), by = 1))) + theme_bw() + geom_point(size = 7)
Total 6 Year SO2 Emissions Galeras vs Popcatepetl
arraysumgaleras <- sum(array4)
arraysumpopocatepetl <- sum(x2array)
print(arraysumgaleras)## [1] 6260.133print(arraysumpopocatepetl)## [1] 20072.05totalemissions <- array(data = c(arraysumgaleras, arraysumpopocatepetl), dim = c(2,2))
ggplot() +
geom_point(aes(x = (2:2), y = totalemissions)) + labs(title="Galeras vs Popocatepetl Total SO2 6 Year Emissions", x = "Galeras (Below) vs Popocatepetl (Above)", y = "SO2 Flux (kg/s)") + scale_x_continuous(breaks= round(seq(min(0), max(2), by = 1))) + theme_bw() + geom_point(size = 7)
Histogram of Galeras and Popocatepetl Data
Popocatepetl <- array2
Galeras <- x4array
hist(Popocatepetl, breaks=20)
hist(Galeras, breaks=20)
Discussion
Galeras and Popocatepetl are two extremely active volcanos. SO2 gas emmisions can hold value in predicting volcanic eruptions and health risk of SO2 in the future. Map of Galeras,Volcano in Colombia and Popocatepetl, Volcano in Mexico to show the location where both SO2 trends and volcanic eruptions may be effecting.
world_map <- map_data("world")
point_data <- data.frame(
lon = c(-77.36, -98.62),
lat = c(1.22, 19.02)
)
ggplot() +
geom_polygon(data = world_map, aes(x = long, y = lat, group = group), fill = "lightblue") +
geom_point(data = point_data, aes(x = lon, y = lat), color = "red", size = 3)
Galeras Trends
Galeras’ Daily Mean of SO2 trend is remarkable throughout the six years because it shows a huge peak in SO2 gas emissions around 2008 and subsiding to lower SO2 after 2009. During 2008 there was a volcanic eruption and this trend in the graph shows that clearly. This trend can be useful in predicting volcanic eruptions for Galeras in the future. We see in the graph that the normal SO2 emission trend is low for Galeras, therefore if we start to see a increase in SO2 all of a sudden in the future, this could be a sign of a volcanic eruption. This also could be a good proxy for other volcanos that exibit behavior like Galeras.
Galeras’ Yearly trend of SO2 is was expected, 2008 and 2009 had the highest amount of SO2 emissions due to the eruption at that point.The total amount of SO2 for the full six years was lower than Popocatepetl which could be explained by the idea that Popocatepetl is more active or due to the histogram showing Galeras had less amount of data recorded. Either way, it is good that there is multiple ways to predict eruptions from Galeras including seismic data.
Popocatepetl Trends
Popocatepetl’s Daily Mean of SO2 emissions differs from Galeras with its trend being consistently high. This is due to Popocatepetl being a lot more active than Galeras. Popocatepetl did have a bigger eruption specifically in 2013 which can explain the biggest spike in the data during that year. A volcano like Popocatepetl might be harder to predict bigger eruptions without outside data due to the constant activity.
Popocatepetl’s Yearly trend of SO2 is as expected, just a lot of data around the same amount. Galera’s and Popocatepetl total of SO2 throughout the six years differed by an order of magnitude with Popocatepetl being higher. Like previously mentioned this could be due to differences in amount of data or just due to Popocatepetl have more emissions. By just looking at all the graphs combined for both volcanos, Popocatepetl seems to have both more data and more total emissions compared to Galeras.
It’s also interesting to know the total amount of SO2 that is being put out by Popocatepetl, this could be important in understand health risk in the surrounding cities in Mexico. SO2 is harmful to respiratory and lung functions therefore this secondary hazard of volcanic eruption is important to record and analyze.