Use Google Earth Engine to Map the Rainfall of Typhoons
Year 2021 has finally passed. It is a year of F****ing Pandemic as the COVID was still lingering, there had been many serious incidents of flooding killing more than 900 people.
Christian Aid’s (2021) annual report, which was published in mid December, ranks the top ten climate disasters of the year, but it was published two weeks before the end of December and there were another two fatal floods in the last two weeks of 2021, one in Malaysia and the other was in Brazil.
Top ten floods in the world in 2021
Here is the list of the top ten floods (and hurricanes and typhoons) in 2021, which cover across five continents: North America, South America, Europe, Asia, and Oceania:
- Floods in Australia on March 10–24–2 people died
- May 14–19 India, Sri Lanka, Maldives hurricane-198 deaths
- May 25–29 India, Bangladesh hurricanes-19 dead
- July 12–18 floods in Germany, France, the Netherlands, Belgium, Luxembourg-240 dead
- July 17–31 floods in Henan, China-302 dead
- July 21–28: Typhoon in China, Philippines, Japan-5 dead
- August 28-September 2 U.S. Hurricane-95 dead
- November 14 Flood in BC, Canada-4 dead (the above 8 items are from Christian Aid, 2021)
- December 19–21 Flood in Malaysia-At least 14 dead (BBC, 2021)
- December 28 Flood in Bahia, Brazil-20 people have died so far (New York Times, 2021)
Using NASA GPM data to draw precipitation map on GEE
To analyze the floods, we must first collect relevant data such as the precipitation data. It is often available from the websites of observatories in various countries, but it can be time-consuming to search one by one.
This article describes how to use NASA’s GPM satellite data to draw precipitation maps on Google Earth Engine (GEE). It is convenient to draw precipitation maps of any places on Earth and on any date without searching for data.
First of all, if you missed the article of how to log in to GEE, you may read it via this link (Yiu, 2021) and log in to GEE first. There are at least three different sets of satellite data on global precipitation. Satellite data of NASA/GPM_L3/IMERG_V06 provides data every half hour and is provided by NASA (Huffman et al., 2019: GPM IMERG Final Precipitation L3 Half Hourly 0.1 degree x 0.1 degree V06, DOI: 10.5067/GPM/IMERG /3B-HH/06).
Precipitation Map of Bahia, Brazil on December 25, 2021
Figure 1 shows the precipitation map of Bahia, Brazil, on Christmas Day on December 25, 2021. The highest rainfall in the red position reached 50 mm per hour, which clearly shows the location and severity of the down pours.
It is difficult to explain step by step in words how to draw the map, a short video is made to demonstrate, which is clearer and easier to understand (Youtube: https://youtu.be/xZefkS3eVJM).
First, the GEE data catalog provides a program that can be run directly on GEE. A brief introduction is as follows (words after // are remarks):
var range = ee.Date(‘2021–12–25’).getRange(‘day’) //Define the time range
var dataset = ee.ImageCollection(‘NASA/GPM_L3/IMERG_V06’) //Define data source
.filter(ee.Filter.date(range)); //Filter the data range
var precipitation = dataset.select(‘precipitationCal’).max(); //Select the highest rainfall data from the specified day
var mask = precipitation.gt(0.5); //Mask data below 0.5 (mm/hr)
var precipitation = precipitation.updateMask(mask); //Define precipitation after removing the masked
var palette = [
‘000096’,’0064ff’, ‘00b4ff’, ‘33db80’, ‘9beb4a’,
‘ffeb00’,’ffb300',’ff6400',’eb1e00',’af0000'
]; //Define the color codes
var precipitationVis = {min: 0.0, max: 30.0, palette: palette}; //Define the minimum and maximum precipitation values
Map.addLayer(precipitation, precipitationVis,’Precipitation’); //Draw the map layer
Map.setCenter(-41, -12, 5); //Positioning map centre at coordinates -41, -12 which is at the northeast of Brazil, and the 5 indicates map size
An illustration — the rainfalls of the Signal 10 Typhoon Mangkhut
The following example shows the impact of Typhoon №10 “Mangkhut” on Hong Kong in 2018. We just need to change the date in the first line of the GEE program to 2018–09–16, and then change the coordinates of the last line to 114, 22 (coordinates around Hong Kong) to get the rainfall map on the day in Hong Kong.
The following Figures 2, 3 and 4 are the precipitation maps of Hong Kong on September 14, 15, and 16, 2018. The red patches of the picture indicate rainfalls as high as 85, 96, and 82 mm/hr. The Hong Kong Observatory also recorded a maximum rainfall of 279 millimeters per hour (Shek Kong Station) on that day.
A Timelapse Video
Put the daily or hourly satellite maps on a Timelapse software can create a time-series dynamic video showing the movement and changes of rainfalls. Readers who are interested in making a timelapse video can read the introduction of Wu (2020). Here I simply use Adobe Video to produce a simpler one with similar effect.
[Use Adobe Video to present the precipitation of Typhoon Mangkhut from 20180914 to 20180916 https://youtu.be/dnNxGQCg2Fo]
References:
Christian Aid (2021) Counting the cost 2021 A year of climate breakdown, December. https://www.christianaid.org.uk/sites/default/files/2021-12/Counting%20the%20cost%202021%20-% 20A%20year%20of%20climate%20breakdown.pdf
Huffman, GJ, EF Stocker, DT Bolvin, EJ Nelkin, Jackson Tan (2019), GPM IMERG Final Precipitation L3 Half Hourly 0.1 degree x 0.1 degree V06, Greenbelt, MD, Goddard Earth Sciences Data and Information Services Center (GES DISC), Accessed: [29 December 2021], 10.5067/GPM/IMERG/3B-HH/06
Wu, Q. (2020) GEE Tutorial #39-How to create Landsat timelapse animations without coding, Geospatial Data Science, Oct 14. https://blog.gishub.org/gee-tutorial-39-how-to-create- landsat-timelapse-animations-without-coding
Yiu, C.Y. (2021) Use Google Earth Engine to Map CH4 concentration, Medium, November 11. https://ecyy.medium.com/use-google-earth-engine-to-map-ch4-concentration-3eaed4241b99
