This week, we are experimenting with a dataset that illustrates the global distribution of surface water. The data set provides a three-way pixel classification with water detected, no water detected, and no data at 30 meters, with records ranging from 1984 through 2020.
The collection is comprised of 4,453,989 Landsat 5, 7, and 8 scenes captured between 16 March 1984 and 31 December 2020 (Google Data Catalog description), covering 442 months (one image per month).
To learn more about this dataset, please see https://www.nature.com/articles/nature20584.
The animation presents an event that is quite meaningful to me, both geographically and historically speaking. I grew up hearing anecdotes from my parents about the 1980s flood in the province of Buenos Aires, specifically in the Flooding Pampas region where I am from. In 1980, 37 Buenos Aires districts (4 million hectares) were flooded by 30 billion liters of water, according to the article from Scarpati and Capriolo. During the same decade, 1984, 1986, and 1987 also saw much above-average precipitation (for a comprehensive listing of years above and below the mean, please see the article), adding to the problem. The authors made an outstanding job of collecting material since 1576, yes 1576, in Argentina (even before Argentina became a country), and although, we do not have satellites that can go so far back in time, thanks to the Landsat mission and the dataset shown before, I was able to recreate the Epecuen Lagoon area in 1986 and 2019.
Villa Epecuen was flooded almost in the blink of an eye, and many stories are tied to this region. Only in recent years did the lagoon's extent diminish, exposing the ruins and showing to the younger generations some of the devastations of that period. The comparison given above illustrates the contrast between the time when the town was already submerged (the major flood happened in 1985, but the dataset does not have data for that region) and the present condition of the area.
This video is helpful if you want to understand more about the causes apart from the weather: https://www.youtube.com/watch?v=iHNS-s35Y3E
References and bibliography:
Pekel, J. F., Cottam, A., Gorelick, N., & Belward, A. S. (2016). High-resolution mapping of global surface water and its long-term changes. Nature, 540(7633), 418-422.
Scarpati, O. E., & Capriolo, A. D. (2013). Sequías e inundaciones en la provincia de Buenos Aires (Argentina) y su distribución espacio-temporal. Investigaciones geográficas, (82), 38-51.
NOTES: the codes to recreate this image, past and future ones will be available on the last post of each month.
コメント