This is a post in reply to Nick Stokes criticism of using the entire Amazon for fire statistics, rather than just using the state of Amazonas.
These are GFED (Global Fire Emissions Database) regions in the Amazon Region (click on region to go to that data). I have added the country of each region, and the numbers of fires, as of Aug 23.
- Acre Brazil, 2427
- Amazonas Brazil, 10494
- El Beni Bolivia, 7410
- Maranhão Brazil, 7301
- Mato Grosso Brazil, 27016
- Pando Bolivia, 2019
- Pará Brazil, 13895
- Peru Peru (duh), 6044
- Rondônia Brazil, 10425
- Santa Cruz Bolivia, 20135
Now, of the 10 regions (don’t’ count “Totals”), 4 are outside Brazil; 3 in Bolivia, and 1 in Peru. I have further broken down each sub region, in this spreadsheet.
Care should be used when quoting Emissions, as per the warnings on the GFED Website, and due to the fact that the totals above do not match the Totals bar chart on the GFED web page.
The state of Amazonas has only 9.8% of total fires in the entire region. The fire per 1000 km2 is second lowest. This ratio of area to fires is about 40% of the ratio for the entire Amazon region. The number of fires is only 4th highest.
Mato Grosso state has the highest number of fires, both as % and as an actual number. It has the 5th highest fires/1000 km2. Yet Nick specifically criticized using this state. Even though it borders on Amazonas State.
If there is a fire problem in the Amazon, it’s actually in Bolivia, specifically Santa Cruz state. Why don’t we hear the condemnation of Bolivia?
But, lets take out Peru and Bolivia and Mato Grosso. Amazonas still has only about 23% of the remaining fires, nearly tied with Rondonia.
Here is a visual of the geographical spread of the fires, from CBC News.
That big area in the Northwest, with relatively few fires? That’s Amazonas. See also that most of the fires in Mato Grosso, are very close to the Amazonas border. (see map below for location of states)
A Google Search of the terms ‘Amazon Fires at Record Number’ returns nearly 300,000,000 hits. ‘Amazonas Fires at Record Number” returns 1/200th that many hits. Even the news sources that use the term Amazonas, do so only in passing, or misleading it as the major cause. This is an example:
See how the news article blames Amazonas for the smoke in Sao Paulo, and ignores the much larger number of fires in the states it mentions; Para and Mato Grosso. The Tweet does as well, though it attributes to another source for the smoke.
My point here, is most of the news mentions “Amazon”, not “Amazonas”. Searching for Rondonia gives 190,000 hits. Searching for Mato Grosso gives 2.48 million. Amazonas gives 1.48 million.
While Amazonas is often blamed for the problem, as in the news clip above, it’s obviously not the only source, and actually has a relatively small contribution of about 10%. Yes, it is near a record number of fires, but so is Santa Cruz, Bolivia. It would be interesting to see what drives this.
It has been suggested to me to only use Amazonas as data for the fires in the Amazon. One Twitter suggested I also only use the years after 2011, because there was drought before that.
If I was so selective in choosing cherries, I am sure I could make a sweet, sweet, cherry pie.
Conclusion 1: It’s a big area, with a large demographic spread. Limiting statistics to one single sub-region may help understand that area, but not the larger whole region. Using only 1 sub region to comment on the state of the larger region, is just wrong. It would be just as wrong, to quote Maranhao 2019 numbers as showing that Amazon fires are decreasing. (Even though that is exactly what GFED emissions numbers show.)
Conclusion 2: Two sub regions are near record numbers, but the entire region is very close to average.
Conclusion 3: OK, not a conclusion, a question. How the heck does Bolivia avoid condemnation?
Note: Again, please be specific with criticism, quoting what you disagree with, and why.
via Watts Up With That?
August 24, 2019 at 08:58PM