PMDI shows an excess of moisture in Georgia. and almost no drought in Arizona, but the Washington DC based US Drought Monitor artist forgot to check the actual data before getting out his airbrush.
Disrupting the Borg is expensive and time consuming!
Google Search
-
Recent Posts
- Global Warming Emergency In The UK
- Mainstream Media Analysis Of DOGE
- Angry And Protesting
- Bad Weather Caused By Racism
- “what the science shows”
- Causes Of Earthquakes
- Precision Taxation
- On the Cover Of The Rolling Stone
- Demise Of The Great Barrier Reef
- Net Zero In China
- Make America Healthy Again
- Nobel Prophecy Update
- Grok Defending Climategate
- It Is Big Oil’s Fault
- Creative Marketing
- No Emergency Or Injunction
- The Perfect Car
- “usually the case”
- Same Old Democrats
- Record Arctic Ice Growth
- Climate Change, Income Inequality And Racism
- The New Kind Of Green
- The Origins Of Modern Climate Science
- If An Academic Said It, It Must Be True
- Record Snow Cover
Recent Comments
- spren on Mainstream Media Analysis Of DOGE
- spren on Global Warming Emergency In The UK
- conrad ziefle on Global Warming Emergency In The UK
- conrad ziefle on Mainstream Media Analysis Of DOGE
- Bob G on Global Warming Emergency In The UK
- Ivan G Wainwright on Global Warming Emergency In The UK
- Robertvd on Global Warming Emergency In The UK
- Independent on Global Warming Emergency In The UK
- Bob G on Global Warming Emergency In The UK
- william on Angry And Protesting
When was the first image published, and do you have a link to the source?
When would you guess that a February, 2013 drought map might have been published? In 1985 perhaps?
http://www1.ncdc.noaa.gov/pub/data/cmb/drought/historical-palmers/201302-pmdi.gif
A better link for Gina: http://www.ncdc.noaa.gov/temp-and-precip/drought/historical-palmers.php?index=pdsi&month%5B%5D=2&beg_year=2013&end_year=2013&submitted=Submit
Plug in your choice of months and years, suggest compare 2012 to 1988, the 1950s and 1930s:
1988 – http://www.ncdc.noaa.gov/temp-and-precip/drought/historical-palmers.php?index=pmdi&month%5B%5D=6&month%5B%5D=7&month%5B%5D=8&month%5B%5D=9&beg_year=1988&end_year=1988&submitted=Submit
1950s – http://www.ncdc.noaa.gov/temp-and-precip/drought/historical-palmers.php?index=pmdi&month%5B%5D=6&month%5B%5D=7&month%5B%5D=8&month%5B%5D=9&beg_year=1953&end_year=1956&submitted=Submit
1930s – http://www.ncdc.noaa.gov/temp-and-precip/drought/historical-palmers.php?index=pmdi&month%5B%5D=6&month%5B%5D=7&month%5B%5D=8&month%5B%5D=9&beg_year=1931&end_year=1937&submitted=Submit
Maybe one of them is short term drought, the oither long term drought? Whatever!
They use two different methods to determine “Drought” conditions! USDA is in the Chicken Little business even more than NOAA, and NOAA is into it over their heads.
Keep posting this stuff and the droughts in the past will start to dissapear, along with strem flow data.