Meteorological reanalysis for the masses!

What was the weather like on the day when you were born?

Have you ever been asking yourself what was the weather like at the day when you were born? And what consequences, if any, could it have for your life? Oddly strange, people tend to develop their own beliefs, including but not limited to that, that movements and positions of celestial bodies correspond somehow to events experienced on a human life time scale. Even more, there are people, who claim that they can somehow understand and interpret correlation between celestial bodies and each individual’s life, especially if that individual is willing to pay for that enlightenment. Funny enough, how people tend to believe that such distant objects could have any influence on their life but they do not take into account mundane phenomena like everyday weather or even political and sociological or economical circumstances at the time they were born. However, I cannot answer these questions and it is not my job to do that.

Being a climate scientist my job is to investigate climate of the past in order to make educated guess about possible climate regimes in the future. Just for fun, I would like to investigate if there is any correlation between weather at the day someone is born and his life afterwards. I doubt that I will have much time to do that seriously, but if enough people get interested maybe something turns out of it. Therefore, I am going to provide ingredients for DIY geeks, willing to try some tweaking on their own. Luckily enough, most of the ingredients are available online and most of them are available in public domain or with some restrictions for commercial use. So, here we go!

Data and Visualisation

The most important component is of course data, that is information about the atmospheric conditions at the day of your interest. There are several sources available but the best applicable should be meteorological reanalysis, these are European Centre for Medium-Range Weather Forecasts Reanalysis (ERA-40)
http://data-portal.ecmwf.int/data/d/era40_daily spanning from September 1957 to August 2002 and NCEP/NCAR reanalysis:
http://www.cdc.noaa.gov/data/gridded/data.ncep.reanalysis.html spanning from January 1948 to present.

So if your birthday happened to be after 1948 you can make plots of the weather conditions for that special day of yours. Temporal coverage of the data are 4-times daily. Spatial coverage is rather coarse, 2.5 degree x 2.5 degree, but it is for the whole globe 0.0E to 357.5E, 90.0N to 90.0S (144x73 global grids). With NCEP/NCAR reanalysis is even possible to produce online plots and print them in postscript, png or gif format. Here, I will demonstrate how to do that for the temperature at the 2 m (T2M) height above the surface, for the Adriatic Sea region, at the 9th of February 1971.

At the web page:
http://www.cdc.noaa.gov/data/gridded/data.ncep.reanalysis.html choose surface and on the next page click on the plot below Create plot/subset in the Air Temperature 4-times Daily row. On the next page click on Make plot or subset and then you have possibility to select the region and time of your interest. There are several plot output options you can play with. I suggest to make plot on white background with fill/shade option. If your region is small it is advisable to scale plot to 150% or even 175% to increase the image. Contour range should also be adjusted to fit the range of actual data. You might need at least twice to plot your data, first time to see what is your actual range of data (minimum and maximum value are printed on the plot, so make sure that your range is between these two value), and next time with adjusted range. After you click on Create plot or subset of data you will get the picture which you can save with the right click of your mouse, but you can also download the data in netcdf format in order to visualise your data offline with visualisation package of your choice.

I will write something about that later, but now lets just plot some other parameters. Same as T2M we can repeat the procedure for other surface parameters, such as precipitable water content and mean sea level pressure. For that purpose click on Specify a new image from the NCEP Reanalysis Surface Level data set and click on the variable that interest you the most.

So from all this pictures you can see that 9th of February 1971, about midday was sunny day with temperature about 5-6 deg Celsius in Zagreb (Croatia). High pressure field dominated above the Adriatic Sea. My mother agree with that!

Now, if you want to make your pictures reflecting your individual preferences you can manipulate them with some of the popular graphical package (GIMP, Photoshop, or similar). Even more, customisations are possible if you download the data and use one of the freely available scientific data visualisation and analysis packages, such as ferret, GrADS, GMT or similar, but how to do that is out of scope of this post for the time being. In some of the next posts, I might write about downscaling and increasing spatial and temporal resolution of your data in order to produce high-resolution images or even animations with decreased time step, perhaps rendered in 3D. Depending on the feedback I might consider to build database based on CMS (such as Joomla!), containg information about weather (from reanalysis and from witness description), socio-political and economic situations and some selected informations from respective person's life.


I hope you enjoyed this trip, have fun!