As scientists, we tend to get carried away with our acronyms. Sometimes they can be clever (such as Verification of the Origins of Rotation in Tornadoes Experiment, or VORTEX), and others there’s an acronym within an acronym (TOVS is an optical vertical sounder, but the ‘T’ stands for TIROS, which stood for Television Infrared Observation Satellite). In a world of scientific jargon, how are users going to figure out what each acronym means and what function it poses? Well have no fear, I’m here to break down some of these acronyms used to describe weather and climate data sets in which I’m calling: Alphabet Soup.
A good number of acronyms in this series are based upon weather and climate datasets, but what about data…about the data? This is also known as metadata. Depending on the dataset, metadata can mean many different things. First it could describe location information, such as latitude, longitude, and elevation. It could provide more information about the station site, such as vegetation, population density, or simply if it exists on land (weather station) or water (buoy). Some datasets also go into great detail about the instrument recording the information. Items can include the type of sensor, how it was calibrated, and when maintenance is applied to the instrument.
This information can sometimes be overwhelming, and worse, overlooked. Most people simply want the data and analyze it, without all the nitty gritty information. While this can be okay most of the time, there are some exceptions to the rule. NOAAs National Centers for Environmental Information understands this, which is why they have built a database to try and track all of this information. It is called the Historical Observing Metadata Repository… or HOMR.
Yes, it is pronounced “Homer” and the first thing that pops into my mind is an early episode of the Simpsons called “Marge on the Lam” when a young teenage Homer Simpson is caught smashing a weather station. While provided for comedic laughter, this actually is a good metaphor for describing the HOMR dataset. There are thousands of weather stations across the country, all recording relevant information. Like the underlying data, metadata can act as a function of time. Weather stations change over time for many reasons, including:
The station moves. Maybe it got moved to a different part of the airport, or the observer taking the measurements moved across town.
The observer changes. Sometimes weather stations are monitored by members of the government (such as the Air Force), and the person taking the measurements can change because of a move, or even retirement.
The instruments need to be replaced. Nothing lasts forever, and the instrument needs to be upgraded, calibrated, or simply replaced with newer technologies. There aren’t reports of people hitting weather stations with baseball bats, but they can be broken by localized high winds or vandalized by animals in the area.
HOMR tries its best to capture this information. Once you enter the webpage, it provides a search function to help users find the station they are looking for. Once a station is found, the database brings up all the information it could find. It can be very overwhelming at first, but believe me when I say it’s important information. Let’s look at an example weather station at the Reno, Nevada airport. First, here is some information it found in regards to the instrumentation used to record the weather elements:
Highlighting over the word provides more information about it (for example, TB means tipping bucket rain gauge). First thing to notice is there are a LOT of instrument changes over time. Nationally, there were a couple of big changes to weather instruments. In the 1980’s, many thermometers were upgraded from a liquid in glass thermometers to a digital thermistor. Then in the 1990's, airport stations got upgraded to the Automated Surface Observing System (ASOS…stay tuned for that Alphabet Soup), which included more upgrades. Not only did this Reno station get an ASOS upgrade in 1995, it has since been upgraded with even newer instruments. For example, the anemometer measuring wind speed was upgraded from a 5-second cup anemometer to a 3-second sonic anemometer.
The most interesting item about this station is that it has moved so many times. According to the database, it has moved at least 7 times during its existence. If one was just grabbing the data for its entire period of record, the user might not know this. HOMR provides location information about each move, and also maps it to provide spatial context.
Now for the most interesting question. WHY do this?. Why should people care about all the station moves or instrument changes? People say “just get me the data!” Well during two station moves (one in the 1930s and the other in the 1990s), there was a shift in the temperature record. The differences are usually small (about a degree or 2), but can have drastic effects when looking at long term climate analysis. A paper by Menne et. al 2009 showed this shift in the temperature time series (or breakpoint) and was able to readjust the data so it compares well with not only its entire period of record, but also neighboring stations. The algorithm is automated, but is assisted by the metadata that is read into the system. Had HOMR not had this information, these breakpoints might not have been resolved.
While it can be a time consuming matter to check on the metadata, it’s important to know information about any data station before an analysis is run. You never know if someone took a baseball bat to a weather station in the past. In fact, in that same episode of The Simpsons an older Homer contemplated smashing the same weather station again (thankfully he did not).