The underlying mission of my job is to safeguard lives and property through climate change adaptation based on science. In other words, to help society to prepare itself for risks connected with more extreme rainfall and temperatures.
For many people, “climate” may seem to be an abstract concept. I have had many conversations about climate, and then realised that people often have different interpretations. In my mind, climate is the same as weather statistics (which I realise can be quite abstract to many).
To avoid miscommunication, I want to make sure that we are on the same page when I discuss climate. Maybe it helps if I talk about more familiar and specific aspects, such as the temperature, rainfall, snow, or wind?
Data, facts, and climate
But I have a challenge because data and facts are often not valued and engaging. This is a general problem when it comes to climate change, as there are probably few other scientific disciplines that have shared more data then the climate science community.
There is a bounty of sites that will give you access to free and open data, however, the access does not necessarily mean that it’s easy make use of the information embedded in the data. Often it requires a bit of work and skills in order to download and visualize it.
For instance, there are some great web-portals, such as NASA/GISS, Global Historical Climate Network (GHCN), and the KNMI ClimateExporer. These portals are extremely useful for scientists and experts (they also give the contrarians numbers on which to build their misconceived ideas), but they may be too complicated for a lot of people.
There are also many stories about climate in the mass media these days, and I have started to ignore many of these reports because they are not all relevant. So, if climate is perceived as both an abstract concept and not always relevant, then it’s a real challenge to engage people in the question about climate change.
See how rainfall and temperature have affected you
On the other hand, people care about local issues where they live and have a direct connection to their lives. So perhaps the message about climate change is perceived as more relevant when people can see the historical temperatures and rainfall near where they live?
Based on these thoughts, we wanted to try to make a simple app without jargon, acronyms and technical terms that enables the (wo)man in the street to explore the precipitation and temperature measured in her/his vicinity.
One intention with this app was to start with a simple overview picture of the measured climate data. It is important not to overwhelm at first sight, but let people understand the depth of the data once they start to explore it.
The data is not perfect
When viewing these data, it is important to be aware that there may be an occasional error in the measurements, but showing the data and letting people explore it may bring such errors to our attention.
You are also likely to come across some records with a misleading trend estimate if you study the data, because there are some data records for stations that have been relocated during the period of measurement, the instrument has been replaced, or the observational practices has been changed.
The Open Climate Data Prototype (OCDP), shown in the iFrame below, has been designed for a project in Mozambique, but is being tested for Norway. We wanted to experiment with ways to make the climate information more easily available for people.
An open and interactive app
You can change the main settings of this app by clicking on the icon with three horizontal lines in the top left corner, for instance to change the region/country or the element. It also lets you study the metadata as well as summary statistics.
The graphics is interactive, and there are three tabs showing different aspects of the data. You can also explore daily, monthly, seasonal or annual temperature or precipitation for a selection of locations in different parts of the world (e.g. a selection from North America, Australia, Eurasia, Africa, Asia).
After we developed this prototype, we realised that there was a similar app called ACD-App (GitHub) of the Southern African Science Service Centre for Climate Change and Adaptive Land Management (SASSCAL). However, I don’t know if it is up and running as a web-based as well.
Both apps are developed in R and R-shiny, and can in principle run on a stand-alone desktop/laptop as well as on a server. SASSCAL is a joint initiative of Angola, Botswana, Namibia, South Africa, Zambia, and Germany in response to the challenges of global change, and is highly relevant to our capacity-building project in Mozambique.
Know the past before you can know the future
The study of past climatic variations and trends is necessary before we can make projections for the future. The historical data provide us with important clues about how different conditions interact, as well as being the basis for model evaluation. They are also important for studying the impact of local climate fluctuations on society, such as crop yields.
23 Responses to "Let’s check your temperature"
C. W. Dingman says
Very Interesting. However, I was able to change the location on the “globe” but the stats I got still pertained to Norway. Probably I didn’t do something I should have done,
In any event, when I learn how to use it, I think some of my “weather conscious”, but skeptical neighbors up here will find it informative.
Hi, try to view this screencast to see if it shows you how to change the region. -rasmus
I applaud the intent; I have long thought the “local” aspect is important. Some example modifications that might be useful:
1. You say
But your app says:
Clearly, we live on different streets. I’ve said in the past that scientists at your level have little experience with, or have forgotten, if they did, teaching “the public”, meaning people who are not science majors at a good university.
Have you done anything like what is called a “Focus Group” with people not of your own educational and socioeconomic background, to constrain your language choices?
2. On the visuals:
-When you connect data points with lines that create sharp peaks and valleys, it distorts the significance of the trend lines. I don’t know if this is a problem with the software– I see this all the time, and I can’t understand why people use that plotting mode.
-I’m not a man in the street, and I’ve followed this topic for a long time, but I really don’t know what the map with a lot of colored dots on it is supposed to be telling me. If you can articulate what you want to convey in words, perhaps you can more readily find a good visual way to communicate it.
Anyway, I hope you take the critique in the positive sense that it is intended– overall, your approach seems overly complicated for the purpose, and off-target for public communication.
[Response: Thanks! The critique is definitively welcome. You are the “focus group” – the app will evolve as I get feedback. The challenge is to make something that suits a wide range of people. -rasmus]
Tom Sager says
That’s because so many Norwegians have come to the United States since Donald Trump invited them all to immigrate last January.
(Let me know if frivolous comments like this are considered unexceptable on this list.) -Yusha
[Response: It’s OK, and I think many Norwegians find the guy really scary. -rasmus]
Mr. Know It All says
All 13 of the folks in Africa who have laptops will be very happy to learn of this new technology.
[Response: It will mainly be for people working in the meteorological and hydrological services, but also universities. And it works on smart phones and tablets too. -rasmus]
Alastair B. McDonald says
To my mind you fell at the first fence. You wrote:
“In my mind, climate is the same as weather statistics (which I realise can be quite abstract to many).”
to me that implies average weather, which is what most meteorologist believe. But the dangers of climate change are not that the temperature will rise by 1.5C on average. The danger is from the more extreme events caused by the temperature rise, which will lead to wild fires and heat waves. The average rise in sea level of 10cm will not flood many places. It is the the storm surges 10cm higher which will cause disaster, just as it did in New Orleans. If sea level had only risen by the average, New Orleans would never have been flooded.
it is the same with hurricanes. They only last for such short periods at any place, roughly 24 hours, that their effect on average is practically nil, but in reality their effect can be catastrophic. If sea surface temperatures rise enough, then hurricanes could strike Norway, even if wind speeds on average do not change.
It is the variability not the mean which presents the greatest dangers.
[Response:Thanks for frank comment! I get your point. However, weather statistics is not just ‘average weather‘, but equally the number of record-breaking events and the probability that the temperature will raise above 110F. In my mind, the practical meaning of statistics is the probability density function, which can be described in terms of a small set of parameters (it’s very mathematical), where the mean and the standard deviation are key ones. But even it you were to limit the statistics to the average, you still would have to deal with the average number of storms and the average duration of droughts. Having said that, I know that statistics is not a subject that engages many people, even though this TED-talk ought to get people change their mind about that. Anyway, as long as I can show the numbers that impress people who are not afraid of numbers, such as engineers, economists, and teachers, then that is a good start. -rasmus]
Can we make it draw isotherms ? In fact does any of the portals allow isotherms, isobars and the like ? or do i have to download the temperature fields and plot isotherms myself, as i have done on occasion ?
[Response: It could be done, but this app does not have that capability (yet). When drawing isotherms, there are some choices to be made, such as which technical solution to use (kriging, triangualtion or bi-linear interpolations; should co-variates such as elevation, distance from the coast, and latitude be included?). anyway, the app will evolve in the future. -rasmus]
Ray Ladbury says
Mr. KIA: “All 13 of the folks in Africa who have laptops will be very happy to learn of this new technology.”
Racist dumbass. Have you ever been to Africa? Hell, have you ever even had a conversation with a person from Africa that lasted more than 10 minutes?
Seriously, Africa is one of the world’s most dynamic regions–both economically and culturally–at present. It has a long way to go, but unlike the US, it is actually making progress.
On the off chance that there might be other readers as stupid as you, here are some guidelines:
1) Africa is NOT a single country.
2) Different African countries differ in terms of terrain, economics, culture, resources, languages…
3) Some of the fastest growing economies last year were in Africa.
4) Not all African countries are under authoritarian strong men. Ghana and several other countries have vibrant democracies.
5) Not all of Africa is jungle!!!
Seriously, dude. Hop on a plane and rectify your ignorance. Not only will you become MUCH less annoying, you might even grow a soul.
Hans Kiesewetter says
Nice tool! I do know where to find the umbers for my home country (the Netherlands) but this tool makes it easy to “travel” to other parts of the world. Handy to show others that global warming does not mean that warming is everywhere the same. It also shows that it is not only the temperature, but also precipitation, or even could cover, is changing in our climate.
Minor glitch (read: improvement potential): the trend line in “Past Weather” does not detect that data of the first years is missing, and assumes “0” for these years in the calculations. (as example: Obdam, in the Netherlands, precipitation data started in 1981). As user you can adjust the time frame to eliminate the blank years from the calculations, but it would be better if the programming does this.
[Response: Thanks! I’ll keep that in mind when I continue work on it. -rasmus]
Russell Doty says
The idea is great, but what shows on my screen is meaningless to me. Also, some grammar differences are distracting, e.g., “statistics is” vs. “statistics are”
One comment re. the UI: should it not display what it is we are looking at? Nowhere do I see a ‘precipitation’ label.
One aspect to consider, if you want this to be used in Africa, is supporting off-line use.
My company develops data driven web sites and apps for stakeholders in (mostly) West-Africa and we get the request to support an off-line mode to this day.
Plus, if your app where to support this, people there like to do peer-to-peer sharing, so a great feature would be if one person who has downloaded the app+data could then pass it onto another person.
[Response: The app is designed so that it already can run off-line, e.g. on a laptop or a desktop. The data files can be generated manually or copied from a website. There is some more information on the info/disclaimer tab, and a link to the GitHub repository (see the wiki-page on GitHub site). -rasmus]
MR KIA @ 5, you are trolling again. Millions of africans own computers. About 13% of Africans have internet access, and ownership of smartphones is at 25 million and is increasing extremely fast.
Great. The NOAA website (Global Historical Climate Network) is not available at this time “due to a lapse in appropriation”. Fahrenheit 11/9 comes into my mind there.
Sheldon Walker says
Global warming temperature distributions
Using a single number to represent global warming, like 1.5 or 2.0 degrees Celsius of global warming, makes it hard to see how bad the problem really is. Is 2.0 degrees Celsius of global warming a major change from what we have now, or is it a minor change?
Using temperature anomalies to represent global warming, removes (or ignores) what is “normal” for temperatures. “Normal”, becomes a single temperature anomaly, 0.0 degrees Celsius. Does 0.0 degrees Celsius, really represent the “normal” temperature distribution on the Earth.
What is the solution to this problem? The answer is to look at temperature distributions, rather than single numbers. Temperature distributions make global warming multi-dimensional, rather than a one-dimensional number. Temperature distributions show how the temperature varies with latitude, elevation, proximity to the ocean, size of landmass, and many other factors.
Comparing the “normal” temperature distribution, to a “global warming” temperature distribution, makes it easier to judge the size of the problem. Are “alarmists” trying to turn a molehill into a mountain? Or are “deniers” trying to turn a mountain into a molehill?
This article will show you the temperature distributions for a range of global warming “amounts”. People with weak hearts should not look at the more extreme amounts of global warming. Seeing 10.0 or 15.0 degrees Celsius of global warming on a graph, may be too much for those with a vivid imagination.
This article offers a choice of global warming simulations.
1) with NO polar amplification
2) WITH polar amplification
I agree with zebra (comment #3). I’m having trouble telling what I’m supposed to get from this app exactly, and it gets pretty technical in the “Past Weather” tab. It definitely needs more explanation.
[Response: That’s fine. The most important objective is to make climatological data more transparent. Think of it more as a prop that experts can use when discussing the climate where people live. So, yes, there would be some need of some explanation to go along with it, and this explanation may differ depending on who you talk to and what your questions are. But it’s also a resource for people working with climate, as it provides an overview of the available stations and makes it easy to examine the climatological data. -rasmus]
Hugh McLean says
Further to comment 6, there’s one app I think most people would find both striking and easy to understand/relate to. If anyone could recreate this New York Times animation (https://www.nytimes.com/interactive/2017/07/28/climate/more-frequent-extreme-summer-heat.html?rref=collection/sectioncollection/climate) in an app that would allow users to choose their own continental/regional/local scale, choose seasons, and view both historical AND projected visualizations — WOW! I think it would provide an instant and dramatic introduction to both the reality of climate change and urgency of reining it in.
#16 Hugh McLean and Felicia,
Hugh, excellent suggestion.
Felicia, I was concerned/surprised that this thread wasn’t getting much traffic, because it seemed like the perfect opportunity for people to provide input like Hugh’s.
Rasmus in response to #3 said: “the challenge is to make something that suits a wide range of people”.
I would say: “the challenge is to resist the impulse to make something that suits a wide range of people.” In science, engineering, and visual and other design, things work better when the effort is clearly targeted/compartmentalized.
So, for, say, meteorologists, easy access to the data in table form is probably just fine, since they are perfectly capable of importing it into spreadsheets and manipulating it.
But, for teachers, who don’t have time or resources, help with visual presentation would probably be most appreciated.
My example would be that instead of the somewhat confusing map with colored dots, giving people the ability to create a kind of 3-dimensional “graph” of the stations with vertical monochrome cylinders representing the variable values would be far easier for viewers to access. I’m sure it’s relatively simple to “code”, but not for people unfamiliar with such skills or access to the software.
Anyway, I hope more people will give inputs that can be organized and incorporated into a useful app with appropriate categories of users.
t marvell says
I suggest that the best use of this app is to peddle it to local newspapers, who can print the relevant facts concerning their towns or cities. Unlike national publications, local papers still enjoy steady, wide readership, and I suspect that few of the readers would otherwise learn about the app, never mind use it.
David Piepgass says
The app is buggy.
– After choosing Calgary Intl (North America) it kept switching by itself to Aberdeen (first entry on list)
– No debounce – dragging the time range causes numerous, slow, updates
– Nonsensical trend: Choosing 1900-1950 gives trend of 0.12 C/10y, 1950-2012 gives trend of 0.29 C/10y, but the entire range 1900-2012 gives trend of 0.11 C/10y
Also, I really want to see trends for a region (e.g. Alberta or Edmonton), not a single temperature station whose name I might not even know.
I’d also suggest that the app should by default show info on climate *change* rather than reporting the current climate, info which you can easily gather with a google search like “average temperature in Calgary”
[Response: Thanks! Yes, the app is not perfect, and I have not managed to fix all bugs yet. I had to limit the data coverage for North America to avoid too much strain on the processor memory-wise. Sorry! It was originally made for Mozambique, which doe not have a problem with too much data, and Norway, with limited amount of data, has also been the primary region used for testing it. My hopes are that trying different ideas will inspire engineers and software providers to develop new apps (I’m a mere scientist playing around with some ideas and limited resources, hence the emphasis on prototype). -rasmus]
Mike Roberts says
Sheldon Walker, temperature anomalies do the opposite of what you claim. They don’t ignore the “normal” temperature at all. They show how the temperature varies from the normal temperature (where normal is the average temperature over some base period; of course, there now really is no “normal” as temperatures have been increasing somewhat for a couple of centuries, more so in the last few decades). Anomalies for different regions show us the distribution of warming.
I quite like Tamino’s alternative of using changing probabilities for looking at the effects of temperature change (and also mentioned in his later post).
Hank Roberts says
Sheldon Walker has already gotten some deserved scrutiny from Tamino
Peter coates says
Climate for any given place is no longer simply a description of typical weather over time, but has now to include current, historic, and predicted rate of change. It is a sad reflection on our state that first and second and possibly third derivatives have to be given.
I respectfully suggest the issue is not easier understanding it is better emotional resonance. Perhaps “More floods, summers without rain and slower storms–think hurricanes that stick around for a week.” would do for starters?