Greenhouse gases are top of the worldwide green agenda and all around the world industrious groups of people are seeking to reduce emissions. Whether by legislating against irresponsible fossil fuel use, encouraging energy efficiency or creating traffic-free zones (and many, many other methods besides), the only way we’ll know whether such schemes are having an effect is by knowing the rate of emission of gases into the atmosphere.  We need to be able to do this to a fine enough level of detail to be able to tell if our reduction strategies are successful or not.

This is no mean feat. The atmosphere is notoriously complicated, as is the land surface with sinks (take up gases) and sources (emitters of gases) springing up all over the place, often as quick as you can count and varying wildly according to environmental conditions.  In spite of these difficulties, scientists, industry and policy have created methodologies and taken steps towards putting some figures on greenhouse gas emission.  There are three main methods, which I will introduce in turn.

1. Emissions inventories.

I like to call this the “let’s just tot up what we know” method.  It relies on accurate figures for the extent of a certain activity (i.e. how many cars there are, how many power stations are running and for how long) and an accurate figure for how much greenhouse gas is emitted by each of these activities. These are called activity data and the emissions factor.

Flux = sum of all (activity data x emissions factor).


I call this the “tot up what we know” method because all that goes into an inventory is the emissions sources we know about.  They also usually only include human activities or have a subjective or modelled term for natural sources and sinks.  We have several uncertainties in the inventory method and they require extensive checking:

  • is the activity data right? i.e.  is our traffic count up to date?, are the power companies telling the truth about what’s coming out of that chimney?, are we up to date with all of the new houses?
  • are the emissions factors right? i.e. are we simply estimating over a large range of possible values?
  • are we summing up all of the sources we know about? what about the sources that we don’t know about? could we be overestimating the total emissions by ignoring the effect of vegetation uptake?

Because of questions like these, inventories work best in a coarse resolution (over a large area, like a country).  They aren’t checked using measurements and are best for making broad decisions (i.e. cutting back on fossil fuel energy in a certain country) than narrow ones (i.e. implementing a neighbourhood road closure and cycling scheme).

2. Direct measurement.

Measurement of greenhouse gas fluxes can happen at all sorts of scales. From a chamber placed over a square metre of soil right up to an instrument mounted on a tall tower above a city.  For the purpose of this discussion I will focus on these tall tower instruments.  The instrument on the new tower in the drawing below is designed to capture fluxes from a wide area.  So all of those terms that we put in our emissions inventory (plus or minus the influence of the sources/sinks that we ‘forgot’) are contributing to the flux measurement.


This particular system measures the difference in concentration of greenhouse gas between two vertical air packets (I will come back to how this works some other time).  This tells us the overall flux.

To be able to interpret this, we need to know where the fluxes measured by the instrument are coming from.  That is, we need to know the area of the ground that is sending gas up to be captured by the instrument.  We don’t want to consider sources that aren’t being included in the measurement and likewise we don’t want to miss any.  To do this, we need a ‘source area’ or ‘footprint’ model.  This can calculate the area of ground we can expect the fluxes to be coming from and then we can investigate how they might be influencing this measurement.

Hm, you might think this is a little vague.  You’re right, when it comes to making decisions about changing the number of cars in your city or increasing the amount of green space, it makes sense to be able to tell more exactly where the sources really are and whether or not the action has made a difference to their extent.  But if you want to monitor what the overall flux is over the city, what it really is, through observations rather than approximations, a direct measurement scheme is what you need.

How about a middle ground between the two?  A method that includes all the knowledge we have about what is out there generating greenhouse gases (inventories) plus real greenhouse gas measurements to tell us if we’re going wrong?  This brings me to the final method, which is newest to science and very much still in development.  This is one of the things that is keeping me busy during the working day right now.

3. Measurements + Model (the “inversion” method)

Inversion methodology is where we start to get really clever.  I will explain this step by step.

a. Take the inventories and make it as accurate as you possibly can for your chosen area.

b. Place an instrument on a tower that measures concentrations of greenhouse gases.  (Note: this is concentration not flux.  The instrument tells us what the concentration of gas is, not the change in that concentration).

c. Use a meteorological forecast model to tell you what the winds are doing throughout the surface km of atmosphere.

d. Place the inventories and the wind data (plus some other information such as the height of the boundary layer and the surface roughness) into a chemistry-transport model.  This model can give you an idea of what the concentration of the greenhouse gas in question is EXPECTED to be.

e. (And here’s the clever bit) Use the real data from the atmosphere, plus the expected concentrations and then run the model BACKWARDS.  This “inversion” of the method uses everything we know to ‘go back to source’ and tell us where we should be checking our inventories a little more closely.


This last method has a lot of potential. But right now it is in its pilot stages.  New science is consistently being published in developing this method and industry is becoming aware of the economic opportunity in providing governments with such a service.

The method that an international group/national govenment/local authority will choose depends on what they wish to get out of the investigation. Benchmark figures? Real measurements? In-depth analysis? Participating in new development of new scientific technology? The future is bright for estimation methodology, so don’t let “how will we even know if it works?” put you off taking your bike to work instead of your car.

In the next article at Ground to Sky, I will be discussing the ups and downs of wading through huge piles of scientific literature and providing some tips from my experience.  Following this I will return to the technical science and will be discussing some of the methods that we can use to cool down our hot cities during the summer months, without compromising winter warmth.