Ecological Footprints – a good idea gone bad

Guest Post by Willis Eschenbach

In “Another day, overshot to hell” Anthony Watts commented on the “Overshoot Day” promoted by Mathis Wackernagel and the Global Footprint Network (GFN). This is based on the idea of the “ecological footprint”. Your “ecological footprint” (EF) is how many acres (hectares) of land it takes to support you, to grow the grain for your bread and the timber for your house and so on. It’s a simple and visual way to measure our impact on the planet.

Unfortunately, the particular form of the EF as advanced by Mathis Wackernagel and the GFN contains three fatal flaws. It wildly underestimates the available rain-fed cropland. It assumes that people in Britain farm like people in Africa. And it arbitrarily assigns huge weighting to CO2.

Figure 1. The effect of CO2 on the Wackernagel version of the “Ecological Footprint”. Image from Bambi meets Godzilla, a cartoon worth watching.

Here’s the stern warning from the Global Footprint Network folks:

Earth’s Overdraft Notice

On August 21, we exceed nature’s budget

It has taken humanity less than nine months to exhaust its ecological budget for the year, according to Global Footprint Network calculations.

Today, humanity reaches Earth Overshoot Day: the day of the year in which human demand on the biosphere exceeds what it can regenerate. As of today, humanity has demanded all the ecological services – from filtering CO2 to producing the raw materials for food – that nature can regenerate this year.  For the rest of the year, we will meet our ecological demand by depleting resource stocks and accumulating greenhouse gases in the atmosphere.

“If you spent your entire annual income in nine months, you would probably be extremely concerned,” said Global Footprint Network President Mathis Wackernagel. “The situation is no less dire when it comes to our ecological budget. Climate change, biodiversity loss, deforestation, water and food shortages are all clear signs: We can no longer finance our consumption on credit. Nature is foreclosing.”

First, the mandatory disclosure of personal interests.

Mathis Wackernagel was the co-developer of the idea of the “ecological footprint”. He has since built it into a business called the Global Footprint Network (GFN), licensing out the software to do the calculations. From 2001 to about 2005 I discussed these issues both in person and by email with Mathis. I invite him to respond on this thread, and I’m sure that I speak for Anthony when I say that he is more than welcome to write a guest post for WUWT. Onwards to the issues.

Ecological Footprint

The basic idea of the ecological footprint is simple, and quite interesting. How much land does it take to support you? How many hectares of land are required to grow the wheat for your bread, or the strawberries for your cereal? Add up your wheat footprint and your strawberry footprint and all the other footprints for whatever you consume, and you have your own ecological footprint.

So for example, suppose the local land annually produces one bushel of wheat per square cubit of area farmed. If you eat three bushels of wheat per year, your wheat footprint is three square cubits. That’s how much land it takes to produce the wheat you ate. (Of course we’d use modern measures.)

The formula for calculating the ecological footprint EF for a given product can then be seen to be

EF = Consumption / Yield

or including units,

EF (hectares) = Consumption (kg or tonnes per year) / Yield (kg or tonnes per hectare per year)

So that’s the footprint plan, and an interesting plan it is. However, as always, the devil is in the details. In this case the details are how Mathis and the GFN define certain values.

ISSUE 1: Underestimating Available Cropland

One of the central questions to be answered is, how much cropland do we have on the planet, used and unused? Available cropland means land that has the rain and the soil and the temperature and the other criteria to allow rain-fed agriculture. Mathis and the GFN say that the world is nearly out of cropland, and that’s one of the reasons that they say we are at the ecological end-of-times.

In the GFN calculation of the ecological footprint, the amount of land on the planet that is available for use as cropland is taken from the United Nations Food and Agriculture Organization (UN FAO) website. What GFN use as available cropland is the FAO category called “Arable land and Permanent crops”. This is a huge misunderstanding. Those FAO categories are defined in the FAO Glossary as: (emphasis mine)

Title

Arable land

Definition

Arable land is the land under temporary agricultural crops (multiple-cropped areas are counted only once), temporary meadows for mowing or pasture, land under market and kitchen gardens and land temporarily fallow (less than five years). The abandoned land resulting from shifting cultivation is not included in this category. Data for “Arable land” are not meant to indicate the amount of land that is potentially cultivable. Data are expressed in 1000 hectares.

Title

Permanent crops

Definition

Crops are divided into temporary and permanent crops. Permanent crops are sown or planted once, and then occupy the land for some years and need not be replanted after each annual harvest, such as cocoa, coffee and rubber. This category includes flowering shrubs, fruit trees, nut trees and vines, but excludes trees grown for wood or timber.

The FAO specifically says that “Arable land” does not mean available potential cropland, as Mathis and the GFN maintain. It is no surprise that they think we’re out of cropland — they are using the wrong figures for available cropland, despite a clear warning from the FAO not to do just that.

The premier study on this question is again from the FAO. It is called the “Global Agro-Ecological Zone Study” or GAEZ Study. It is a study using global digital terrain maps, soil maps, precipitation and temperature maps, and other global databases. They identified “agro-ecological zones”, that is, zones which have common types of vegetation and soil, and thus are suitable for one or more agricultural crops or crop combinations. The result is a database with a huge amount of information on everything from desert areas to amounts of cropland to the area occupied by cities and roads. Here, for example, are the climate constraints (as opposed to say soil constraints) on plant growth:

Figure 2. Climate constraints on plant growth, per GAEZ. Plate Source.

The GAEZ study has some fascinating results in a variety of charts, tables, and maps. For example, there is more available rain-fed cropland sitting unused in South America than there is land under cultivation in North America. And there is enough unused rain-fed cropland in Sudan (over 75 million hectares) to feed every person in Africa. Some areas are short of available cropland, but the world as a whole is not short of available cropland.

Meanwhile, Mathis and the GFN claim that the Sudanese have no more land to farm. They say the Sudanese are farming almost 100% of the cropland available. By contrast, the FAO says that Sudan has 92,391,000 hectares of land suitable for rainfed agriculture, and it also says that they are farming 16,433, 000 hectares, or only 18%, of that land.

So when the GFN folks say we’re running out of cropland for agriculture, don’t be fooled. The Earth still has a lot of cropland. Our footprint is large … but not that large. I encourage everyone interested in the subject to read at least the GAEZ Summary.

ISSUE 2: Do the Germans farm with wooden plows?

Mathis and I are deeply divided on the next question. When you calculate say the wheat footprint of a country like say Russia, or Cambodia, should you use the local yield of the wheat they actually ate, or the global average wheat yield?

Mathis uses global average yield. I say use the actual yield from wherever the food was grown.

I say that if the folks in Scotland grow three thistles per hectare of land, and they eat a twelve thistles per year, that their thistle footprint is four hectares. It doesn’t matter if people somewhere else on the planet get six thistles per hectare of land, or if they can only grow one thistle per ten hectares. It doesn’t matter what the global average thistle yield is. I say the hardy Scots have a thistle footprint of four hectares.

Mathis says that … well, I’m not sure what his arguments are these days. He used to argue that using global yield rates compensated for the fact that some countries have good land for thistles, and some don’t. In fact, it was in researching that claim that I came across the GAEZ study. The GAEZ study actually has on-line, country-by-country data regarding cropland availability and cropland quality. Using that data, I was able to show that the use of global average yield did not compensate for varying cropland quality. That news was not well received by Mathis, and may not be unrelated to his cutting off all further communication with me.

However, there is a more subtle and devastating problem with their use of global average yields. In our example above, suppose the global average thistle yield is only one thistle per hectare. Using Scottish yields, we get the Scottish thistle footprint (consumption / yield) of four hectares. That’s a consumption of twelve thistles per year, divided by the Scottish yield of three thistles per hectare per year, equals four hectares.

But if we use the global average thistle yield of one thistle per hectare, suddenly the Scottish thistle footprint jumps up to twelve hectares! In other words, using their method, Scotland is getting penalized because they are better at growing thistles than the world average.

So when the Global Footprint folks say that England, or the US, or any industrialized countries have huge footprints, that’s absolute nonsense. The footprints of all the industrialized countries are artificially inflated by GFN, purely because they are efficient and get high yields. And the higher the yield, the higher the penalty. A country producing wheat at four times the global average wheat yield has its wheat footprint multiplied by four.

And conversely, if a country has a yield that is lower than the global average, their footprint gets artificially shrunk. Shrunk! Some countries use more hectares than the rest of the world for a unit of production, and for that the number of hectares of their footprint is reduced? I don’t think so …

I pointed this all out to Mathis. He ignored it. My conclusion was that any measuring system that penalized efficiency and high yield, and rewarded low efficiency and low yield, was off the rails. Too bad, the ecological footprint was such a good idea at the start.

So yes, the Scots would have a large thistle footprint … but only if they farmed with oxen and wooden plows like the global average farmer. But they don’t, and that’s the point. Scots farmers are both hard-working and canny, it’s a verifiable and well-attested fact. By farming hard and farming smart they have reduced their thistle footprint. It takes less land to produce their thistles, and good on them. Inflating their footprint because they are more efficient than the world average is nonsense.

ISSUE 3: Overestimating CO2

In any measure intended to show total impact, like the ecological footprint (measured in hectares or acres), we may choose to include things that can’t be measured in hectares or acres, things that have no “area”. For example, one might want to include say river pollution in the footprint … but how does one measure dead fish in hectares? You can’t.

As a result, if you want to include those incommensurate factors, I say you need to divide your analysis into measurable (wheat yield) and incommensurate (river pollution) sections. Because once we leave the measurable, we are in the world of “pick a number, any number”. Someone who is passionate about rivers will say river pollution should translate into a large ecological footprint, lots of hectares per unit of pollution. Someone who is not passionate may give it a smaller number. Mathis and the GFN folks give river pollution a value of … well … zero. Pollution, according to them, has no ecological footprint. Curious, huh? The ecological footprint as used by Mathis and GFN assigns zero footprint to air, land, or water pollution.

What they are passionate about, of course, is CO2. So it is accorded a huge footprint. The “CO2 footprint” is the reason that their model says we’ve … what was it? … “exhausted our ecological budget”.

But that’s just picking a number. They picked zero for the ecological footprint of say a mine that kills all riverine life downstream of the mine. They picked a big number for CO2. It’s just picking numbers, because there is no way to measure either dead fish or CO2 in hectares.

CO2 is in the incommensurate section of the EF indicator, not the measurement-based section. If we look at actual measurements, we’re nowhere near exhausting our ecological budget. That is an illusion sustained by the high conversion numbers for CO2 into hectares.

CONCLUSIONS.

1. Mathis and the GFN say we are running out of cropland. Not true.

2. Their use of global average yields artificially reduces the footprints of inefficient nations practising antiquated, low yield agriculture. At the same time, it artificially increases the footprints of high-yield nations. Double not true, or more.

3. The true footprint for any product is calculated using the actual yield figures for wherever that actual product was grown. Using any other yield than the actual yield figures for that particular product gives us distorted results, as discussed immediately above.

4. They combine actual measurable data with CO2 data, which cannot be measured in hectares. Not wrong, just 100% subjective, and should be flagged “WARNING: Contains Absolutely No Science.” …