This is the first of two posts on some large issues connected with global fire regimes, biomass flows, and food security.
Boverty is the human impact of too many bovines overwhelming the local biosphere’s ability to feed them … the bovines are usually cattle and more than a few African countries have precisely this problem. Their livestock is a millstone around their necks and helping to keep them poor. Well-meaning aid organisations often contribute to the problem.
The ecosystem impacts of cattle spread far and wide but it may not be the owners of the animals who suffer the impacts. Indeed, the animals can buffer their owners against the worst impacts of boverty. This is analogous to the way that drivers of large SUVs do well in collisions with smaller vehicles. The entire community suffers from the presence of the vehicles, but the owners may be the least affected.
But these conclusions are just the end point of a longish discussion. We need to start at the beginning. But before we get to the beginning, here is a MODIS satellite firemap of the planet during the last days of December 2009. The sub-Saharan cattle countries are ablaze.
This post surveys the impacts of livestock, firstly at a very general level on the biosphere due to its domination of global biomass consumption, proceeding through the cattle-specific annual planetary conflagrations as people ignite the world’s grasslands to prevent reforestation. Lastly, we look at more intimate and sometimes more indirect bovine impacts, like the accelerated degradation of arable soil, the tens or hundreds of thousands of children killed by cooking with dung, and the global increase in respiratory and heart disease from ozone increases caused by rising methane levels.
Cattle are a major causal component in all these problems. The planet’s 1.4 billion cattle have a liveweight biomass exceeding that of humans and dominate many of our adverse impacts on planetary eco-systems.
Over the past couple of decades, a variety of measures of our impact on the planet have emerged. Eco-footprints seem to be the sexiest of these, with great logos and a resonance with old folk-lore about treading lightly on the planet. I’ve always felt this measure was conceptually flawed. Converting things as different as what you had for breakfast along with electricity and water usage into a square kilometer figure is bizarre. Happily, an older, clearer measure is now making a comeback, thanks to the increasing power of those remarkable spies in the skies, the satellites that between them can weigh Antarctica, measure fire scars, spot ocean bottom trawling damage, and check out which of your neighbours has a swimming pool.
The old measure is based on photosynthesis. Photosynthesis produces plant growth and underpins almost all life on the planet. Every year the planet produces a huge quantity of plant growth, we eat some of it, other species eat some of it, and the rest either forms long-lived things like trees, or is more quickly broken down, returning its nutrients back to the soil, water and air. How much of this growth do we use?
An Austrian team has been turning out some remarkable papers cutting and dicing our usage in considerable detail using both satellite data and more normal statistical sources. Initial work by Helmut Haberl and the team culminated in a 2007 paper estimating that we use about 23.8 percent of the planet’s annual plant growth … otherwise known by the catchy name of net primary productivity (NPP). The language of the preceding sentence was a little sloppy, … we don’t actually use all of this, but we certainly appropriate it in ways that will slowly become clear.
NPP is measured as dry matter (DM). This is what’s left when you get rid of the water. If you didn’t do this, 10 kilograms of water melon would count the same as 10 kilograms of rice. The choice of this as a measurement unit is important and sensible because dry matter plant material is about 50 percent carbon. This means you can easily convert between carbon and dry matter quantities. So when I say that global NPP was 118.4 billion tonnes of dry matter in 2000, this means that about 59.2 billion tonnes of carbon was sucked out of the sky that year by plants using photosynthesis.
Even though this measure is conceptually clearer than eco-footprint areas, there are plenty of gotchas which can trick new jugglers of these numbers. First, only about half the NPP is above ground. So plenty is simply unavailable. We dig out potatoes but not tree roots. So when you cut down a tree for timber, the roots aren’t used, but they still count as appropriated. This is reasonable because the tree certainly can’t use them any more. Secondly, the NPP of an area isn’t fixed. Humans do things which change the NPP of land. They pour on fertiliser and pump in water which raises NPP. They pave paradise with parking lots or graze it to a dustbowl and NPP drops to zero.
The Austrians estimate our land degradation and pavement have reduced annual global plant growth (NPP) by about 6.2 billion tonnes of carbon. That’s a pretty significant number for a couple reasons. First, because it’s not that much below the amount of fossil fuel carbon we emit annually … which is up around 8 billion tonnes! Second, because its higher than the net annual land clearance emissions of 1.47 billion tonnes. The estimated carbon released to the atmosphere due to deforestation over the industrial period is 200 billion tonnes. These numbers indicate that managing the biosphere to draw down a couple of decades of fossil fuel emissions is possible and we have two mechanisms at our disposal: reforesting areas we have deforested and thickening current vegetation … enhancing NPP. These mechanisms, used to their theoretical maximum, won’t make rebuilding our energy infrastructure unnecessary, but they can buy time. We shall see in part two of this post that rebuilding and extending our energy infrastructure to poor nations is probably essential if we are to successfully reforest the planet.
Another Austrian permutation, headed this time by Fridolin Krausmann, has done more work on these datasets and has produced biomass flow data with breakdowns by country.
This shows that globally, we only eat about 12 percent of the 12.1 billion tonnes of plant material that we either crop or have our livestock graze. This provides 83 percent of global food calories. Livestock eat 58 percent of that 12.1 billion tonnes and provide the other 17 percent of calories. What about fish? Fish are just 1 percent of global calories and part of the 17 percent.
Note that the 12.1 billion tonnes doesn’t include biomass incinerated in deliberately lit fires … this is important later on.
Australia, all by itself, appropriated 468 million tonnes of plant growth in 2000. We harvested it for paper, animal feed and timber and much else besides. Our livestock eat about twice as much of what we harvest as we do and that obviously counts as our appropriation. But they don’t just eat the bulk of the harvest, they graze another 30 times more. But that’s not the end of their impact.
In my last BNC post, I referred to an estimate just published as a WorldWatch report that put the impact of livestock on greenhouse forcings at about 51 percent of the global anthropogenic total. I didn’t analyse this figure, but suggested it wasn’t unreasonable to think that land clearing, feeding, watering, housing, slaughter, transport and cooking implicit in dealing with 700 million tonnes of livestock biomass could conceivably be responsible for half of the total climate impact of 335 million tonnes of human biomass. But I was waiting for more expert analysis and still am. Many of the additions over and above the 2006 Livestock’s Long Shadow (LLS) estimate rely on close knowledge of the precise details of the FAO’s statistical data collection processes.
But on theoretical grounds, one of the most contentious inclusions in the 51 percent figure is livestock respiration … the carbon dioxide that livestock exhale. On the face of it, this seems plain wrong. All of the carbon dioxide in livestock respiration comes from the atmosphere via photosynthesis in plants. So it’s simply part of the carbon cycle. Isn’t it? The WorldWatch authors have subsequently justified this a little further, re-citing evidence given in LLS which states that animal respiration plus soil carbon oxidation (co2 flowing into the atmosphere) exceeds the drawdown due to photosynthesis by one or two billion tonnes of carbon annually. In many cases it is livestock driving the loss of soil carbon by deforestation and desertification and given that the planet’s 700 million tonnes of livestock dwarf wildlife by a ratio of about 23:3, it is possible that the planet’s total plant biomass may be shrinking under livestock’s onslaught. This is the implication of the reduction of NPP noted above and the carbon flow imbalance just mentioned.
I say may be shrinking because it’s tough to measure things like global photosynthesis or global respiration, and the figures in LLS are not the same as the figures in the Austrian work. Close, but not the same. But if the respiration plus soil carbon losses really are outstripping photosynthesis, then including at least some livestock respiration in the ledger isn’t just reasonable, but mandatory.
In any case, not all parts of the carbon cycle are currently excluded from national greenhouse inventories. Livestock methane is part of the carbon cycle and everybody includes that in their inventories … for good reason. Turning carbon dioxide (CO2) into methane (CH4) doesn’t increase the carbon in the atmosphere but, in effect, puts it on steroids for a decade as far as its warming effect is concerned.
Similar considerations apply to fire. Under IPCC accounting principles, CO2 emissions from fire are ignored unless the fire changes the underlying vegetation. For example, a fire in a savanna doesn’t permanently change anything, the grass comes back. But a fire that clears a tropical forest to make a pasture results in a net permanent reduction in standing carbon (the trees!) which is added to the atmosphere.
Deforestation also produces soil changes. Soil can be viewed as an organism in its own right. Its microbial inhabitants transform soil matter and emit or absorb the greenhouse gases that dominate our current concerns. There are many types of soil and zillions of types of microbes in constant evolutionary flux so getting a handle on what is happening is like holding a bowl of jelly with chopsticks and no bowl.
Anyway, most tropical soils under forest act as methane sinks but lose this property when the forest is gone. Similar results have recently been demonstrated in Australia in temperate, Mediterranean and subtropical regions. When paired sites at various stages of forest and pasture growth were compared, the trend was for nitrous oxide emissions to be lower from forests than pasture, with methane absorption also lower in pasture than in forests. So forests did more of what we want than pastures in both cases. Again, this is complex soil chemistry and other studies have found the opposite with regard to nitrous oxide.
Back in 2006 a study shocked the scientific community by claiming that living plants can produce methane. This prompted an immediate claim from a New Zealand scientist, probably with an eye on his local sheep industry, to claim that forests may have produced as much methane as the ruminants which displaced them. Unfortunately for the New Zealand sheep industry, someone was rude enough to actually do the calculations, and based on the proposed new methane source, show that the livestock emissions were 16 times bigger than the forests they replaced. As it turns out, it seems plants don’t produce methane, but they can transport methane generated in the soil.
The unquantified false claim about ruminants producing less methane than the forests they replaced is a great example of an idea which sounded plausible until the numbers showed otherwise. I’ve written previously about Tim Flannery’s plans to provide abundant meat to the planet by expanding cattle production. This is another example of a plan that becomes laughable (or more correctly cryable) when you do the numbers. Apart from the fact that the current 1.4 billion cattle provide just 1.4 percent of global calories, the injection of another 96 million tonnes of methane into the atmosphere by providing Australian levels of beef to most of the planet (excluding India) would make winding back climate forcings even harder than it is presently.
Apart from any nitrous oxide that may be emitted by soils, once cattle are added to the pasture, the nitrous oxide emissions from the cattle droppings are substantial. A global study estimated that livestock waste represents 30-50 percent of global agricultural nitrous oxide emissions. This is in addition to the emissions from the feed crops, many of which are now fertilised with nitrogenous fertiliser.
Note that for either a savanna fire or a forest fire, the methane and black carbon from the fires generate net climate warming. Methane, and a few other gases from such fires are recorded in national greenhouse inventories, but black carbon isn’t because it isn’t regulated by the Kyoto protocol. More on black carbon later. Methane from savanna burning is listed by Australia in its greenhouse inventory, but not by some developing countries, even when they do massive amounts of burning. For example, Sudan lists no methane from savanna burning in its only communication with the UNFCC in 2003, but Nigeria and Ethiopia do.
In most places in the world, most fires (80-90 percent) are deliberately lit by people. The major exceptions are Russia, the US and Canada where Boreal forests are regularly ignited by lightning. Australia has some of these kinds of fires also, but less commonly because we have less lightning … as indicated in this global map.
Most lightning runs from cloud to cloud, so is irrelevant to ground fires and, as far as I know, satellites can’t pick a ground strike from a cloud to cloud flash, and this map (despite the title) is actually of flashes, not ground strikes.
Tim Flannery recently speculated that removing or reducing herbivores would lead to more fires and a paper last year pointed out that wildfire and insects have turned Canadian forests into a source of carbon rather than a sink. The same paper estimates that the historical deforestation of the planet has added 200 billion tonnes of carbon to the atomosphere. Can everybody see the blazing flaming contradiction here? If we had 200 billion tonnes of carbon worth of forests before we deforested the planet for livestock and the much smaller areas that we crop and live on, where were all the wildfires back then? Certainly we had no firefighting planes and helicopters back when those billions of tonnes of forest were standing. Certainly we had no huge armies of cattle and sheep in Australia at the time before we cleared 100 million hectares. Why didn’t fires burn it all back then? Maybe we did have more natural fires, but with so much more forest, the carbon impact was of no consequence.
The main traditional driver of deliberate human fires has been to clear land and keep it cleared for livestock grazing or cropping. The latter is usually called slash and burn, or shifting cultivation. It’s a cheap and effective method. The collateral damage is generally limited to wildlife and provided Steve Parish has been and taken his pictures for all those airport tourist calendars, what other use does wildlife have? Traditionally, hunting wildlife was the third prime driver of burning. We shall see below that scientists estimate that currently about 2/3 of burning is for livestock grazing.
From a climate perspective, all three kinds of fires represent foregone biosequestration, with the first being a direct climate cost of livestock.
More recent work in the Austrian series refines the estimates of biomass burned through anthropogenic fires with better estimates on the type of burning and better country level breakdowns. Lauk and Erb’s estimates slice fires into two kinds: big fires and little fires. The big fires are almost entirely the livestock fires we have discussed. The small fires are shifting cultivation … plant food fires.
Estimating the extent and impacts of both is difficult and only possible because of new datasets on global vegetation. The satellite data showing what is on the ground can be compared to other global data on potential vegetation and also with satellite data on burn scars and actual fire detection using thermal imaging. Big brother is not just watching you, but watching your back paddock as well. The data on potential vegetation is derived from a global vegetation model which models a raft of processes using input such as current cover, soil type data, temperature and rainfall.
Globally, the big fires release about 2.5 billion tonnes of carbon. N.B. this is a carbon figure. The small fires release between 1 and 1.4 billion tonnes of carbon. There is a largish range because it’s much tougher to estimate the small fires.
If this carbon was balanced by photosynthesis it wouldn’t be a problem would it? Yes and no. Provided the quantity burned each year is constant and vegetation levels are globally maintained, then it’s not causing a net carbon increase in the atmosphere. But are both these quantities constant? The technology is a long way short of giving a real-time read out. Most of the figures I’m presenting are for a single year, 2000. The fires will of course put additional carbon on steroids and produce plenty of other nasties. The Edgar methane inventory lists methane from savanna burning at about 7 million tonnes, probably a little under the true value, but close. This is equivalent to a population of about 60 million cattle grass fed cattle.
Included in the total of plant growth appropriated by Australia is biomass we deliberately just burned. Apart from firewood, most burning in Australia is in deliberate fires set in large regions in the north of Australia every year. The now renamed Australian Greenhouse Office calculated that some 75 percent of this burning was for cattle. This is pasture burned to keep forest regrowth at bay. We are, of course, happy for Indonesians and Brazilians to have tropical forests, but we’d rather do something more useful with our northern regions than merely mop up carbon and provide habitat for wildlife. So we set fire to it. Rainforests can and are expanding in North Queensland into areas no longer subject to human burning. In other areas of tropical Queensland grasslands have changed to closed forests with the cessation of human burning.
That mass of top end burning counts as part of the Australia’s total appropriation of 468 million tonnes of plant growth. How big a part? About 40 percent … some 139 million tonnes DM. All up, we burn slightly more biomass in northern Australia than our livestock graze over the entire continent during the whole year.
But in the burning stakes (or should that be steaks), we are small fry. The global burning picture is massive and has implications for both climate change and food security. Here is a MODIS satellite fire map from the end of July 2009. It’s worth visiting the NASA website to look at when different regions of the planet get burned. Higher resolution maps would show individual fires and not the solid contiguous region that is shown in this image.
What could limits on global burning regimes do? Globally, we burn about 3.7 billion tonnes of dry matter annually. If we reduced this burning to perhaps 2 billion tonnes, which is possible (but hard) and desirable for many reasons, then we could absorb about 1 billion tonnes of carbon. In the first year we did this, we would sequester about an eigth of the fossil fuel carbon emitted each year. As time went on, forests would regrow and absorption rates would slowly fall. As a mitigation strategy, this is significant. Not a single handed planet saver, but useful.
The next part of this two part blog deals with the continent which burns 2 billion tonnes of dry matter annually, a country of chronic undernutrion, poverty and large scale boverty. The next post is on Africa.