Peatland and modelling

Have CSERGE developed Earth's very own Deep Thought?

Peatlands are important. It is estimated that Northern peatlands store more than 455Pg of carbon (Gorham 1991), which, according to my calculations, is roughly twice as much as the carbon stored in the world’s tropical rainforests[1]. So yeah, these soggy, waterlogged landscapes are pretty important.

The large swathes of peatland that now exist in the circum-arctic region are believed to have developed during the late Pleistocene and early Holocene as ice sheets retreated and the climate became warmer and wetter. The hypothesis that climate has been the prime driver of peat formation is supported by Macdonald et al (2006) who use radiocarbon dating to establish the timing of peat development across northern Europe. They find that peatland expansion across all continents occurred in a pattern that reflects fluctuations in global temperatures with initial expansion during the Bølling-Allerød warm period (c.14.7–14.5 ka bp) slowing during the colder conditions of the Younger Dryas (c.12.65–11.5 ka bp) before rapidly increasing after the end of the Younger Dryas at around 11.5 ka bp.

Evidence from the UK, however, indicates anthropogenic impacts have also been important causes of peatland formation. For example, Oldfield (in Walker and West (eds) 1970) has shown that Blelham Bog in the Lake District was artificially established in the early 19^th century as a result of human flooding of former peat cuttings with subsequent stream diversion throughout the 19^th century encouraging further silt deposition and, hence, peat formation.

While the contribution of humans to peatland therefore appears variable, it is clear that anthropogenic impacts are the overwhelming drivers of contemporary peatland destruction. In Europe alone, it is estimated that out of a total mire and peatland area of 617.000km2, 52% has been converted over last century with the main areas of conversion including agriculture (50%), forestry (30%) and peat extraction (10%) (Byrne et al 2004). It's clear that future land-use decisions should preserve this significant carbon sink, but how can we ensure that such decisions are made?

Here I believe that recent developments in environment-economy modelling can, again, be instructive. Work is being undertaken at The Centre for Social and Economic Research on the Global Environment to develop integrated land-use models that will allow decision-makers to measure the impacts of land-use changes not only in terms of changes in market-priced commodities, such as agricultural output and timber, but also non-market benefits such as the recreational services of forests and, importantly, the carbon sequestration benefits of peatland.

Bateman et al (2013) provides an example of how such modelling developments can promote pro-peat land-use decisions. The authors use a series of decision models to explore how land use patterns change under various scenarios defined in terms of the magnitude of climate change and the strength of environmental regulations. Valuation models are then used to determine the magnitude of changes in market and non-market benefits that are the result of land use in each scenario. The images below show the results from the ‘world markets’ scenario in which environmental regulation and policy are weakened unless they coincide with improved agricultural production. Looking at the maps, the difference between market (agricultural) and non-market (ghg emissions and recreation) outcomes is explicit. Chopping down trees and converting peatland to agriculture would be expected to increase agricultural output but the maps below show how this also causes increased carbon emissions and losses in recreational benefits (going for a walk in a field of potatoes isn’t quite the same as going for a walk in an ancient forest).

Images from Bateman et al (2013)

These maps might look nice but what’s the point of this? Bateman et al (2013) show that basing land-use decisions solely on market values will have considerable detrimental effects and, while this result is pretty unsurprising, the model developed by CSERGE makes the point impossible to ignore.

I do, however, have some reservations. The novel aspect of Bateman et al’s modelling approach is the combination of a land-use decision model with a valuation model. This allows all the impacts of land use changes to be expressed in terms of a single monetary unit (note that the maps above are all in terms of £/ha/yr). To me, there seem to be two issues with such an approach. The first is that this means that modelling results are inevitably going to depend on the monetary values that are placed on non-market benefits such as the carbon sequestration services of peatland. I’ve written before about the importance of such environmental valuation but it’s important to recognise that the model’s conclusions will always be contingent on the valuation methods used.

While I believe that environmental valuation has many benefits, my second, and more significant, concern is that policy makers place too much emphasis on such modelling exercises. I worry that politicians are far too keen to hold up the results of such models as facts that make further deliberation redundant, ignoring that the results are actually contingent on subjective valuation method decisions.

So yes, environment-economy models such as that developed by CSERGE can aid pro-peatland decision-making but it’s vitally important to use them correctly as tools to inform further debate rather than infallible oracles that always produce The Truth.

---

[1] To put the peatland figure in context I used NASA’s estimate of the total carbon stored in tropical forests (247 billion tons) and used this converter to express the figure in petagrams. I am pretty ignorant about comparative carbon storage estimates so if this appears way off to anyone please let me know!