Will air pollution from biomass heating damage our health?

Biomass boilers are low carbon but will they bring back smog?

Image from http://www.gilliesgas.co.uk/palazzetti/stoves/

Our two main low carbon heating options are electric heat pumps and biomass boilers and of the two biomass heating is a lot more low carbon than heat pumps, at least until we get more low carbon electricity. However, burning biomass is much dirtier than gas or oil in other ways. In particular it generates tiny particles – like smoke – that get into our lungs and can cause health problems. They are particularly bad for people with breathing problems such as asthma or chronic obstructive pulmonary disease (COPD). This issue came up in a discussion with the Transition Cambridge Energy group recently and it also relates to my research on heating in rural areas. So I have been looking into it, to see if there is a potential problem or not. It seems domestic biomass boilers are not too much of a concern, but wood stoves could be.




NOx and SOx from biomass looks OK.
As well as the particulates issue, wood burning of any kind generates nitrogen oxides and sulphur oxides. Sulphur emissions depend on how much sulphur there is in the fuel and for biomass that is typically around 0.1% [1]. This is much more than in gas but it is similar to the permissible levels in oil [2] and considerably less than in coal, which is typically 0.4-0.8% [3]. Nitrogen oxide emissions depend mainly on the temperature of combustion because there is always plenty of nitrogen present in the air. Judging by tests on a range of boilers in the USA, biomass boilers typically generate less NOx than gas or oil boilers [4].

We already have noticeable particulates emissions from wood stoves.
However, you may have noticed the smell of wood smoke on cold winter evenings – the levels of particulates in the air in some areas do show trends strongly suggesting that wood stoves are a significant source. The chart shows mean levels of PM10 (particulates smaller than 10 µm) in Norwich, with strong peaks in the evenings and especially at weekends, when people are most likely to be using their stoves [5]. Of course some of this could be due to coal burning, or wood on open fires: they are not so clean burning as stoves.

Mean winter time levels of PM10 Norwich, 2009-2011 [5]

But the levels seen are well within allowed limits.
However, the highest levels seen in Norwich were only 6 µg/m3. This is barely more than a tenth the level allowed under EU air quality standards - the limit for PM10 is
50 µg/m3, averaged over 24 hours. There is also a tighter limit for the smaller PM2.5 particles, 25 µg/m3 averaged over the year. So it seems that pollution from wood stoves isn’t a health issue yet.

Biomass boilers are much cleaner than wood stoves.
There is a big difference between biomass boilers, usually burning wood pellets or wood chips, and wood stoves. Biomass boilers are designed to be highly efficient and burn the fuel in tightly controlled conditions. To qualify for subsidy under the RHI, your boiler has to be certified to emit no more than 30g/GJ of particulate matter [6]. For an average home, in January, that means about 9g/day [7]. For wood stoves, on the other hand there are no limits at all except in clean air zones where you are supposed to use only approved stoves. For a small 5 kW stove, the permitted level is about 370g/GJ: more than ten times that from the biomass boilers [8]. Other wood stoves not approved for use in clean air zones are presumably even worse.

The overall increase in PM10 due to biomass boilers under the RHI will be negligible.
The overall impact of biomass boilers subsidised by the RHI is expected to be something like 0.1 µg/m3. This is the increase in PM10 population weighted mean concentration calculated using the National Pollution Model [9]. Under this scenario, biomass boilers are responsible for about 1500 tonnes of particulate emissions annually. The report did not say how many homes this represents – I reckon it would be about 1.3 million homes or about 5%.

In bad weather conditions, the effects will be greater.
However the effect of emissions is incredibly weather dependent. The dreaded pea-souper London smogs only lasted a few days. The problem comes when there is no wind to speak of and you get a temperature inversion that keeps the air from mixing upwards. I have done some simple modelling of a worst case scenario:

• 65 homes/ha – typical for Victorian terraces
• 10% of homes using biomass boilers
• 61 kWh/day heating requirement – the mean for terraced homes in January, the worst month [7]
• Emissions at the limit (30g/GJ) 
• No wind
• Air mixing to 200m altitude (during the London pea-soupers the inversion ceiling was 100-200m [11])

With these assumptions I calculate the increase in particulate emissions from 1 day of heating to be about 25 µg/m3 – so half the limit. But these conditions are rare. Even a 1 km/hour wind is enough to disperse the emissions thoroughly. In Cambridge there were 5 days in the last 5 Januarys when the average wind speed was below 1 km/hour. There was only one instance when the wind was so low for two days running. If there is just a little wind so that the emissions disperse across Cambridge, the impact is much lower because the mean housing density is only about 12 homes/ha [12]. Even allowing for the slightly higher average home energy demand, the PM10 levels increase by only 5 µg/m3.

However, a similar calculation with wood stoves, assuming 10% of homes have a 5 kW wood stove (certified for the clean air zone) and run it at 2/3 capacity for 5 hours, gives an increase in PM10 of 87 µg/m3 for Victorian terraced housing. Dispersed over Cambridge this would be 16 µg/m3. (See chart).

Impact of particulate emissions from biomass boilers or wood stoves (clean air zone approved) from one day of heating in January averaged over the whole of Cambridge, or just over high density housing such as Victorian terraces.

It looks as though biomass boilers are unlikely to cause serious air pollution problems, at least as long as we don’t have more than 10% take-up in high density housing areas. Wood stoves, however, could be a serious problem if they become popular.

In our report for CPRE we discussed the impacts of low carbon heating on our landscape such as vast areas given over to energy crops. Here is another reason why simply replacing our energy use with renewable energy is not a silver bullet solution. If we improve on energy efficiency too, so we don't need to so much energy in the first place, then the impacts of low carbon energy are more manageable.

Corrections
24/April: the figures for PM10 concentrations were not consistent with the chart, they read 145 and 27, corresponding to full capacity instead of 87 and 16, for 2/3 capacity.

[1] Emissions (Biomass energy centre)
[2] Sulphur content of liquid fuels (Air pollution information system)
[3] Classification of Coal (Engineering Toolbox)
[4] MdDonald (2009) Evaluation of Gas, Oil and Wood Pellet Fueled Residential Heating System Emissions Characteristics (Brookhaven National Laboratory)
[5] Solid Fuel and Air quality: an update for local authorities (Environmental Protection UK) April 2013
[6] RHI Emissions Certificates (HETAS) April 2014
[7] 30g/GJ is 0.11g/kWh. Mean January space heating requirement is 68kWh/day calculated from the Cambridge Housing Model
Assuming 85% efficiency, 68/0.85*0.11 =8.6
[8] Appliance Exemption from section 21 of the Clean Air Act 1993 (DEFRA)
6.7g/hour for 5 kW stoves is 6.7/5 =1.3g/kWh, with is 370g/GJ
[9] Introduction of air quality requirements into the Renewable Heat Incentive (RHI) (DEFRA)
From Table 2, 0.1 g/GJ is the difference between two modelled scenarios with and without the RH 30 g/GJ limit. This limit approximately halves the emissions (table 1), so the difference between having the limit and not is the same as the difference between having the limit and no boilers at all.
[10] Based on 11,000 kWh/year space heating need per home, at 85% efficiency, generating emissions at the limit rate 30g/GJ
[11] The Great Smog of 1952 (MET)
[12] Cambridge is about 40km2 (Wikipedia) and 48,000 dwellings (https://www.cambridge.gov.uk/sites/www.cambridge.gov.uk/files/documents/Population,%20Housing%20and%20Employment%20Forecasts%202013.pdf)