5 August 2010, by Tom Marshall

Research is needed on how trees will respond to the changing climate, a new paper suggests.

Temperate forest

Scientists know a fair amount about how individual factors like higher temperatures or more carbon dioxide (CO2) in the air may affect plants. But there are severe limits on applying this knowledge directly to the real world.

Current knowledge mostly comes from experiments on individual plants, rather than long-term work on whole ecosystems, and we don't know much about how the different factors will interact with each other.

'So far we have mostly been looking at the impact of single factors such as temperature or CO2 on tree growth; this was largely because it is easier to manipulate one factor at a time,' explains Dr Martin Lukac, a researcher at Reading University's Department of Agriculture and lead author of the paper, published in Tree Physiology.

'But this isn't what happens in the real world - one variable doesn't just change on its own while everything else stays the same,' he adds, 'What we really need is long-term experiments on how these factors will interact and affect each other across entire ecosystems.'

On top of this, experiments so far have usually been carried out using seedlings or young trees. These are easier to manipulate under lab conditions, but may not reflect how mature trees will respond to a changing environment. More work on how whole ecosystems will react is urgently needed.

In the real world, one variable doesn't just change on its own while everything else stays the same'

Dr Martin Lukac, Reading University

For example, in the cold boreal forests that cover much of Russia and Canada, growth is at present limited by temperature.

Initially, at least, warmer conditions and air richer in CO2 are likely to mean faster growth; this in turn will increase production of biomass and the amount of CO2 being absorbed.

But this may not last; by growing faster, trees will use the stocks of nutrients in the soil more quickly, and may eventually deplete them. This would lead to a negative feedback loop that would cause growth and the absorption of CO2 to slow down again. Due to the large contribution of forests to the global carbon cycle, such shifts in CO2 uptake of may themselves determine how quickly climate change takes effect.

There are similarly complex feedbacks and interactions affecting other variables such as temperature, and its effects on the activity of soil fungi and microbes, with direct implications for the pool of nutrients held in the soil. Hotter summers may also make drought more common, and this will have ramifications both for the availability of nutrients, and for how well trees can use them. For example, very dry conditions may reduce the availability of nutrients by limiting the rate of their release and hinder nutrient acquisition by affecting root physiology.

A further problem is that researchers have tended to focus on forests in temperate climates. Other woodland types, like rainforests or boreal forests, have so far been neglected with regard to large-scale climate change experiments. Such research is expensive and can be logistically difficult, while most research money tends to be available in countries with temperate climates; this has meant resources have been concentrated in work on temperate forests.

Lukac says the best way to understand this intricate interplay between the numerous consequences of climate change is to with large-scale experiments. Only this will show what the overall effect of different climate-change scenarios will be on forests in different parts of the world.

Some research of this kind is going on, but more is needed according to Lukac. Many experiments being done at present are observational, making use of natural temperature or precipitation gradients by taking measurements at regular intervals along a line stretching from lowland rainforest to mountain cloud forest, for example. But Lukac believes we also need manipulation experiments, in which environmental variables are manipulated to uncover fundamental mechanisms or relationships.

This can be complex. For example, CO2 levels can be manipulated by surrounding trees in a network of pipes that emit the gas in controlled quantities. Or drought can be simulated by roofing off bits of woodland to prevent any rain reaching the soil and then watering artificially to supply the desired amount of moisture.

Temperature is harder to manipulate at ecosystem scale; the soil temperature can readily be changed by burying heating elements, but the temperature of the air mass surrounding the trees is much more difficult to change reliably without enclosing whole sections of forest entirely - a very expensive proposition. Climate change will affect air and not just soil temperature, though, so scientists will need to manipulate this variable somehow to understand its effect on forests.

Forests are long-lived ecosystems, and the experiments will also need to take place over long periods, Lukac says. Otherwise, they may miss important variations over time in forest processes. 'For example, certain types of soil microorganisms are associated only with specific stages of forest development, so short-term experiments may not provide a complete picture of how nutrient availability in forest soils will respond to environmental change.'