Sir

William Laurance, in his News & Views article 'Forests and floods' (Nature 449, 409–410; 2007), highlights a paper by C. Bradshaw and colleagues, claiming that it provides correlative evidence that native forests reduce the frequency and severity of floods in developing countries.

The 'forest and floods' debate goes back at least to the nineteenth century. Now forest hydrologists generally agree that, although forests mitigate floods at the local scale and for small to medium-sized flood events, there is no evidence of significant benefit at larger scales and for larger events.

Laurance also recognizes the omission of extreme events in the Bradshaw analysis. But we argue that this seriously weakens the policy importance of the results. It is these extreme events that matter: economic damage and loss of life grow exponentially with flood magnitude. The authors also excluded Chinese data because of outliers — an unfortunate omission, given that China has undergone large changes in forest cover where the flood 'signal' should be strong.

Laurance did not discuss Bradshaw and colleagues' other conclusion, namely that increasing the number of forest plantations can lead to longer and more frequent floods.For China, where initiatives such as the sloping-lands conversion programme — heavily promoted on the basis of flood-mitigation benefits — are leading to forest plantation over areas comparable to those of afforestation by the rest of the world put together, the conclusion is particularly important. Other studies have warned that forest-management activities can aggravate flood risk (see J. A. Jones and G. E. Grant Water Resources Res. 32, 959–974; 1996).