HYDROLOGICAL PROCESSES Hydrol. Process. 22, 919–940 (2008) Published online 28 February 2008 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/hyp.7003 A comparison of forest and moorland stream microclimate, heat exchanges and thermal dynamics David M. Hannah, 1 * Iain A. Malcolm, 2 Chris Soulsby 3 and Alan F. Youngson 2 1 School of Geography, Earth and Environmental Sciences, University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK 2 Fisheries Research Services (FRS) Freshwater Laboratory, Faskally, Pitlochry, Perthshire, PH16 5LB, UK 3 School of Geosciences, University of Aberdeen, Elphinstone Road, Aberdeen, AB24 3UF, UK Abstract: Although the importance of riparian forest in moderating stream temperature variability is recognized, most previous research focuses on conifer harvesting effects and summer maximum temperature with highly variable findings. This article compares stream temperature, microclimate and heat exchange dynamics between semi-natural forest and moorland (no trees) reaches in the Scottish Cairngorms over two calendar years to provide a longer-term perspective. Mean daily water column temperature is warmer for moorland than forest in late winter–early spring, but cooler in summer. Daily water column temperature range is greater for moorland than forest. Streambed temperature dynamics are markedly different between reaches, reflecting contrasting groundwater–surface water (GW–SW) interactions. Mean, minimum and maximum daily air temperature is cooler, humidity is lower, and wind speed is much higher for moorland than forest on average. Net radiation is the dominant heat sink in autumn–winter and major heat source in spring–summer for moorland and summer for forest. Net radiation is greater in summer and lower in winter for moorland than forest. Sensible heat is an energy source in autumn–winter and sink in spring–summer, with loss (gain) greater in summer (winter) for moorland than forest. Latent heat is predominantly a sink for both reaches, with magnitude and variability higher for moorland than forest. Streambed heat flux is much smaller than fluxes at the air–water interface, with moorland and forest illustrating seasonal and between-reach differences attributable to different GW–SW interactions. Seasonal patterns in stream energy budget partitioning are illustrated schematically. To our knowledge, this is the first such study of mixed woodland, which generates notably different results to work on coniferous forest. This research provides a process basis to model stream thermal impact of changes in forest practice, and so inform decision making by land and water resource managers. Copyright  2008 John Wiley & Sons, Ltd. KEY WORDS water temperature; riverbed; thermal regime; energy balance; heat budget; hydrometeorology; riparian woodland; Cairngorms; Scotland Received 24 September 2007; Accepted 9 January 2008 INTRODUCTION Stream temperature is an important and highly sen- sitive variable affecting physical, chemical and bio- logical processes (Poole and Berman, 2001; Malcolm et al., 2008; Caissie, 2006). It is controlled by dynamic energy (heat) and hydrological fluxes at the air–water and water–riverbed interfaces (Figure 1; Hannah et al., 2004). Land and water management impact on these drivers and, thus, modify river thermal characteristics (Webb et al., 2008). Recent work has attempted to dis- entangle the multivariate influence of the wide range of factors that control river temperature (Isaak and Hubert, 2001; Gu and Li, 2002). Numerous, mainly North Amer- ican, studies have highlighted the importance of ripar- ian forest in moderating stream thermal dynamics (e.g. Chen et al., 1998; Johnson and Jones, 2000; Macdonald et al., 2003; Malcolm et al., 2004a; Danehy et al., 2005; Moore et al., 2005b; Gomi et al., 2006). This research has focused on timber harvesting effects and summer maximum temperature (Johnson, 2003); and findings * Correspondence to: David M. Hannah, School of Geography, Earth and Environmental Sciences, University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK. E-mail: d.m.hannah@bham.ac.uk have been highly variable in terms of amount of warm- ing and rates of recovery to pre-harvest conditions (as reviewed by Moore et al., 2005a). Thus, there is a need to consider all aspects of the thermal regime and adopt a longer-term perspective, rather than focusing on summer extremes, and to systematically compare spatio-temporal heat budget dynamics between forest and open river reaches. To protect streams against direct insolation, wooded riparian buffer strips are now deemed best manage- ment practice in North America (Zwieniecki and Newton, 1999; Moore et al., 2005a). The current UK Forest and Water Guidelines (Forestry Commission, 2003) recom- mend 50% of each river reach be shaded by riparian trees, if salmonids predominate. In Scotland, environ- mental organizations are promoting (re)introduction of native/semi-natural woodlands to improve quality of river habitat for fish (Urquhart, 2004). However, there have been few medium to long-term investigations of forestry effects on stream temperatures in the United Kingdom (Webb and Crisp, 2006), and there is a global paucity of such studies of semi-natural woodland. Thus, the scien- tific basis for these management decisions is limited with Copyright  2008 John Wiley & Sons, Ltd.