TY - JOUR
AU - Connolly, Thomas
AB - Abstract There is a strong policy move in Britain to improve forest resilience to climate change by increasing stand structural and species diversity. Although currently little used in Britain, the technique of underplanting allows regeneration and diversification of stands while avoiding some of the disadvantages of clearfelling. Two experiments were examined: (1) the growth and survival of five underplanted conifer species of differing shade tolerance in a shelterwood and (2) compared performance of underplanted and open-grown Douglas-fir seedlings on restocking sites. Underplanted Sitka spruce, Norway spruce, noble fir, European silver fir and Douglas-fir were all able to survive and grow. However increased exposure following overstorey removal resulted in some damage and ‘socketing’, especially to taller seedlings, particularly Douglas-fir. This may be linked to poor root development when growing under an overstorey. Microclimate conditions on some underplanted sites were more sheltered from extreme climatic conditions, and in some cases this improved survival of Douglas-fir seedlings. However, seedling growth rates were reduced compared with those on open sites probably due to lower light levels. Underplanting may help to improve establishment success of some species, particularly in exposed areas. However, the shelter benefits of underplanting must be carefully balanced against the trade-off with lower light, and underplanting is likely to be more successful where low canopy density is maintained. Introduction Evidence suggests that forests with a diverse structural and species composition are more resilient to the effects of climate change and pests and disease (e.g. Su et al., 1996; Gerlach et al., 1997; Bengtsson et al., 2000; Hanewinkel et al., 2014; Lafond et al., 2014; Jactel et al., 2017). The potential benefits of increasing the structural and species diversity of planted conifer forests have been reported (Meason and Mason, 2014; Cameron, 2015), and there is strong policy support for silvicultural methods to improve the resilience of forests, including diversifying the range of species being planted (e.g. Grant et al., 2012; Natural Resources Wales, 2017). Despite this, the dominant silvicultural system used in British forestry remains patch clearfelling followed by planting, usually with a single species, predominantly Sitka spruce (Picea sitchensis (Bong.) Carr.) (Mason, 2015); managers wishing to change species usually clearfell the stand at rotation age and replant with an appropriate species or species mixture. Restocking of clearfelled sites can be a costly and high-risk phase of forest management, particularly when using less well-known alternative species. The negative impacts of drought, heat exposure and frost on the growth and survival of seedlings on restock sites have been reported by many (Grossnickle, 2000; Langvall and Löfvenius, 2002; Paquette et al., 2006). Under climate change conditions, restocking is likely to become increasingly challenging; summer droughts are likely to increase in severity, frequency and length; winter storms and heavy rainfall events may increase physical damage and waterlogging (Lowe et al., 2018), and there may be new pests and diseases (van Lierop et al., 2015). Models predict a reduction in the number of frost days and in the diurnal temperature range in winter (Lowe et al., 2018), which may reduce frost damage in spring, depending on the timing (Cannell and Smith, 1986; Murray et al., 1994). However, although autumn frosts are expected to occur less frequently, they may cause more damage due to generally warmer temperatures and later hardening of shoots. These changes are all likely to increase stress to young planted trees on exposed restock sites and reduce seedling survival and growth (Desprez-Loustau et al., 2006). Continuous cover forestry (CCF) is an approach to forest management which involves establishing the next generation of trees before removing the overstorey. CCF techniques have become more widely practised in Britain over the past two decades (Mason et al., 1999; Mason, 2015) and may address some of the challenges of planting on open restock sites. The presence of an overstorey can potentially reduce impacts of drought, high temperatures, strong winds and heavy precipitation (rainfall or snow) on seedlings (Childs and Flint, 1987; Calder et al., 2003; Erefur et al., 2008; Jiménez et al., 2008). Sellars (2006) reported that increased canopy openness resulted in higher wind speeds and larger fluctuations in air temperature, particularly close to ground level. Temperature near ground level can be several degrees higher under a shelterwood than in open exposed sites on clear calm nights, reducing frost damage to young trees (Groot and Carlson, 1996; Langvall and Örlander, 2001; Nilsson et al., 2006; Erefur et al., 2008). Hungerford and Babbitt (1987) also demonstrated the impact of the overstorey sheltering effect on survival of Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) seedlings, identifying the importance of the frequency and severity of extreme high temperature and low temperature events, rather than the monthly average temperature. The presence of an overstorey may also result in microclimate conditions which are less favourable to pests and diseases (Mason et al. 2004a). The trade-off is that light levels beneath the canopy are lower, reducing photosynthetic rates and growth of tree seedlings. Managers using CCF techniques in Britain generally aim to establish the next generation of trees by using natural regeneration. However, this restricts managers to the existing overstorey species and provenance, which may not be suitable for the current or future site conditions. Natural regeneration is not always reliable, or of good genetic quality, and can be affected by the age and health of the overstorey trees. The technique of underplanting (planting young tree seedlings underneath the canopy of an existing stand) allows managers to adjust species or provenance choice, increase diversity and introduce mixtures by establishing the trees in a less extreme microenvironment than is found on a restock site (Paquette et al., 2006). There is limited recent experience or knowledge of underplanting in British forestry, although it was used prior to World War II, and in the 1950s for restoring derelict woodland. The technique is more widely used in continental European and North American forestry (e.g. Smidt and Puettmann, 1998; Holgén and Hånell, 2000; Brandeis et al., 2001; Maas-Herbner et al., 2005). The requirements of different species in underplanting conditions in Britain are generally not well understood, particularly for some of the ‘emerging’ species being considered for future climatic conditions or for relatively shade-tolerant species (such as western hemlock (Tsuga heterophylla (Raf.) Sarg), Douglas-fir, grand fir (Abies grandis (Douglas ex D. Don) Lindley), European silver fir (Abies alba (Mill.)), noble fir (Abies procera (Rehder)), western redcedar (Thuja plicata (Donn ex D. Don)) that could be used to increase species diversity in existing stands (Cameron, 2015). As the light requirements and tolerance of different species to frost and drought vary, some may be better able to grow and survive under an overstorey than others (Hale, 2004; Mason et al. 2004b; Vanoni et al., 2016). This paper summarizes results and experience gained from two experiments on underplanting in Britain. The first examined growth and survival of a range of species planted under a uniform shelterwood in North Wales. The second compared growth and survival of underplanted and open-grown Douglas-fir seedlings on three sites and examined the impact of microclimate conditions. The objectives were to: compare survival and growth of five shade-tolerant conifer species growing under a Sitka spruce overstorey; determine the impact of thinning operations and shelterwood removal on the survival and development of underplanted species; and compare survival and growth of underplanted seedlings with open-grown seedlings. Methods Experiment 1: Clocaenog underplanting species trial Experiment 1 was established in a stand of Sitka spruce planted in 1948 at Clocaenog Forest, North Wales (Figure 1); site details are summarized in Table 1. The stand lies within an operational research forest where implementation of a range of transformation to CCF techniques began in 2001. The overstorey was even-aged and had been thinned regularly until 1999, after which it was managed as a uniform shelterwood; a summary of stand operations between 2001 and shelterwood removal is shown in Table 2. The underplanting experiment was planted in March 2007; permanently marked extraction racks and timber processing and stacking areas were used in all operations, and felling directions were marked on trees to minimize damage to the underplanted seedlings. Figure 1 Open in new tabDownload slide Location of experiment sites at Clocaenog (Experiment 1) and at Great Glen, Wythop and Clocaenog for Experiment 2. Figure 1 Open in new tabDownload slide Location of experiment sites at Clocaenog (Experiment 1) and at Great Glen, Wythop and Clocaenog for Experiment 2. Table 1 Summary of site details at Experiment 1 (Clocaenog) and Experiment 2 sites (Wythop, Clocaenog and Great Glen). . Experiment 1 . Experiment 2 . . Clocaenog . Wythop# . Clocaenog . Great Glen . Latitude, longitude 53.08 N, 3.43 W 54.65 N, 3.22 W 53.08 N, 3.43 W 55.22 N, 3.62 W Elevation (m) 400 120–260 315 90 Aspect NW NE—SE SE SE Exposure (DAMS)† 19 8.9–12.8 12.7 10.7 Rainfall (mm)† 1100 1585 1020 1335 SMR† Moist Fresh Wet Very moist SNR† Poor Poor Medium Very poor Soil type Iron-pan Predominantly brown earths Typical brown podzol Podzolic brown earth Underplanted Spring 2007 Spring 2011 Spring 2012 Spring 2012 . Experiment 1 . Experiment 2 . . Clocaenog . Wythop# . Clocaenog . Great Glen . Latitude, longitude 53.08 N, 3.43 W 54.65 N, 3.22 W 53.08 N, 3.43 W 55.22 N, 3.62 W Elevation (m) 400 120–260 315 90 Aspect NW NE—SE SE SE Exposure (DAMS)† 19 8.9–12.8 12.7 10.7 Rainfall (mm)† 1100 1585 1020 1335 SMR† Moist Fresh Wet Very moist SNR† Poor Poor Medium Very poor Soil type Iron-pan Predominantly brown earths Typical brown podzol Podzolic brown earth Underplanted Spring 2007 Spring 2011 Spring 2012 Spring 2012 #Range given covering the four sites. †According to Ecological Site Classification (Pyatt et al., 2001) Open in new tab Table 1 Summary of site details at Experiment 1 (Clocaenog) and Experiment 2 sites (Wythop, Clocaenog and Great Glen). . Experiment 1 . Experiment 2 . . Clocaenog . Wythop# . Clocaenog . Great Glen . Latitude, longitude 53.08 N, 3.43 W 54.65 N, 3.22 W 53.08 N, 3.43 W 55.22 N, 3.62 W Elevation (m) 400 120–260 315 90 Aspect NW NE—SE SE SE Exposure (DAMS)† 19 8.9–12.8 12.7 10.7 Rainfall (mm)† 1100 1585 1020 1335 SMR† Moist Fresh Wet Very moist SNR† Poor Poor Medium Very poor Soil type Iron-pan Predominantly brown earths Typical brown podzol Podzolic brown earth Underplanted Spring 2007 Spring 2011 Spring 2012 Spring 2012 . Experiment 1 . Experiment 2 . . Clocaenog . Wythop# . Clocaenog . Great Glen . Latitude, longitude 53.08 N, 3.43 W 54.65 N, 3.22 W 53.08 N, 3.43 W 55.22 N, 3.62 W Elevation (m) 400 120–260 315 90 Aspect NW NE—SE SE SE Exposure (DAMS)† 19 8.9–12.8 12.7 10.7 Rainfall (mm)† 1100 1585 1020 1335 SMR† Moist Fresh Wet Very moist SNR† Poor Poor Medium Very poor Soil type Iron-pan Predominantly brown earths Typical brown podzol Podzolic brown earth Underplanted Spring 2007 Spring 2011 Spring 2012 Spring 2012 #Range given covering the four sites. †According to Ecological Site Classification (Pyatt et al., 2001) Open in new tab Table 2 Summary of operations carried out in Experiment 1 at Clocaenog. Date . Operation . June 2004 Overstorey age 54; frame trees selected, crown thinned to residual basal area of 25 m2 ha−1. October 2006 Dense natural regeneration of Sitka spruce cleared using motor manual and harvester mounted flails. March 2007 Experiment planted into unfenced plots. June 2009 Overstorey crown thinned (3rd growing season following underplanting) from 31 down to 25 m2 ha−1 June 2012 Overstorey crown thinned (6th growing season following underplanting) from 30 down to 25 m2 ha−1 September 2013 Vigorous natural regeneration of Sitka spruce controlled with a directed application of 1.8 kg a.i. ha−1 glyphosate (as 5 l ha−1 Roundup Pro Biactive (360 g l−1 glyphosate) Monsanto). October 2015 End of 9th growing season; basal area was 30 m2 ha−1; overstorey clear felled by a combination of motor-manual and harvester felling, with forwarder extraction. Date . Operation . June 2004 Overstorey age 54; frame trees selected, crown thinned to residual basal area of 25 m2 ha−1. October 2006 Dense natural regeneration of Sitka spruce cleared using motor manual and harvester mounted flails. March 2007 Experiment planted into unfenced plots. June 2009 Overstorey crown thinned (3rd growing season following underplanting) from 31 down to 25 m2 ha−1 June 2012 Overstorey crown thinned (6th growing season following underplanting) from 30 down to 25 m2 ha−1 September 2013 Vigorous natural regeneration of Sitka spruce controlled with a directed application of 1.8 kg a.i. ha−1 glyphosate (as 5 l ha−1 Roundup Pro Biactive (360 g l−1 glyphosate) Monsanto). October 2015 End of 9th growing season; basal area was 30 m2 ha−1; overstorey clear felled by a combination of motor-manual and harvester felling, with forwarder extraction. Open in new tab Table 2 Summary of operations carried out in Experiment 1 at Clocaenog. Date . Operation . June 2004 Overstorey age 54; frame trees selected, crown thinned to residual basal area of 25 m2 ha−1. October 2006 Dense natural regeneration of Sitka spruce cleared using motor manual and harvester mounted flails. March 2007 Experiment planted into unfenced plots. June 2009 Overstorey crown thinned (3rd growing season following underplanting) from 31 down to 25 m2 ha−1 June 2012 Overstorey crown thinned (6th growing season following underplanting) from 30 down to 25 m2 ha−1 September 2013 Vigorous natural regeneration of Sitka spruce controlled with a directed application of 1.8 kg a.i. ha−1 glyphosate (as 5 l ha−1 Roundup Pro Biactive (360 g l−1 glyphosate) Monsanto). October 2015 End of 9th growing season; basal area was 30 m2 ha−1; overstorey clear felled by a combination of motor-manual and harvester felling, with forwarder extraction. Date . Operation . June 2004 Overstorey age 54; frame trees selected, crown thinned to residual basal area of 25 m2 ha−1. October 2006 Dense natural regeneration of Sitka spruce cleared using motor manual and harvester mounted flails. March 2007 Experiment planted into unfenced plots. June 2009 Overstorey crown thinned (3rd growing season following underplanting) from 31 down to 25 m2 ha−1 June 2012 Overstorey crown thinned (6th growing season following underplanting) from 30 down to 25 m2 ha−1 September 2013 Vigorous natural regeneration of Sitka spruce controlled with a directed application of 1.8 kg a.i. ha−1 glyphosate (as 5 l ha−1 Roundup Pro Biactive (360 g l−1 glyphosate) Monsanto). October 2015 End of 9th growing season; basal area was 30 m2 ha−1; overstorey clear felled by a combination of motor-manual and harvester felling, with forwarder extraction. Open in new tab The experiment had three replicate blocks with each block containing 50 × 50 m randomized single-species plots of each of the five species, planted at 2 × 2 m spacing with 2-year-old plants. The species planted were noble fir, Douglas-fir, Sitka spruce, Norway spruce (Picea abies L. (Karst.)) and European silver fir, listed in order of increasing shade tolerance (according to Niinements and Valladares, 2006). However, due to limited space on the site, all three European silver fir plots, one Norway spruce plot, one noble fir plot and one Sitka spruce plot, were reduced in size to 50 × 25 m (Figure 2). Supply problems with European silver fir meant that two plots of 50 trees and one plot of only 10 trees were planted, and planting of this species was delayed until April 2007. Figure 2 Open in new tabDownload slide General layout of species plots at Clocaenog (Experiment 1). Species codes: DF, Douglas-fir; NF, noble fir; NS, Norway spruce; SF, European silver fir; SS, Sitka spruce. Number indicates the replicate block. Large plots were 50 × 50 m, and smaller plots were 50 × 25 m. Assessments were carried out on 50 permanently marked trees in the centre of each plot. Only 50 trees each were planted in plots SF1 and SF2, and only 10 trees in plot SF3. Figure 2 Open in new tabDownload slide General layout of species plots at Clocaenog (Experiment 1). Species codes: DF, Douglas-fir; NF, noble fir; NS, Norway spruce; SF, European silver fir; SS, Sitka spruce. Number indicates the replicate block. Large plots were 50 × 50 m, and smaller plots were 50 × 25 m. Assessments were carried out on 50 permanently marked trees in the centre of each plot. Only 50 trees each were planted in plots SF1 and SF2, and only 10 trees in plot SF3. Assessments were carried out on 50 permanently marked and numbered trees in the centre of each plot; for the smaller SF plots, all 50 trees were assessed in two plots and 10 trees in the third plot where insufficient planting stock had been available. At the end of growing seasons 2, 4, 6, 8, 9 and 10, the survival, height (cm), stem diameter at 5 cm above ground level (mm) and length of the leader and longest lateral branch in the uppermost whorl (cm) were recorded, allowing calculation of leader–lateral ratio. Health score was also recorded for each tree using a 1–3 scale, where 1 = generally healthy-looking or with minor damage or defects, 2 = damaged or unhealthy but likely to recover and 3 = damaged or unhealthy and unlikely to recover. At the end of growing seasons 9 and 10, a lean score was also recorded for each tree using a stick and protractor, where 1 = tree is vertical ±5°, 2 = tree is leaning at an angle of <45° and 3 = tree is leaning at an angle of >45°. Unfortunately, no measurements were made on the assessment trees at planting or at the end of year 1, and it has not been possible to correct for any systematic differences in tree size between species at planting. Analysis Data were prepared and analysed using statistical software (Genstat, 2013). Analysis used a REML mixed model on individual tree data for height, diameter and leader–lateral ratio, with fixed effect for species and random effects for variation between blocks, between plots within blocks and between trees within plots. Response variables were assumed to be normally distributed. The response variables of survival, health score and lean score were analysed using a generalized linear mixed model (GLMM) based on plot counts, with fixed effect for species and random effect for variation between blocks. Response variables were assumed to have a binomial distribution with logit link, i.e. logit(p) = log(p/(1−p)) where p is the proportion of trees attaining an outcome (i.e. for the variable of Health Score 1,2 (HS12) p is the proportion of trees scoring either 1 or 2). A dispersion factor was also fitted to account for observed extra binomial variation; one consequence of this is to increase the variation around predictions for fixed effects, making predictions less precise. Post hoc comparison of species performance was carried out using the Bonferroni method of multiple comparisons where the species effect was significant (P < 0.05). The Bonferroni method leads to conservative intervals for differences between means, i.e. it is sometimes the case that predicted means for a significant fixed effect are found not to differ significantly using the Bonferroni method and can all share the same ‘significance letter’. Experiment 2: Performance of underplanted vs open-grown Douglas-fir seedlings In the second experiment, three trials were established to compare the performance of open-grown and underplanted Douglas-fir seedlings. Plots of 10 × 10 Douglas-fir seedlings were planted at 2 × 2 m spacing in open restock areas and at 1 × 1 m spacing in underplanting areas (following guidance to plant at closer spacing when underplanting, to facilitate rapid establishment, Kerr and Haufe, 2016). There was one main experiment site at Wythop, which had five replicate plots of each treatment (‘open-grown’ and ‘underplanted’), and two smaller demonstration sites at Clocaenog and Great Glen (Figure 1). Plots at all sites were unfenced. Site details are summarized in Table 1. The Wythop site had 5 underplanted plots of 100 trees each (Figure 3). These were located in two Douglas-fir stands that were planted in 1927 and had been undergoing transformation to irregular shelterwoods since 2001. Three of the plots were located in a stand that was thinned in 1996 and 2009 and had a basal area of 31 m2 ha−1 in 2011 (Figure 3; ‘Underplanted 1’). The other two plots were located in a nearby stand that was thinned in 1993 and 2006, with a basal area of 25 m2 ha−1 (Figure 3; ‘Underplanted 2’). In addition, there were 5 open-grown plots of 100 trees each. These were located in two nearby compartments that had previously been Sitka spruce, clearfelled in 2006 (three plots in one compartment ‘Open grown 1’, and two in the other ‘Open grown 2’, Figure 3). The restocking sites were excavator mounded prior to planting; no ground preparation was carried out in the underplanting areas. Figure 3 Open in new tabDownload slide Map and plot location for Experiment 2 Wythop Wood, showing locations of the underplanted stands and open-grown areas, and positions of the five plots within them (plots not shown to scale). Location of each weather station is indicated by a star. Figure 3 Open in new tabDownload slide Map and plot location for Experiment 2 Wythop Wood, showing locations of the underplanted stands and open-grown areas, and positions of the five plots within them (plots not shown to scale). Location of each weather station is indicated by a star. At Great Glen, 2 underplanted plots of 100 trees each were located in an even-aged Douglas-fir seed stand, planted in 1927. The stand was thinned in 2011 and had a relatively low basal area of ~20 m2 ha−1 to promote seed production. Two open-grown plots of 100 trees each were located in the adjacent compartment; the previous conifer crop had been clearfelled some years earlier, and there was a large amount of harvesting residue. No ground preparation was carried out in either the underplanted area or the open-grown area. At Clocaenog, four underplanted plots of 100 trees each were located in an even-aged stand of Sitka spruce planted in 1946. The stand was in the early stages of transformation to a uniform shelterwood and was last thinned in 2010. The basal area was ~38 m2 ha−1 at the start of the experiment. Two open-grown plots of 100 trees each were located in the adjacent compartment, from which the previous crop of Sitka spruce had been clearfelled, leaving a large amount of harvesting residue. The two open-grown plots and two of the four underplanted plots were excavator mounded prior to planting; the remaining two underplanted plots received no ground preparation. Environmental monitoring equipment was installed at each site to record microclimate conditions during the first 5 years after planting; details are shown in Table 3. The periods of coldest and warmest air temperatures, and highest wind speeds during the first growing season at each site were identified as representing periods of particularly high risk for the recently planted seedlings. The microclimate measurements made in the underplanted and open-grown locations at each site during these periods were compared. Monthly averages for the meteorological measurements at each site (and monthly totals for rainfall) were also calculated for the first year following planting. Table 3 Summary of meteorological monitoring equipment at Experiment 2 sites. Site . Monitoring locations . Equipment . Record frequency . Wythop Two weather stations in underplanted and two in open-grown locations, see Figure 3 Weather stations (model WS-GP1, Delta-T Devices Ltd, Burwell, Cambridge, UK). Comprising Wind speed and direction sensor (model D-043B-CA) 1 min, hourly average Relative humidity and air temperature sensor (model RHT3nl-CA) 5 min, hourly average Solar energy flux sensor (model PYRPA-03) 1 min, hourly average Rain gauge (model RG2 + WS-CA) Hourly total Clocaenog Two weather stations; one in underplanted and one in open-grown location As described above As above Great Glen Two dataloggers; one in underplanted and one in open-grown location Tinytag temperature and humidity sensors (model TGP-4500, Gemini Dataloggers, Chichester, UK) Half-hourly average Site . Monitoring locations . Equipment . Record frequency . Wythop Two weather stations in underplanted and two in open-grown locations, see Figure 3 Weather stations (model WS-GP1, Delta-T Devices Ltd, Burwell, Cambridge, UK). Comprising Wind speed and direction sensor (model D-043B-CA) 1 min, hourly average Relative humidity and air temperature sensor (model RHT3nl-CA) 5 min, hourly average Solar energy flux sensor (model PYRPA-03) 1 min, hourly average Rain gauge (model RG2 + WS-CA) Hourly total Clocaenog Two weather stations; one in underplanted and one in open-grown location As described above As above Great Glen Two dataloggers; one in underplanted and one in open-grown location Tinytag temperature and humidity sensors (model TGP-4500, Gemini Dataloggers, Chichester, UK) Half-hourly average Open in new tab Table 3 Summary of meteorological monitoring equipment at Experiment 2 sites. Site . Monitoring locations . Equipment . Record frequency . Wythop Two weather stations in underplanted and two in open-grown locations, see Figure 3 Weather stations (model WS-GP1, Delta-T Devices Ltd, Burwell, Cambridge, UK). Comprising Wind speed and direction sensor (model D-043B-CA) 1 min, hourly average Relative humidity and air temperature sensor (model RHT3nl-CA) 5 min, hourly average Solar energy flux sensor (model PYRPA-03) 1 min, hourly average Rain gauge (model RG2 + WS-CA) Hourly total Clocaenog Two weather stations; one in underplanted and one in open-grown location As described above As above Great Glen Two dataloggers; one in underplanted and one in open-grown location Tinytag temperature and humidity sensors (model TGP-4500, Gemini Dataloggers, Chichester, UK) Half-hourly average Site . Monitoring locations . Equipment . Record frequency . Wythop Two weather stations in underplanted and two in open-grown locations, see Figure 3 Weather stations (model WS-GP1, Delta-T Devices Ltd, Burwell, Cambridge, UK). Comprising Wind speed and direction sensor (model D-043B-CA) 1 min, hourly average Relative humidity and air temperature sensor (model RHT3nl-CA) 5 min, hourly average Solar energy flux sensor (model PYRPA-03) 1 min, hourly average Rain gauge (model RG2 + WS-CA) Hourly total Clocaenog Two weather stations; one in underplanted and one in open-grown location As described above As above Great Glen Two dataloggers; one in underplanted and one in open-grown location Tinytag temperature and humidity sensors (model TGP-4500, Gemini Dataloggers, Chichester, UK) Half-hourly average Open in new tab Height (cm), stem diameter at 5 cm above ground level (mm) and survival of the 100 trees in each plot were recorded at the end of each of the first three growing seasons at Wythop (after which unintended harvesting operations caused severe damage to the underplanted plots, which were abandoned) and each of the first five growing seasons at the other two sites. Analysis Data for Clocaenog and Great Glen were not statistically analysed as the sites were established as demonstrations with limited replication and are included here for interest and comparison only. Data for the Wythop experiment site were prepared and analysed using the Genstat (2013) statistical software. Analysis used a mixed model for height and diameter and a generalized linear mixed model (GLMM) for survival. The two growth variables were assumed to be normally distributed, whereas survival was modelled assuming a binomial distribution with logit link, i.e. logit(p) = log(p/(1−p)) where p is the proportion of plants alive. In all the models the fixed effect was forest cover (underplanted vs open-grown) and variation between plots was fitted as a random effect. Results Experiment 1: Clocaenog underplanting species trial Performance of shade-tolerant conifer species Survival Generally survival of all species was high, >80 per cent to year 9, and was not affected by thinning (Figure 4). Overstorey removal caused minor reductions in percentage survival except for Douglas-fir which had a sharp drop in survival between year 9 and 10, from 93 to 76 per cent. There were no significant differences in survival between species in any year, except year 10 (borderline significance P = 0.049, Figure 4 and Supplementary Table S1). Post hoc Bonferroni tests could not distinguish which species had significantly different survival from the others in year 10. The small increase in survival of Norway spruce between years 2 and 4 (from 93 to 95 per cent) was due to measurement error. Figure 4 Open in new tabDownload slide (a–e) Summary of survival (proportion), stem diameter (mm), height (cm), leader: lateral ratio and proportion of seedlings achieving health score 1 at the Clocaenog underplanting species trial (Experiment 1). Figure 4 Open in new tabDownload slide (a–e) Summary of survival (proportion), stem diameter (mm), height (cm), leader: lateral ratio and proportion of seedlings achieving health score 1 at the Clocaenog underplanting species trial (Experiment 1). Stem diameter All species had reduced rates of diameter growth during the 9th growing season (during which the overstorey was removed) followed by a high rate of diameter growth during the 10th growing season (Figure 4). The overstorey thinning operations during the third and sixth growing seasons did not affect stem diameter growth rate. Significant differences in diameter between species were seen in year 2 only (P = 0.0039; Figure 4 and Supplementary Table S1); European silver fir (SF) and Sitka spruce (SS) had a significantly larger mean stem diameter than noble fir (NF) and Norway spruce (NS); diameter of Douglas-fir (DF) was intermediate. Height The main finding was that height growth rate appeared reasonably constant for all species throughout the period of assessment and there was no clear impact of the overstorey thinning during the third and sixth growing seasons or the overstorey removal during the ninth growing season (Figure 4). Growth rates were generally good, with species increasing between 2.5 and 4.2 times in height in the first 10 years. In all years except 9 and 10, there were significant differences in height between species (Supplementary Table S1), although in year 8 post hoc Bonferroni tests could not identify which species had significantly different heights from each other. At the end of year 2, SS and DF were significantly taller than NF and NS (SF was intermediate and was significantly shorter than SS). At the end of year 4, DF and SS remained significantly taller than NF (the shortest and least shade-tolerant species), and at the end of year 6 only DF was significantly taller than the other species. Impact of overstorey management interventions Leader–lateral ratio Leader–lateral (Le:La) ratio was significantly different between species in all years (Figure 4 and Supplementary Table S1). DF had the highest Le:La but not significantly higher than SS in any year, both being >1 throughout the assessment period and following a similar trend over time. NF and SF had low Le:La, with NS having an intermediate Le:La of close to 1 throughout. Between the end of year 2 and the end of year 4, the Le:La of DF and SS fell, and that of NF increased. Le:La ratios of all species increased between the end of year 6 (following the June 2012 thinning) and the end of year 8. At the end of year 9 and year 10, following the overstorey removal during the ninth growing season, there was limited evidence that the Le:La ratios of all species appeared to be gradually converging close to 1.0. Health and lean scores For all species there was a small drop in the proportion of trees achieving health score 1 (HS1) between year 4 and year 6, following the second overstorey thinning operation (Figure 4). At the end of growing seasons 9 and 10, the proportion achieving HS1 had fallen for all species, following the overstorey removal during year 9; this was particularly marked and worsened between years 9 and 10 for DF. Only a quarter of DF trees achieved HS1 at the end of year 10 (although 72 per cent of DF trees achieved either HS1 or 2 (HS1,2) indicating that the damaged trees were likely to recover). There were no significant differences between species in the proportion of trees achieving HS1 or HS1,2 in any year (Supplementary Table S1). DF had the lowest proportion of trees receiving lean score 1 (LN1) and lean score 1 or 2 (LN1,2) in both years 9 and 10, following overstorey removal (Supplementary Table S1). However, this was only significant for LN1 in year 10, and post hoc Bonferroni tests could not distinguish which species had significantly different scores from each other. Lean scores worsened between years 9 and 10 for DF and improved between years 9 and 10 for NS but remained similar for the other species. The proportion of trees with LN1 was lower than would be desired in normal forest management. Experiment 2: Performance of underplanted and open-grown Douglas-fir seedlings The mean survival, stem diameter and height of underplanted and open-grown seedlings at Wythop are shown in Figure 5. There was no significant effect of canopy cover on either survival or height of seedlings in any year (P > 0.1 in all cases). There was little difference in stem diameter between underplanted or open-grown seedlings in years 0–1, but a small difference was apparent in year 2 (P = 0.0253, s.e.d = 0.234). However, in year 3 stem diameter of open-grown seedlings was significantly larger than that of underplanted seedlings (P < 0.001). Figure 5 Open in new tabDownload slide Mean percentage survival, mean stem diameter (mm) and mean height (cm) of underplanted and open-grown Douglas-fir seedlings at the Wythop site after the first three growing seasons. Error bars are standard error of the mean for stem diameter and height. Figure 5 Open in new tabDownload slide Mean percentage survival, mean stem diameter (mm) and mean height (cm) of underplanted and open-grown Douglas-fir seedlings at the Wythop site after the first three growing seasons. Error bars are standard error of the mean for stem diameter and height. The results at Great Glen (Figure 6) contrast sharply with those at Wythop; survival of open-grown seedlings was very much lower than that of underplanted seedlings, with most deaths of open-grown seedlings occurring in the first year after planting. Despite the poor survival, both height and diameter growth of the surviving open-grown seedlings at Great Glen were markedly larger than for the underplanted seedlings. At Clocaenog, higher survival was seen in the open-grown plots than the underplanted plots, but the difference was not marked (Figure 7). However, as was seen at Great Glen, both the height and diameter growth of open-grown seedlings was very much larger than that of underplanted seedlings. There did not appear to be any clear benefit of mounding to underplanted trees. Figure 6 Open in new tabDownload slide Mean percentage survival, mean height (cm) and mean stem diameter (mm) of underplanted and open-grown Douglas-fir seedlings at the Great Glen site after the first five growing seasons. Error bars are standard error of the mean for stem diameter and height. Figure 6 Open in new tabDownload slide Mean percentage survival, mean height (cm) and mean stem diameter (mm) of underplanted and open-grown Douglas-fir seedlings at the Great Glen site after the first five growing seasons. Error bars are standard error of the mean for stem diameter and height. Figure 7 Open in new tabDownload slide Mean percentage survival, mean height (cm) and mean stem diameter (mm) of underplanted and open-grown Douglas-fir seedlings at the Clocaenog site after the first five growing seasons. Error bars are standard error of the mean for stem diameter and height. *Stem diameter in open-grown treatment was recorded at 1.3 m above ground level in year 5, rather than at 5 cm above ground level. Stem diameter was not recorded in year 0. Figure 7 Open in new tabDownload slide Mean percentage survival, mean height (cm) and mean stem diameter (mm) of underplanted and open-grown Douglas-fir seedlings at the Clocaenog site after the first five growing seasons. Error bars are standard error of the mean for stem diameter and height. *Stem diameter in open-grown treatment was recorded at 1.3 m above ground level in year 5, rather than at 5 cm above ground level. Stem diameter was not recorded in year 0. Microclimate On the hottest days during the first growing season, there was little difference in maximum temperature between the underplanted and open-grown sites at Wythop (Figure 8). However, on the coldest day during the first winter (1 February 2012), temperatures on the open sites fell to a lower value than on the underplanted sites (Figure 8). Peak wind speeds at Wythop were also lower on the underplanted sites than the open-grown sites and peak light transmission was reduced to around 50 per cent of that on the open sites (Figure 8) despite relatively low basal areas of 25–31 m2 ha−1. Mean monthly temperatures did not differ between underplanted and open sites during the first year after planting (Supplementary Figure S1), but mean monthly wind speeds, radiation and total monthly rainfall were all higher in the open sites than the underplanted sites. Figure 8 Open in new tabDownload slide Temperature during the coldest period of the first winter for Wythop (30 January–1 February 2012), Clocaenog (10–12 March 2013) and Great Glen (20–22 February 2013). Temperature and radiation (W m−2) during the hottest period of the first growing season for Wythop (25–27 July 2011), Clocaenog and Great Glen (both 23–25 May 2012). Wind speed during windiest period of the first year for Wythop (3–5 January 2012) and Clocaenog (23–25 April 2012). Radiation and wind speed not recorded at Great Glen. Figure 8 Open in new tabDownload slide Temperature during the coldest period of the first winter for Wythop (30 January–1 February 2012), Clocaenog (10–12 March 2013) and Great Glen (20–22 February 2013). Temperature and radiation (W m−2) during the hottest period of the first growing season for Wythop (25–27 July 2011), Clocaenog and Great Glen (both 23–25 May 2012). Wind speed during windiest period of the first year for Wythop (3–5 January 2012) and Clocaenog (23–25 April 2012). Radiation and wind speed not recorded at Great Glen. At Great Glen (the most Northerly site) during the coldest period of the first winter, the temperature range on the open-grown sites was much wider than on the underplanted sites, with rapid swings from very cold to very warm temperatures each day, while the temperature range on the underplanted sites at Great Glen was smaller (Figure 8). This buffering of extreme temperatures in the underplanted sites was also seen during the period of highest summer temperatures in the first growing season, with maximum temperatures recorded on the open-grown sites as much as 18°C higher than those in the understorey sites (Figure 8). Mean monthly temperature records at Great Glen during the first year following planting also indicated that the open sites experienced wider ranging temperatures, with higher mean temperatures than the underplanted sites in the summer months and lower mean temperatures in the winter months (Supplementary Figure S3). Similar patterns were seen during cold periods in January 2015 and January 2016 (the third and fourth winters following planting; data not shown) where minimum temperatures recorded on the open-grown sites were consistently around 2°C colder than those recorded in the underplanted sites. At Clocaenog there was almost no difference in temperature between the open-grown and underplanted sites during either the hottest or the coldest periods (Figure 8) or the mean monthly records for the first year following planting (Supplementary Figure S2). Light transmission was a lot lower in the underplanted site, with peak radiation values only around 30 per cent of those on the open site (Figure 8 and Supplementary Figure S2) and mean monthly rainfall was lower (Supplementary Figure S2). The peak wind speed during the windiest period of the first year was also lower in underplanted sites (Figure 8), although mean monthly wind speeds were not lower (Supplementary Figure S2). Discussion Performance of conifer species All of the five underplanted species trialled at Experiment 1 were able to establish under the canopy, despite having a range of shade tolerances (Niinements and Valladares, 2006). Survival rates of all species did reduce over the period, with SF (the most shade tolerant) and NF (the most light demanding) having the lowest survival but remained comparable with rates that would be acceptable on a restock site, where some replacement (‘beating up’) would be expected in standard forestry practice. Five years after planting, the survival of SS, NS and DF was all in the region of 95 per cent, comparable with 5 year survival of 96 per cent on the open-grown sites at the nearby Experiment 2. Sitka spruce and Douglas-fir performed well, as might be expected for species that are known to be well suited to the site (the SS overstorey productivity is ~18–20 m3 ha−1 year−1). The similar height growth of DF and SS growing under larch overstories of different densities was also reported by Neustein and Jobling (1970). Differences in height between species are likely to be influenced by differences in size of planting stock, at least in the early years. Although not formally recorded for the assessed trees in this experiment, the mean height of trees of the same nursery stock that were underplanted in the surrounding area were recorded during the first growing season. These data indicate that in addition to having the largest heights at 10 years, SS and DF also had the largest 10-year height increments (Supplementary Table S2). Abies species are known to establish slowly, with growth rates increasing after the first few years (Macdonald et al., 1957; Kerr et al., 2015); the SF planting stock was a similar size to that of DF, and although SF had a smaller 10-year increment (Supplementary Table S2) the increasing growth rates of SF seen between years 6 and 10 support this. Despite differences in survival and growth rates, all of the species did establish under the overstorey in sufficient numbers to fully stock a future stand, indicating that underplanting a shelterwood with any of the species would be a viable method of diversifying an existing forest. Impact of overstorey management interventions Many variables have been studied in an attempt to predict the response of naturally regenerated seedlings to release from overstorey competition (e.g. Reich et al., 1998; Metslaid et al., 2007). The physiological characteristics of the species, shade tolerance, growth rate prior to release, growth plasticity and several other factors affect the acclimation ability of a species and response to release (Metslaid et al., 2007). The different responses of the underplanted species in this study to thinning and overstorey removal are consistent with this. Changes in the relative growth rates of above- and below-ground tissues have been reported following release, with increased growth usually seen in the roots first (Kneeshaw et al., 2002); it has been suggested that the changes in allocation following release maintain a balance between crown and root functions (Nikinmaa, 1992). Youngblood (1991) reported a lag of several years before an increase in diameter growth was seen following release of Picea glauca ((Moench) Voss). The sharp reduction in survival of DF during the 10th growing season suggests that the species, which has lower shade tolerance than SS, NS or SF (Niinements and Valladares, 2006), was not able to function efficiently in the environmental conditions following overstorey removal. It is possible that all available resources were being allocated to DF root growth to balance crown and root functions, with little excess available for above-ground growth, while the other species were able to respond to the changed conditions with an increase in diameter growth rate. Although trees were not killed during operations, health scores decreased, particularly for DF, both during the 9th and 10th growing seasons; DF also had the highest proportion of leaning trees of all species. Harvesting had been planned for early summer to take advantage of dry ground conditions and to allow the seedlings a growing season to consolidate before winter. However, operations were delayed until October and were immediately followed by number of storms with very strong winds and heavy rainfall throughout winter 2015/2016. Many trees developed holes at the base caused by excessive rocking movement in very wet ground (known as ‘socketing’). The socketing and leaning damage was variable across the site but was most severe in taller plants, particularly DF. Empirical observations from forest experiments and managers’ experiences also suggest that understorey DF saplings have a high risk of uprooting during strong winds or heavy snow fall; this is thought to be caused by poor development of the structural root system in conditions of light limitation and overstorey competition (Kuehne et al., 2015). Douglas-fir seedlings require current photosynthate for new root growth, rather than using carbohydrate reserves (Philipson, 1987). Therefore, although young DF can tolerate low light levels and maintain steady height growth (as observed in this study), this is at the expense of root growth, leading to unfavourable root-to-shoot ratios, particularly in dense or highly competitive stands (Devine and Harrington, 2008; Kuehne et al., 2015). The Le:La ratio has long been used as an indicator of light limitation, with trees growing in high light conditions producing longer leader than lateral branches, resulting in ratios greater than 1 (e.g. Chen et al., 1996). However, Le:La ratio is also strongly influenced by the morphology and physiology of a species. Le:La ratios of SS and DF fluctuated more widely than those of the other species, particularly in response to the thinning in year 6, and overstorey removal, but remained greater than or equal to 1 throughout the study. These species are less shade tolerant than SF and NS and may invest more heavily in height growth as a survival strategy in low light conditions. However, our results contrast with those reported by others that species with lower shade tolerance tend to show less morphological change in response to changing light conditions (Chen et al., 1996; Williams et al., 1999; Claveau et al., 2002). The two species with the highest shade tolerance in our study, SF and NS (according to Niinements and Valladares, 2006), had lower Le:La ratios than DF and SS, remaining below 1.0 throughout the study and appeared to show less response to changes in light conditions; Le:La ratios of SF and NS have been shown to be similar in other studies (Grassi and Giannini, 2005). Low Le:La ratio may be a physiological trait of some shade-tolerant species, and one of the reasons that they have high shade tolerance (rather than being an indicator of light limitation). Trees that have long lateral branches are likely to have high light capture efficiency per unit of above-ground biomass and lower self-shading, while those with shorter lateral branches and a more columnar form, such as SS and DF, are likely to have lower light capture efficiency and lower shade tolerance (Kohyama, 1987). Regardless of the Le:La ratio values of each species, all were able to establish under an overstorey in sufficiently high numbers to restock, indicating that light limitation was not a prohibitive factor for any species in this study. Therefore the ‘rule of thumb’ that Le:La ratios <1 indicate light limitation, sometimes applied in British forestry conditions, and probably based largely on observations of Sitka spruce, does not appear to be appropriate for all species during the early years following planting. Performance of open-grown and underplanted seedlings The degree of shelter and shading is dependent on the properties of the overstorey and varied between the three sites in our study. At Wythop although there was little buffering effect of the overstorey on peak temperatures during the hottest period of the first growing season, the minimum temperatures were less extreme under the canopy. However, as the presence of the overstorey did not significantly improve the survival or height of the seedlings, this suggests that the colder temperatures on the open sites were not limiting the seedlings. In contrast, the reduced stem diameter of underplanted seedlings compared with open-grown seedlings may indicate that in the understorey environment, with lower wind speeds and light levels, the available resources had been allocated to favour height or branch growth at the expense of diameter growth. Morphological plasticity of tree growth to suit the environmental conditions in which they are growing has been shown by Williams et al. (1999), MacFarlane and Kane (2017) and others. Under the microclimate conditions at Wythop, there was no apparent benefit of the shelter effect to the seedlings. In contrast, at Great Glen the height and diameter growth of seedlings were better on the open-grown site than the underplanted site, but the survival at the end of the first year was very much lower on the open-grown site (30 per cent compared with 78 per cent) with most of the deaths occurring during the first year. Löf et al. (2005) also reported higher diameter growth and lower survival rates for seedlings planted on clear-cut sites than those under canopies with 15 and 20 per cent light transmission. The cause of the deaths on the Great Glen open-grown site is not known, and it is possible that factors such as deer browsing or damage by the large pine weevil Hylobius abietis were involved; however, mortality was very much lower in the nearby underplanted plots, where similar biotic impacts might have been expected. The high mortality may be due to the high level of exposure on the open site at Great Glen (the most Northerly site); the overstorey was shown to reduce the temperature extremes compared with the open-grown sites. However, the reduced light environment resulted in slow growth, and the establishment phase will be extended. The overstorey at Great Glen was relatively light, open and well-spaced, with basal area of only 20 m2 ha−1, having been well thinned as a seed stand, so it is interesting that the seedling growth rate was so affected. For managers, the trade-off on underplanted sites is whether the increased seedling survival outweighs the penalty of reduced growth rates. The benefits of shelter during seedling establishment may become increasingly important as extreme weather events, such as unseasonal frosts, or summer drought worsen or become more frequent with climate change (Moffat et al., 2012). Although the presence of the dense overstorey had little impact on temperatures at Clocaenog, the light transmission and rainfall were much lower than on the open site, and growth rates were very reduced. A meta-analysis of underplanted seedling performance across four biomes showed that survival and growth improves as stand density decreases to an intermediate level, below which survival and growth either drop or stabilize (Paquette et al., 2006). In the trade-off between sufficient light for seedling growth and sufficient shelter (Man and Lieffers, 1999; Langvall and Örlander, 2001; Paquette et al., 2006; MacFarlane and Kane, 2017), the dense canopy cover at the Clocaenog stand appeared to restrict growth rates of the underplanted seedlings. As this stand is in the early stages of transformation to a shelterwood, the basal area requires significant reduction before the next generation of trees could be safely established, whether naturally regenerated or underplanted. However, there is also a trade-off between the advantages of early establishment of the next generation and negative impacts of reducing the overstorey basal area on stand volume production. Management aspects The overstorey basal areas at Experiment 1 (c. 25 m2 ha−1) and at Wythop in Experiment 2 (25–31 m2 ha−1) were sufficiently low to allow growth of underplanted seedlings, while the higher basal area of 38 m2 ha−1 at Clocaenog in Experiment 2 appeared to reduce light levels and seedling growth significantly. Maintaining low basal area required frequent thinning at Experiment 1, and operations were very carefully managed to minimize damage to the underplanted trees. Felling directions were marked on each overstorey tree, permanently marked racks were used, trees were processed in the rack, and timber was stacked in designated stacking areas. Some unavoidable damage occurred where felled trees had been dragged to the racks for processing, but the damage rates were lower than those reported in Stokes et al. (2009) where such steps were not taken. However, the underplanted trees in the assessment plots were favoured during thinning operations, and some trees outside the assessment plots were killed or severely damaged. Within a uniform shelterwood, where underplanting may be carried out throughout the stand, some damage to planted trees is very likely, but in systems where not all areas are planted, i.e. group selection or group shelterwood, damage will be easier to limit with careful management. Where thinning operations are not carefully controlled, as was seen at the Wythop site, damage to the underplanted seedlings may be severe; the Wythop plots were abandoned following severe harvesting damage in the third growing season. Permanently marked access racks had not been designated at Wythop and the machine operator used the underplanted areas to access the stand. Frequent thinning of both shelterwoods and CCF systems is essential to maintain a suitable light environment for seedling growth, and access and operations must be well planned if underplanted seedlings are to survive. It may not be possible to underplant continuously beneath uniform shelterwoods without sacrificing some of the area for operational access; in some cases this may mean that that objective of uniform and complete restocking cannot be achieved. Conclusions All of the species trialled were able to survive and grow under the overstorey at Experiment 1 where the overstorey basal area was maintained close to 25 m2 ha−1. The overstorey was managed using simple, operationally practical methods that resulted in low levels of damage to underplanted seedlings. Early growth rates differed between the species trialled, and managers accustomed to the rapid establishment of SS may need to adjust their expectations. Some of the slower establishing species, such as SF, appeared to be catching up with SS in later years and may invest more heavily in root growth during the early years. The importance of considering ongoing growth after overstorey removal is clear; despite good early performance under the canopy, DF had a poor response to the increased exposure on the site. Douglas-fir may be better suited to underplanting on very sheltered sites where poorly developed root structures are less important during the period of overstorey removal or in CCF systems where the overstorey is not removed. Canopy shelter during establishment can be beneficial if exposure is a limiting factor; however, there is a trade-off against light levels which may result in lower growth rates. If exposure is not limiting, seedlings may not derive any benefit from the presence of an overstorey, although other benefits such as reduced landscape impact and continuity of habitat may result from restocking without clearfelling. In addition, the use of underplanting may allow managers to start to establish the next crop and increase species diversity at an earlier stage, rather than waiting for rotation age to clearfell and restock. Funding This work was funded by the Forestry Commission’s Science and Innovation Strategy Programme 3 on Managing Resilient Forests. Experiment 1 was established under the ‘Tyfiant Coed’ project funded by the then Forestry Commission Wales. Experiment 2 was co-funded by the Forestry Commission and the European Union programme FEDER-INTERREG IV Atlantic ‘REINFFORCE’ project. Acknowledgements We dedicate this paper to our friend and colleague Tom Connolly, for his wise advice on all aspects of life and statistics. We would like to thank many members of the Forest Research Technical Services Unit who carried out the field assessments at all sites. Experiment 1 was designed and set up by staff and students at the University of Wales Bangor, led by Dr. Arne Pommerenning. The sites at Great Glen and Wythop were hosted by Forestry and Land Scotland and Forestry England respectively and we thank them for use of the sites. Ewan Mackie, Bill Mason, Helen McKay and two anonymous reviewers provided helpful comments on earlier drafts of the manuscript. The authors would particularly like to thank Dave Williams, Natural Resources Wales for his innovative support and management of the CCF National Trial site at Clocaenog Forest over many years. References Bengtsson , J. , Nilsson , S.G., Franc , A. and Menozzi , P. 2000 Biodiversity, disturbances, ecosystem function and management of European forests . For. Ecol. Manage. 132 , 39 – 50 . Google Scholar Crossref Search ADS WorldCat Brandeis , T.J. , Newton , M. and Cole , E.C. 2001 Underplanted conifer seedling survival and growth in thinned Douglas-fir stands . Can. J. For. Res. 31 , 302 – 312 . Google Scholar Crossref Search ADS WorldCat Calder , I.R. , Reid , I., Nisbet , T. and Green , J.C. 2003 Impact of lowland forests in England on water resources – Applications of the HYLUC model . Water Resour. Res. 39 , 1319 – 1328 . Google Scholar OpenURL Placeholder Text WorldCat Cameron , A.D. 2015 Building resilience into Sitka spruce (Picea sitchensis [bong.] Carr.) forests in Scotland in response to the threat of climate change . Forests 6 , 398 – 415 . Google Scholar Crossref Search ADS WorldCat Cannell , M.G.R. and Smith , R.I. 1986 Climatic warming, spring budburst and frost damage on trees . J. Appl. Ecol. 23 , 177 – 191 . Google Scholar Crossref Search ADS WorldCat Chen , H.Y.H. , Klinka , K. and Kayahara , G.J. 1996 Effects of light on growth, crown architecture, and specific leaf area for naturally established Pinus contorta var. latifolia and Pseudotsuga menziesii var. glauca saplings . Can. J. For. Res. 26 , 1149 – 1157 . Google Scholar Crossref Search ADS WorldCat Childs , S.W. and Flint , L.E. 1987 Effect of shadecards, shelterwoods, and clearcuts on temperature and moisture environments . For. Ecol. Manage. 18 , 205 – 217 . Google Scholar Crossref Search ADS WorldCat Claveau , Y. , Messier , C., Comeau , P.G. and Coates , K.D. 2002 Growth and crown morphological responses of boreal conifer seedlings and saplings with contrasting shade tolerance to a gradient of light and height . Can. J. For. Res. 32 , 458 – 468 . Google Scholar Crossref Search ADS WorldCat Desprez-Loustau , M.-L. , Marcais , B., Nageleisen , L.-M., Piou , D.A. and Vannini , A. 2006 Interactive effects of drought and pathogens in forest trees . Ann. For. Sci. 63 , 597 – 612 . Google Scholar Crossref Search ADS WorldCat Devine , W.D. and Harrington , T.B. 2008 Belowground competition influences growth of natural regeneration in thinned Douglas-fir stands . Can. J. For. Res. 38 , 3085 – 3097 . Google Scholar Crossref Search ADS WorldCat Erefur , C. , Bergsten , U. and de Chantal , M. 2008 Establishment of direct seeded seedlings on Norway spruce and scots pine: Effects of stand conditions, orientation and distance with respect to shelter tree, and fertilisation . For. Ecol. Manage. 255 , 1186 – 1195 . Google Scholar Crossref Search ADS WorldCat GenStat . 2013 GenStat for Windows 16th Edition . VSN International . Web page: GenStat.co.uk Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gerlach , J.P. , Reich , P.B., Puettmann , K. and Baker , T. 1997 Species, diversity, and density affect tree seedling mortality from Armillaria root rot . Can. J. For. Res. 27 , 1509 – 1512 . Google Scholar Crossref Search ADS WorldCat Grant , A. , Worrell , R., Wilson , S., Ray , D. and Mason , B. 2012 Achieving diversity in Scotland’s forest landscapes. Forestry Commission Scotland Practice Guide . Forestry Commission , p. 30 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Grassi , G. and Giannini , R. 2005 Influence of light and competition on crown and shoot morphological parameters of Norway spruce and silver fir saplings . Ann. For. Sci. 62 , 269 – 274 . Google Scholar Crossref Search ADS WorldCat Groot , A. and Carlson , D.W. 1996 Influence of shelter on night temperatures, frost damage and bud break of white spruce seedlings . Can. J. For. Res. 26 , 1531 – 1538 . Google Scholar Crossref Search ADS WorldCat Grossnickle , S.C. 2000 Ecophysiology of Northern spruce species: The performance of planted seedlings . NRC Research Press , p. 409 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hale , S. 2004 Managing Light to Enable Natural Regeneration in British Conifer Forests. In Forestry Commission Information Note 63 . Forestry Commission , p. 6 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hanewinkel , M. , Kuhn , T., Bugmann , H., Lanz , A. and Brang , P. 2014 Vulnerability of uneven-aged forests to storm damage . Forestry 87 , 525 – 534 . Google Scholar Crossref Search ADS WorldCat Holgén , P. and Hånell , B. 2000 Performance of planted and naturally regenerated seedlings in Picea abies-dominated shelterwood stands and clearcuts in Sweden . For. Ecol. Manage. 127 , 129 – 138 . Google Scholar Crossref Search ADS WorldCat Hungerford , R.D. and Babbitt , R.E. 1987 Overstorey removal and residue treatments affect soil surface, air and soil temperature: implications for seedling survival . USDA Forest Surface, Intermountain Research Station Research Paper INT-377 , p. 24 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Jactel , H. , Bauhus , J., Boberg , J., Bonal , D., Castagneyrol , B., Gardiner , B., et al. 2017 Tree diversity drives forest stand resistance to natural disturbances . Curr. Forestry Rep. 3 , 223 – 243 . Google Scholar Crossref Search ADS WorldCat Jiménez , E. , Vega , J.A., Pérez-Gorostiaga , P., Cuiñas , P., Fonturbel , T., Fernández , C., et al. 2008 Effects of pre-commercial thinning on transpiration in young post-fire maritime pine stands . Forestry 81 , 543 – 557 . Google Scholar Crossref Search ADS WorldCat Kerr , G. and Haufe , J. 2016 Successful Underplanting . Forest Research , Alice Holt Lodge, Farnham, Surrey GU10 4LH . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kerr , G. , Stokes , V., Peace , A. and Jinks , J. 2015 Effects of provenance on survival, growth and stem form of European silver fir (Abies alba mill.) in Britain . Eur. J. For. Res. 134 , 349 – 363 . Google Scholar Crossref Search ADS WorldCat Kneeshaw , D.D. , Williams , H., Nikinmaa , E. and Messier , C. 2002 Patterns of above- and below-ground response of understory conifer release 6 years after partial cutting . Can. J. For. Res. 32 , 255 – 265 . Google Scholar Crossref Search ADS WorldCat Kohyama , T. 1987 Significance of architecture and allometry in saplings . Funct. Ecol. 1 , 399 – 404 . Google Scholar Crossref Search ADS WorldCat Kuehne , C. , Karrié , C., Forrester , D.I., Kohnle , U. and Bauhus , J. 2015 Root system development in naturally regenerated Douglas-fir saplings as influenced by canopy closure and crowding . J. For. Sci. 61 , 406 – 415 . Google Scholar Crossref Search ADS WorldCat Lafond , V. , Lagarrigues , G., Cordonnier , T. and Courbaud , B. 2014 Uneven-aged management options to promote forest resilience for climate change adaptation: Effects of group selection and harvesting intensity . Ann. For. Sci. 71 , 173 – 176 . Google Scholar Crossref Search ADS WorldCat Langvall , O. and Löfvenius , M.O. 2002 Effect of shelterwood density on nocturnal near-ground temperature, frost injury risk and budburst date of Norway spruce . For. Ecol. Manage. 168 , 149 – 161 . Google Scholar Crossref Search ADS WorldCat Langvall , O. and Örlander , G. 2001 Effects of pine shelterwoods on microclimate and frost damage to Norway spruce seedlings . Can. J. For. Res. 31 , 155 – 164 . Google Scholar Crossref Search ADS WorldCat Löf , M. , Paulsson , R., Rydberg , D. and Welander , N.T. 2005 The influence of different overstory removal on planted spruce and several broadleaved tree species: Survival, growth and pine weevil damage during three years . Ann. For. Sci. 62 , 237 – 244 . Google Scholar Crossref Search ADS WorldCat Lowe , J.A. , Bernie , D., Bett , P., Bricheno , L., Brown , S., Calvert , D. et al. 2018 UKCP18 Science Overview Report . The Met Office . https://www.metoffice.gov.uk/pub/data/weather/uk/ukcp18/science-reports/UKCP18-Overview-report.pdf ( accessed on 10 September, 2019) . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Maas-Herbner , K.G. , Emmingham , W.H., Larson , D.J. and Chan , S.S. 2005 Establishment and growth of native hardwood and conifer seedlings underplanted in thinned Douglas-fir stands . For. Ecol. Manage. 208 , 331 – 345 . Google Scholar Crossref Search ADS WorldCat MacFarlane , D.W. and Kane , B. 2017 Neighbour effects on tree architecture: Functional trade-offs balancing crown competitiveness with wind resistance . Funct. Ecol. 31 , 1624 – 1636 . Google Scholar Crossref Search ADS WorldCat Macdonald , J. , Wood , R.F., Edwards , M.V. and Aldhous , J.R. 1957 Exotic forest trees in Great Britain. Forestry Commission Bulletin 30 . HMSO , p. 168 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Man , R. and Lieffers , V. 1999 Effects of shelterwood and site preparation on microclimate and establishment of white spruce seedlings in a boreal mixedwood forest . Forest. Chron. 75 , 837 – 844 . Google Scholar Crossref Search ADS WorldCat Mason , B. , Kerr , G. and Simpson , J. 1999 What is continuous cover forestry? In Forestry Commission Information Note 29 . Forestry Commission . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Mason , B. , Kerr , G., Pommerening , A., Edwards , C., Hale , S., Ireland , D., et al. 2004a Continuous cover forestry in British conifer forests. In Forest Research Annual Report and Accounts 2003-2004 . Forestry Commission . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Mason , W.L. , Edwards , C. and Hale , S.E. 2004b Survival and early seedling growth of conifers with different shade tolerance in a Sitka spruce spacing trial and relationship to understorey light climate . Silva Fenn. 38 , 357 – 370 . Google Scholar Crossref Search ADS WorldCat Mason , W.L. 2015 Implementing continuous cover forestry in planted forests: Experience with Sitka spruce (Picea sitchensis) in the British isles . Forests 6 , 879 – 902 . Google Scholar Crossref Search ADS WorldCat Meason , D.F. and Mason , W.L. 2014 Evaluating the deployment of alternative species in planted conifer forests as a means of adaptation to climate change—Case studies in New Zealand and Scotland . Ann. For. Sci. 71 , 239 – 253 . Google Scholar Crossref Search ADS WorldCat Metslaid , M. , Jõgiste , K., Nikinmaa , E., Moser , W.K. and Porcar-Castell , A. 2007 Tree variables related to growth response and acclimation of advance regeneration of Norway spruce and other coniferous species after release . For. Ecol. Manage. 250 , 56 – 63 . Google Scholar Crossref Search ADS WorldCat Moffat , A.J. , Morison , J.I.L., Nicoll , B. and Bain , V. 2012 Climate Change Risk Assessment for the Forestry Sector. Defra , p. 182 . http://www.defra.gov.uk/environment/climate/government/ Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Murray , M.B. , Smith , R.I., Leith , I.D., Fowler , D., Lee , H.S.J., Friend , A.D., et al. 1994 Effects of elevated CO2, nutrition and climatic warming on bud phenology in Sitka spruce (Picea sitchensis) and their impact on the risk of frost damage . Tree Physiol. 14 , 691 – 706 . Google Scholar Crossref Search ADS PubMed WorldCat Natural Resources Wales . 2017 Forest Resilience Guide 2: Improving the tree species diversity of Welsh woodlands . Natural Resources Wales , p. 30 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Neustein , S.A. and Jobling , J. 1970 Artificial regeneration: Underplanting of larch . Rep. Forest Res. London , 99 . Google Scholar OpenURL Placeholder Text WorldCat Niinements , U. and Valladares , F. 2006 Tolerance to shade, drought and waterlogging of temperate northern hemisphere trees and shrubs . Ecol. Monog. 76 , 521 – 547 . Google Scholar Crossref Search ADS WorldCat Nikinmaa , E. 1992 Analyses of the growth of scots pine; matching structure with function . Acta For. Fenn. 235 , 1 – 68 . Google Scholar OpenURL Placeholder Text WorldCat Nilsson , U. , Örlander , G. and Karlsson , M. 2006 Establishing mixed forests in Sweden by combining planting and natural regeneration – Effects of shelterwood and scarification . For. Ecol. Manage. 237 , 301 – 311 . Google Scholar Crossref Search ADS WorldCat Paquette , A. , Bouchard , A. and Cogliastro , A. 2006 Survival and growth of under-planted trees: A meta-analysis across four biomes . Ecol. Appl. 16 , 1575 – 1589 . Google Scholar Crossref Search ADS PubMed WorldCat Philipson , J.J. 1987 Root growth in Sitka spruce and Douglas-fir transplants: Dependence on the shoot and stored carbohydrates . Tree Physiol. 4 , 101 – 108 . Google Scholar Crossref Search ADS WorldCat Pyatt , G. , Ray , D. and Fletcher , J. 2001 An Ecological Site Classification for Forestry in Great Britain, Forestry Commission Bulletin 124 . Forestry Commission . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Reich , P.B. , Tjoelker , M.G., Walters , M.B., Vanderklein , D.W. and Buschena , C. 1998 Close association of RGR, leaf and root morphology, seed mass and shade tolerance in seedlings of nine boreal tree species grown in high and low light . Funct. Ecol. 12 , 327 – 338 . Google Scholar Crossref Search ADS WorldCat Sellars , H.C. 2006 Undercanopy Microclimatology of Sitka Spruce Plantation Forests: Implications for Natural Regeneration Ph.D. Thesis . University of Liverpool , Liverpool, UK , p. 187 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Smidt , M.F. and Puettmann , K.J. 1998 Overstory and understory competition affect underplanted eastern white pine . For. Ecol. Manage. 105 , 137 – 150 . Google Scholar Crossref Search ADS WorldCat Stokes , V. , Kerr , G. and Ireland , D. 2009 Seedling height and the impact of harvesting operations on advance regeneration of conifer species in upland Britain . Forestry 82 , 185 – 198 . Google Scholar Crossref Search ADS WorldCat Su , Q. , MacLean , D.A. and Needham , T.D. 1996 The influence of hardwood content on balsam fir defoliation by spruce budworm . Can. J. For. Res. 26 , 1620 – 1628 . Google Scholar Crossref Search ADS WorldCat van Lierop , P. , Lindquist , E., Sathyapala , S. and Franceschini , G. 2015 Global forest area disturbance from fire, insect pests, diseases and severe weather events . For. Ecol. Manage. 352 , 78 – 88 . Google Scholar Crossref Search ADS WorldCat Vanoni , M. , Bugmann , H., Nötzli , M. and Bigler , C. 2016 Drought and frost contribute to abrupt growth decreases before tree mortality in nine temperate tree species . For. Ecol. Manage. 382 , 51 – 63 . Google Scholar Crossref Search ADS WorldCat Williams , H. , Messier , C. and Kneeshaw , D.D. 1999 Effects of light availability and sapling size on the growth and crown morphology of understory Douglas-fir and lodgepole pine . Can. J. For. Res. 29 , 222 – 231 . Google Scholar Crossref Search ADS WorldCat Youngblood , A.P. 1991 Radial growth after a shelterwood seed cut in a mature stand of white spruce in interior Alaska . Can. J. For. Res. 21 , 410 – 413 . Google Scholar Crossref Search ADS WorldCat © Crown copyright 2020. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Underplanting is a practical silvicultural method for regenerating and diversifying conifer stands in Britain
JF - Forestry
DO - 10.1093/forestry/cpaa027
DA - 2021-03-04
UR - https://www.deepdyve.com/lp/oxford-university-press/underplanting-is-a-practical-silvicultural-method-for-regenerating-and-ez1D3ufkiT
SP - 219
EP - 231
VL - 94
IS - 2
DP - DeepDyve
ER -