Research at Europe Regional Climate Centre
Trees, leaf litter, decomposition
By June 2008 trees had been cencused in two plots, with a total of 1964 stems measured. Deadwood surveys were carried out along transects, and a full cencus of deadwood was done within sub-plots. A further plot had been cencused by December, and leaf litter collection begun: monthly in the summer and fortnightly during autumn. Deadwood surveys continued. In the first quarter of 2009 seven more plots were established. An example of a fully mapped plot is seen in figure 4, which shows tree location, tree species and diameter class.
Figure 4 Tree map of Stroud plot
In the first quarter of 2009, seven more sample plots were identified. GIS (geographical information systems) analysis of forests in the upper Thames region showed that 65% of woodland was within 60 m of the forest edge, highlighting the need to study forest edges and fragments. Data collection continued throughout 2009, including the measurement of 2773 trees in the new sample plots. Leaf litter collected in 2008, which had been dried, was sorted throughout spring of 2009. Dendrometer bands in the Mixed Core plot were measured monthly, bags were prepared for the leaf litter decomposition experiment, and leaf litter traps established in new plots. By December of that year the leaf litter decomposition experiment was set up, dendrometers established in three more plots and coarse woody debris was surveyed in two plots.
In the first quarter of 2010 the first set of leaf litter decomposition bags were retrieved, and dendrometer bands fitted in a further five plots. Initial results from the leaf litter decomposition experiment (fig 5) show that litter in bags with larger mesh decay more quickly, i.e. that soil macro invertebrates play a large role in leaf litter decomposition. Decomposition was slower at the forest edge than the interior, and slightly slower in control than watered plots. As this project is based on forest fragmentation, it is interesting to see that leaf litter has a different decomposition rate at the edge compared to the interior of the plot.
Re-censusing of the plots began in the second quarter of 2010, which, when completed, will allow analysis of how the forest changes over time. Volunteers assisted with dendrometer measurements and assessment of the forest canopy with canopyscopes, both techniques proving good for working with volunteers. Soil respiration collars were installed in the plots, allowing soil respiration to be measured every three weeks.
In the third quarter of 2010 regular measurements continued, including monthly measuring of dendrometers and surveys of tree canopy cover, fortnightly leaf litter collections and measurement of soil respiration every three weeks.
The 2nd census of beech core plot completed in September, two years after the 1st census.
Bait lamina sticks, which assess feeding activity of soil invertebrates by measuring food eaten at different depths in the soil, were retrieved, showing less activity than previous retrievals. This was attributed to the dryness of the soil. The experiment will be repeated in December.
Results of leaf litter decomposition experiments were presented at the British Ecological Society annual conference in a talk entitled “The effect of proximity to forest edge, moisture conditions and soil macrofauna on leaf litter breakdown and decomposition”.
Figure 5 Decomposition of the leaf litter in the watering experiment in the forest edge and interior. C, W, control and watered, respectively. Int, interior of plot. Error bars are standard deviations.
Between October and December 2010, the last leaf litter decomposition bags were collected, and data analysis begun. The bait-lamina study worked well, and appears to correlate with macrofauna abundance across the site – data analysis is ongoing. Leaf litter decomposition and bait-lamina studies are conducted in collaboration with Eleanor Slade. Dendrometer bands and canopy cover continued to be measured monthly in all plots, and soil respiration measurement has continued at three weekly intervals. Measurement of tree height and crown characteristics of all trees with dendrometer bands has begun, with three plots completed so far. This provides a more detailed picture of a subset of trees across the full range of species and size classes. A field trip from University of Oxford School of Geography and the Environment, organised by Terhi, gave 65 masters students a chance to see and hear about our research at Wytham.
Work went as planned between January and March 2011, with dendrometer and canopy scope measurements continuing monthly, and soil respiration observed every three weeks. In March, results to date were presented at a three day conference on forests and global change, hosted by the British Ecological Society.
Despite Climate Champions not being present at this time of year, regular measurement of dendrometers, canopy cover and soil respiration continued through the second quarter of 2011. The annual intensive spring campaign, during which stem and soil respiration are measured twice a week in the core plot, went according to plan, as did the second censuses of three plots, and deadwood and tree height surveys in a number of plots. Following analysis of data collected by “volunteers” and “experts”, a presentation was given at Oxford University, highlighting the amount of error in both sets of measurements.
The third quarter of 2011 saw the summer field season progressing as planned, with monthly dendrometer and canopy scope readings, and three-weekly soil respiration measurements. The 2nd census of Higgins plot, two years after it was first censused, will provide information as to how the carbon stocks have changed there. 2nd censuses for two more plots are planned for the final quarter of 2011. The team are now busy analysing the data collected over the last years, and preparing papers for publication. Visiting scientists from India and the USA have been introduced to the plots, giving the Indian visitors an insight into how to set up this type of plot, and bringing possibilities of collaboration with the American scientists.
Moths and small mammals
In spring and summer of 2008 animal surveys were carried out in three plots. It was found that there was considerable variation in the species of moths present in the different plots, although overall species abundance did not differ between plots. Woodmice and bank voles were the most common mammals caught. It was discovered that woodland connectivity is important for some species of moth, and a comprehensive list of moth species found on the plots was established. After moth trapping in two areas in two consecutive weeks, the importance of carrying out comparative surveys at the same time was clearly realised – climatic variation between weeks was thought to seriously affect the moths species caught.
Over the winter of 2009 data analysis from the previous season was completed, the design of the moth study finalised and a moth field guide collated for volunteers.
In the moth survey experiment in summer of 2009, 14,719 individual moths were caught, representing 87 of the 92 species known to be present at Wytham Woods. A number of uncommon moths were found, including several first sightings for the region for that year. Data analysis showed that large fragments of woodland, particularly the core (central) area, are good for woodland specialist species. Small fragments have nearly as many species as large fragments, but not as many species of woodland specialists are found there. Hedgerows and even isolated oak trees were found to be important for moths moving across the landscape; woodland species were less mobile than ubiquitous species.
Two years of small mammal data found that there were differences in overall population sizes between sampled areas. For both bank voles and woodmice there were more males than females found. However, it cannot be certain whether this reflects the actual population, or whether it is due to differences in male and female activity that result in more males being trapped.
In the 1st Quarter of 2010 data analysis for the moths continues, with a peer-reviewed scientific publication planned.
In the second quarter of 2010 very few small mammals were trapped, which may have been due to the cold winter. Validation of data collection by volunteers continued. It was found that abundance of woodlice and millipedes had no clear link with fragment size or decomposition rate in that area. The rare garden tiger moth was sited more than once, which was the only record for that species in the Upper Thames Basin so far in 2010.
Between July and September 2010 capture-release-recapture experiments with moths were carried out. Results included the finding that that moths fly straight towards hedgerows and forest edges on release. Moths moved up to 2.5 km over a number of nights. Previous work on moths was presented at the British Ecological Society annual conference in a talk entitled “A landscape-scale study of wodland moths: the importance of patch size and connectivity”.
Woodlice and millipede surveys found lower numbers than previously, attributed to the dry weather. As usual, there were more woodlice than millipedes in all sites.
One team assisted with checking bat boxes. Many boxes showed signs of recent habitation, and four had bats inside – including a single box with 28 natterer bats.
In the 4th quarter of 2010, as well as the collaborative work with Terhi Riutta on leaf litter decomposition and the bait-lamina study, woodlice and millipede surveys in 9 sites were conducted by a 7 day team. A total of 314 individuals were found! This is much higher than the numbers in August, due to the wetter conditions. It seems that there is a strong correlation between soil moisture and the number of species found at a site; in terms of the total numbers of individuals, there is a link between millipede abundance and soil moisture, but there is no such link for woodlice. Data from 2009 and 2010 moth surveys is being analysed and modelled in collaboration with researchers at Helsinki University.
During January to March 2011 data from the previous season on moths and mammals was analysed, with several papers currently in preparation. Moving into the next quarter, work will focus on mammal trapping in forest cores and edges, as well as hedgerows.
In the second quarter of 2011 climate champions continued mammal trapping in the permanent plots, catching many more individuals than in spring 2010. The new mammal trapping experiment in hedgerows, investigating how small mammals move around the landscape, got off to a great start, with 149 mammals found, including the first field vole (Microtus agrestis) found on the project. Although logistically challenging (four sites to visit!), the new experiments are going well.
Between July and September 2011 there were unusually high numbers of bank voles found in all sites, which matches findings across the UK. Conversely, wood mice numbers were lower than usual, which could possibly be due to competition between wood mice and bank voles for resources within their habitats. As well as the long term trapping in the forest edges, a new trapping survey has begun, to see how small mammals use hedgerows to move about the landscape. Members of the British Entomological and Natural History Society benefited from Dr Eleanor Slade’s knowledge of moths at a moth trapping night in July. Eleanor and colleagues are now busy analysing data collected over the last years, and preparing papers for publication.
Data quality
Using volunteers means that data are collected by a large number of people, all learning new techniques. Assessing data quality is therefore essential. To this end, a number of trees are regularly re-measured to understand the amount of error in measurements. An example of one set of re-measurements is shown in fig 6. With the exception of one clear error, 1st and 2nd measurements were very similar to each other, with most measurements being within 1 mm of each other, and none more than 4 mm different.
Figure 6 Measurement uncertainty in tree diameter