Deadwood in managed and protected forest in southern Sweden : in the wake of storm

Linköping University | Department of Physics, Chemistry and Biology Bachelor thesis, 14 hp | Biology programme: Physics, Chemistry and Biology Spring term 2019 | LITH-IFM-G-EX--19/3701--SE

Deadwood in managed and

protected forest in southern

Sweden: in the wake of storm

Paula Jonsson

Examinator, Anders Hargeby, IFM Biologi, Linköpings universitet Supervisor, Per Milberg, IFM Biologi, Linköpings universitet

Department of Physics, Chemistry and Biology Linköping University

URL för elektronisk version

ISBN

ISRN: LITH-IFM-G-EX--19/3701--SE

_________________________________________________________________

Serietitel och serienummer ISSN

Title of series, numbering ______________________________

Språk Language Svenska/Swedish Engelska/English ________________ Rapporttyp1 Report category Licentiatavhandling Examensarbete C-uppsats D-uppsats Övrig rapport _____________ Titel

Deadwood in managed and protected forest in southern Sweden: in the wake of storm

Författare

Paula Jonsson

Keyword/Nyckelord

Deadwood, production forest, protected forest, storm, volume

Sammanfattning

Abstract

Deadwood has increased over the last 25 years, but it remains unclear to what extent this is driven by forestry practices or storms. Therefore, I wanted to study the change in volume, decay stage and tree species during a 22-year period, to see if there was a correlation between increase of deadwood and storm. This study included data from southern Sweden, collected by the Swedish National Forest Inventory between 1994-2016. Deadwood in production forest have doubled over the last 25 years and almost quadrupled in protected forest. The increase does not depend on storm since much of the fallen wood was probably removed following year. In protected forest there was an increase in deadwood of broadleaved trees and a drastic decrease in Pinus sylvetstris. While in production forest, conifer trees dominate and there was no lasting effect due to the storm Gudrun (2005) on

Picea abies. Hard deadwood decreased in production forest, possibly due to increased removal of branches and

treetops, used as forest fuel in forest management. Possible reasons for the increase in deadwood could be the awareness in forestry, especially certification system and voluntarily set asides. Though, there is still necessary to increase the volume of deadwood in production forest, since it covers the largest parts of Swedish forests and does not seem to reach the national environment objective in 2030.

Content

1 Abstract... 1

2 Introduction... 1

3 Material and methods ... 3

3.1 Study area ... 3

3.2 Data collection ... 3

3.3 Data management ... 4

3.5 Statistical analyses ... 5

4 Results ... 5

5 Discussion ... 8

5.1 Conclusions ... 10

5.2 Social and ethical aspects ... 10

6 Acknowledgement ... 11

1

1 Abstract

Deadwood has increased over the last 25 years, but it remains unclear to what extent this is driven by forestry practices or storms. Therefore, I wanted to study the change in volume, decay stage and tree species during a 22-year period, to see if there was a correlation between increase of deadwood and storm. This study included data from southern Sweden, collected by the Swedish National Forest Inventory between 1994-2016. Deadwood in production forest had doubled over the last 25 years and almost quadrupled in protected forest. The increase did not depend on storm since much of the fallen wood was probably removed following year. In protected forest there was an increase in deadwood of broadleaved trees and a decrease in Pinus sylvetstris. While in production forest, conifer trees dominate and there was no lasting effect due to the storm Gudrun (2005) on Picea abies. Hard deadwood decreased in production forest, possibly due to increased removal of branches and treetops for fuel. Possible reasons for the increase in deadwood could be the awareness in forestry, especially certification system and voluntarily set asides. A further increase of the volume of deadwood in production forest is needed to achieve the national environment objective in 2030.

2 Introduction

Forest in Sweden, which covers a large part of the country, is an important part of the natural environment. This because many species are adapted to forest structures, such as deadwood. Out of 28 million hectare of forest, 23,5 million ha is used as

production forests (Swedish University of Agricultural Sciences 2017). Two percent of all protected areas are located in south of Sweden (Simonsson et al. 2016). Sweden has an intense forestry that include action such as thinning, pre commercial thinning and clearcut, where thinning is the most common action (Swedish University of Agricultural Sciences 2017). Because of intense forestry, old- growth forest with variation and different structures are rare (Stokland et al. 2012). In protected forests the average volume of deadwood is between 10 and 20 m3_{, depending on forest type}

there is an average of 5.3 m3 deadwood per ha in Svealand and 4.3 m3 per ha of deadwood in Götaland (Swedish University of Agricultural Sciences 2017). Over the last decades, there is an increase in the amount of deadwood (Swedish University of Agricultural Sciences 2017). The most likely factor for the increase could be the actions within forestry to improve biodiversity. In Sweden, there are two certifying system, Forest Stewardship Council (FSC) and Programme for the

Endorsement of Forest Certification (PEFC). These associations strive towards a more substantiable forest management and the certification systems are supposed to be positive for increased levels of deadwood (Simonsson et al. 2015). In forestry, tree retention for conservation have an impact on the increased amount of deadwood in young forests (Kruys et al. 2013). Another reason could be that landowners can voluntarily set asides forestland to promote biodiversity (Skogsstyrelsen 2017). Such voluntarily set asides provides structural factors, that is necessary and important for biodiversity (Simonsson et al. 2016). The third factor with likely impact on increase in deadwood is storm damages. One storm in particular, Gudrun, effected southern Sweden in January 2005, felling about 75 million m3 of trees (Skogsstyrelsen 2006). The presence of deadwood is an important factor in forest ecosystems. Deadwood includes standing dead trees (snags) and downed deadwood (logs).Since deadwood is a large part of biomass in natural forests, many of species are dependent on deadwood (Nilsson et al. 2002), so called saproxylic species (Stokland et al. 2012). Due to increased forest management over the last 150 years (Linder & Östlund 1998), the abundance of valuable qualities of deadwood has decreased (Stokland et al. 2012). Earlier studies have pointed to a lack of deadwood in large dimensions in Swedish forests (Fridman and Walheim 2000). Another observation by Jonsson and colleagues (2016) was that the increase of deadwood since first inventory (1994 until 2012), depends mostly of the storm in 2005 (Jonsson et al. 2016). All these studies including my own, are based on the same data, collected by Swedish National Forest Inventory (SNFI) from Swedish University of Agricultural Sciences.

In the present study, I wanted to investigate the increased amount of deadwood, during a study period of 22 years, where I focused on sample plots that have been revisited. In my study, given that during the study period the storm Gudrun was

3

present in 2005, my aim was to study how much the storm contributed to the increased levels of deadwood, maximum diameter, decay stage and tree species, in both managed production and protected forest. The same data was used by Widen (2017), who analysed decay of individual objects of deadwood.

3 Material and methods 3.1 Study area

The study included southern Sweden with a delimitation along Limes Norrlandicus and thereby included 15 out of 21 counties in Sweden. The forest in southern Sweden is heterogenous, with a mixture of conifers and deciduous (hemi-boreal and nemoral forest) trees. Götaland, which is included in this study contributes with the largest proportion of harvest in volume (Swedish University of Agricultural Sciences 2017). 3.2 Data collection

All data had been collected by Swedish National Forest Inventory (SNFI), a sample inventory of forest in Sweden, that has been ongoing since 1923. The inventory covers all types of forest, and are done in so called tracts, with both permanent and temporary tracts. The permanent tracs are visited every five years and the temporary tracs are only visited once. A tract is made up of a cluster with sample plots. The current study include region 4 (Counties of Stockholm, Södermanland, Uppsala, Västmanland, Örebro, Östergötland, Jönköping, Kronoberg, Kalmar, Västra Götaland (excluding the former Göteborgs and Bohuslän)) and region 5 (Counties of Gotland, Blekinge, Skåne, Halland and Göteborgs and Bohuslän). Tracts in region 4 are 800 m (side) and tracts in region 5 are 300 m (side). For further information regarding tracts and sample plots see Fridman et al. (2014). Deadwood has been recorded since 1994 and among the variables recorded, I have used volume, tree species identity and decay stage. When deadwood is being inventoried, decay stage is recorded in one of five different groups: 0 = raw wood, 1 = hard deadwood, where 90 % of bark cover still occur, 2 = slightly decomposed deadwood, 3 = decomposed deadwood and 4 = very decomposed deadwood.

3.3 Data management

In this study I focused on sample plots from the permanent tracts and arranged the data in two different ways. First, I grouped data according to time periods covered, and named them after the first year of inventory (Table 1). For this merge, I included tree species and decay stage (number of dead trees/ha)

Table 1. Groups of years with records on deadwood collected by SNFI.

Time period 1(1994) 2(2003) 3(2008) 4(2013) Years 1994-2002 2003-2007 2008-2012 2013-2016

I used these time periods, except for change in volume per ha. For this, I instead focused on three different cohorts (where, a cohort is a set of plots that have been revisited), to get a better precision in data (Table 2).

Table 2. Cohorts of plots that were visited four times by SNFI.

Earlier studies included all collected data from the deadwood inventory, while I was focusing on revisited sample plots. Furthermore, I also separated production forest and protected forest, i.e. forest that was under official protection in 2016 (but not always so when the first sampling occurred).

Data were handled in Excel. First, I divided the volume (m3) of deadwood, with the sample plot area (ha). A sample plot is 0.314 ha, or smaller if it was a divided plot (a sample plots can be divided into multiple subsplots if vegetation or management differs in a plot).

As most tree species were rare in data, I merged some species to allow meaningful analyses. Ulmus glabra, Fraxinus excelsior, Fagus sylvatica, Carpinus betulus, Prunus avium, Tilia cordata and Acer platanoides were merged to “broadleaved

Cohort Years

1 1997 2004 2009 2014

2 1998 2005 2010 2015

5

trees”. Populus tremula, Sorbus aucuparia, Alnus spp. and Sorbus intermedia were merges into ”other tree species”. Betula pubescens and Betula pendula were merged into Betula spp.

3.5 Statistical analyses

Estimates of volume of deadwood per ha were accompanied by 95 % confidence intervals. For the model predicted I merged all mean for each cohort and therefore used a linear model to calculate future increase. These predictions are based on production forest and protected forest.

4 Results

Deadwood had increased over time, both in production and protected forest and for all three cohorts (Figure 1). Deviating from the overall trend of steady increase was seen only in the second cohort where there was a clear peak in deadwood for both forest types in 2005 (Figure 1), i.e the year when Gudrun hit Sweden in January.

Figure 1. Volume of deadwood per ha (95 % confidence interval) in three cohorts of sample plots visited on four occasions.

An estimated prediction from my data analysis (Figure 1) with a linear model, shows that in the year of 2030 there would be 22.55 m3/ha in protected forest and 8.73 m3/ha in production forest.

Production forest: y = 0.1767x – 349.97 Protected forest: y = 0.5682x – 1130.9

As for tree species there was a decrease in number of deadwood of Betula spp. in production forest (Figure 2a), from first time period (1994) to the latest (2013). Number of P. abies increased over the last time period (Figure 2a). Another observation was that there was an increase in numbers of P. abies the year of 2005 (Gudrun), and a decrease the year after (Figure 4). Q. robur and broadleaved trees increased from second time period (2003) to last time period (2013).

In protected forest, P. sylvestris decreased from first to last time period (Figure 2b). For P. abies, Betula spp., Q. robur, broadleaved trees and other tree species, there was an increase in number in second time period (2003). P. abies increased from first to third time period (1994-2008), and a minor decrease in last time period (2013). Number of deadwood of broadleaved trees had increased from first cohort (1994) to last (2013).

7

Figure 2. Numbers of dead trees/ha at different time periods. Figure 2a shows number of dead trees in production forest. Figure 2b shows number of dead trees in protected forest. Broad-leaved deciduous forest was merged as broadleaved trees and all other inventoried tree species where merged as other hardwood.

In both production and protected forests, decay stage “1” dominated (Figure 3a, 3b). In both forests, decay stage “0” was not recorded at the first time period (1994). Variation in decomposition class were similar in production and protected forest.

0 2 4 6 8 10 12

Picea abies Pinus sylvestris Betula spp. Quercus robur Broadleaved trees Other tree species N u m b e r o f tre e s Tree species 1994 2003 2008 2013 2b 0 2 4 6 8 10 12 14

Picea abies Pinus sylvestris

Betula spp. Quercus robur Broadleaved trees Other tree species N u m b e r o f tre e s Tree species 1994 2003 2008 2013 2a

Figure 3. Number of dead trees/ha at different time periods. Different decay class are estimated over time measured in numbers/ha. Figure 3a shows numbers of dead trees in different decay stage in manage production forest. Figure 3b show numbers of dead trees in different decay stage in protected forest.

5 Discussion

Deadwood in production forest had doubled over the last 25 years and almost quadrupled in protected forest. Both forests had a clear peak in 2005, the year of the storm. On the other hand, there was no corresponding peak, in either forest type, in the cohort involving sampling in 2006. It was therefore likely that storm-felled wood was removed in production forest as well as in protected. Note that some areas were not yet protected during the storm damage, and that deadwood were removed from some protected areas for forest sanitation purposes. It might have expected that there should be some form of equilibrium in protected forest, but this was not the case. This could mean that protected areas will bear more dead wood in the future than today.

Possible reasons for the increase could be the adjustments in forestry, especially certification system and voluntarily set asides.

Storms in Sweden may contribute with a temporary increase of deadwood (volume and number). But it did not seems that it contributed to the increase for the last 22 years. The storm that felled the most trees and gave a visible peak in deadwood was Gudrun, January 2005. Other significant storms (Figure 1, Table 4) had no visible impact in data. This was probably because they generated much less timber than Gudrun (2005), and partly because no cohort data existed from the season following the storm. 0 4 8 12 16 20 0 1 2 3 4 N u m b e r o f tre e s /h a Decay stage 1994 2003 2008 2013 0 4 8 12 16 20 0 1 2 3 4 N u m b e r o f tre e s /h a Decay stage 1994 2003 2008 2013 3a 3b

9

Table 4. Storms in Sweden and affected area. Some of the storm were not visible until the year after. Information is given through Swedish University of Agricultural Sciences (2017) and Skogsstyrelsen (2005).

There is an ongoing trend that forest owners in southern Sweden avoid to plant pine, because of grazing damage, and therefore plant spruce instead (Skogsstyrelsen 2017). Since many of the sample plots in protected forest, probably belonged to previous production forest, this could be one reason for the reduction of numbers in P.

sylvestris in protected forest (Swedish University of Agricultural Sciences 2017). Another important observation was the large increase of broadleaved trees. One possible reason could be the diseases of U. glabra and F. excelsior. In production P.

abies and P. sylvestris dominated (Figure 3a), which was expected, since conifers

trees dominate in production forest (Larsson et al. 2011). One possible reason for decrease in numbers of P. abies, the year after Gudrun (2005), might be the fear of attack from Ips typographus, and therefore a great effort to remove spruce wood that was felled during that storm.

Figure 4. Numbers of dead trees/ha (Picea abies), in different years.

10 11 12 13 14 15 16 N u m b e r o f d e a d t re e s /h a Year

Storms in Sweden Affected area

2005 Gudrun (Jan) South of Sweden

2007 Per (Jan) South of Sweden

2011 Dagmar (Dec- inventoried in 2012) South of Sweden

2013 Sven (Feb), Simone (Oct- inventoried in 2014) South of Sweden 2015 Egon (Jan), Gorm (Nov- inventoried in 2016) South of Sweden

One possible reason for the decrease in decay stage ”1” in production forest, could be the removal of branches and treetops used as forest fuel (Skogsstyrelsen 2008). When removing these substrates, 20 % must be left behind, since these are valuable to organism (Skogsstyrelsen 2008).

The national environmental objective say that deadwood in production forest should increase to 10 m3 per ha, before 2030 (Naturvårdsverket 2018), and there should also be an average of 20 m3 _{in protected areas (Almstedt and De Jong 2005). According to}

my calculations, protected forest in southern Sweden had an average of 15.9 m3 per ha in 2016, and production forest 5.9 m3 per ha. Assuming a continued increase in volume protected forest would probably reach the national environment objective by 2030 (20 m3 _{per ha). The same fulfilment does not apply to production forest (10 m}3

per ha). For further studies it would be interesting to know more about what kind of deadwood there is, especially location of coarse deadwood and estimate how much that is needed for some species survival, especially threatened ones.

5.1 Conclusions

Overall, there has been an increase of deadwood over the last 20 years. Although the large storm event added deadwood, it was not the primary reason for the increase which instead is likely due to adjusted forestry practices. If the increase continues at its current rate, the national objective for the volume of deadwood in production forest will not be reached by 2030.

5.2 Social and ethical aspects

The data that The Swedish National Forest Inventory from Swedish University of Agricultural Sciences collect every year is a great source for further analysis. It is therefore important that we use these data for monitoring changes in boreal forest in Sweden as a result of forestry, since the forest and especially dead wood give rise to many valuable habitats. This study is built exclusively on already existing data and has no apparent impact on any ethical aspects.

11

6 Acknowledgement

I would like to give a huge thanks to my supervisor Per Milberg for helping me with my data and all of his guidance and help during this project. I would also like to thank Jonas Fridman who have provided me with all data collected by SNFI. Lastly, I would like to thank Andrey Höglund for helping me in R.

7 References

Almstedt M, De Jong J (2005) Död ved i levande skogar - Hur mycket behövs och hur kan målet nås? Naturvardsverket, Rapport 5413

Fridman J, Holm S, Nilsson M, Nilsson P, Ringvall A, Ståhl G (2014) Adapting National Forest Inventories to changing requirements - The case of the Swedish National Forest Inventory at the turn of the 20th century. Silva Fennica 48. 1–29 Fridman J, Walheim M (2000) Amount, structure, and dynamics of dead wood on managed forestland in Sweden. Forest Ecology and Management 131, 23–36

FSC Sweden (2010) Swedish FSC Standard for Forest Certification including SLIMF indicators - FSC-STD-SWE-02-02-2010 Sweden Natural, Plantations and SLIMF EN. FSC

Jonsson B.G, Ekstöm M, Esseen P, Grafström A, Ståhl G, Westerlund B (2016) Dead wood availability in managed Swedish forests – Policy outcomes and implications for biodiversity. Forest Ecology and Management 376, 174–182

Kruys N, Fridman J, Götmark F, Simonsson P, Gustafsson L (2013. Retaining trees for conservation at clearcutting has increased structural diversity in young Swedish production forests. Forest Ecology and Management 304, 312–321

Larsson A (red) 2011 Tillståndet i skogen – rödlistade arter i ett nordiskt perspektiv. ArtDatabanken Rapporterar 9. ArtDatabanken SLU Uppsala

Linder P, Östlund L (1998) Structural changes in three mid-boreal Swedish forest landscapes 1885-1996. Biological Conservation. 85, 9–19

Naturvårdsverket (2018) Miljömålen – Årlig uppföljning av Sveriges nationella miljömål 2018 – Med fokus på statliga insatser. Naturvårdsverket, Rapport 6833 Nilsson SG, Hedin J, Niklasson M (2002) Biodiversity and its assessment in boreal and nemoral forests. Scandinavian Journal of Forest Research 16, 10–26

Simonsson P, Gustafsson L, Östlund L (2015) Retention forestry in Sweden: driving forces, debate and implementation 1968–2003. Scandinavian Journal of Forest Research 30, 154–173

Simonsson P, Gustafsson L, Östlund L (2016) Conservation values of certified-driven voluntary forest set-asides. Forest Ecology and Management 375, 249–258

Skogsstyrelsen (2005) Sammanställning av stormskador på skog i Sverige under de senaste 210 åren. Rapport 9/2005

13

Skogsstyrelsen (2008) Rekommendationer vid uttag av avverkningsrester och askåterföring. Meddelande 2/2008

Skogsstyrelsen (2017) Friviliga avsättningar. Meddelande 4/2017 Skogsstyrelsen (2017) Projektplan Mera Tall. Dnr 2016/1145

Stokland JN (2012) Dead wood and substainable forest management. pp 302–331, Threatened saproxylic species. pp 356–361 in: Stokland JN, Siitonen J, Jonsson BG (eds) Biodiversity in Dead Wood. Ecology, Biodiversity and Conservation.

Cambridge University Press, NY

Swedish University of Agricultural Sciences (2017) Forest statistics 2017. Official statistics of Sweden, Umeå