- Email: [email protected]
Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden
Forest Ecology and Management xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Forest Ecology and Management journal homepage: www.elsevier.com/locate/foreco
Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden Sofie Hellsten a,⇑, Johanna Stadmark a, Gunilla Pihl Karlsson a, Per Erik Karlsson a, Cecilia Akselsson b a b
IVL Swedish Environmental Research Institute, Box 53021, SE-400 14 Gothenburg, Sweden Lund University, Department of Physical Geography and Ecosystem Science, Sölvegatan 12, SE-223 62 Lund, Sweden
a r t i c l e
i n f o
Article history: Received 19 April 2015 Received in revised form 2 July 2015 Accepted 10 July 2015 Available online xxxx Keywords: Storm felling Soil water Nitrate Nitrogen leakage Eutrophication SWETHRO
a b s t r a c t In January 2005, south-west Sweden was hit by a severe storm that caused large damage to the forests through massive windthrow. The aim of this study was to assess the effect of this windthrow on nitrate concentrations in the soil water below the root zone on 33 forest monitoring plots within the Swedish Throughfall Monitoring Network (SWETHRO). These sites were damaged to different extents by the storm. The analysis showed increased levels of nitrate concentrations in the soil water as a consequence of storm damaged forest. The present study concerned forest ecosystems with relatively low levels of nitrogen deposition, as compared with forests analysed in previous studies. The maximum soil water nitrate concentrations occurred 1–4 years after the storm. After 5–6 years, the concentrations were back at the same levels as before the storm event. It was not possible to demonstrate a significant correlation between increased nitrate concentrations in the soil water after the storm and the level of nitrogen deposition at the site. The potential impact on ground- and surface waters due to elevated nitrate concentrations in soil water after storm events is discussed. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction The frequency and intensity of natural disturbances, such as storm events, has increased in many forest ecosystems around the world (Schelhaas et al., 2003). During the last century damage due to storms in Swedish forests has increased (Nilsson et al., 2004). The main driving forces are expected to be a combination of climate change and changes in forest cover, tree species composition as well as changes in management regimes (Nilsson et al., 2004; Seidl et al., 2011; Usbeck et al., 2010). In Sweden, storm statistics have not shown clear indications of increased frequency and intensity of storms (Blennow et al., 2010). However, climate change may increase the probability of windthrow also due to a reduced resilience of forest ecosystems to wind damage. Finnish studies have shown that warmer winters in areas where ground frost occur increase the windthrow risk of trees at northern latitudes through reduced stability (Gregow et al., 2011; Peltola et al., 1999). In January 2005, south-west Sweden was hit by a severe storm with wind speeds up to 42 m s1 (Swedish Forest Agency, 2006). The storm caused large damage through massive windthrow, corresponding to 110% of the average annual harvest rates
⇑ Corresponding author. E-mail address: sofi[email protected] (S. Hellsten).
(1998–2004) in Sweden derived from only 16% of the country’s forest area (Seidl and Blennow, 2012). The drastic increase in deforested areas following the storm could be expected to result in increased levels of nitrate leaching, which may potentially contaminate groundwater and cause eutrophication of surface waters (Galloway et al., 2004; Gundersen et al., 2006; Kreutzweiser et al., 2008; Smith, 2003; Smith and Schindler, 2009). Furthermore, nitrate leaching also contributes to soil acidification, which causes potentially toxic metals to be released in soil and surface water (Reuss and Johnsson, 1986). Leaching of inorganic nitrogen (primarily nitrate) to the soil water increases as the nitrogen uptake to trees is disrupted. Disturbances in forest ecosystems, whether they are natural (e.g. storm events, insect attacks), or anthropogenic (forestry practices, such as clear-cut logging), can affect soil conditions and nutrient processes within the soil resulting in nutrient losses. Several studies carried out in boreal forests indicate that logging affects the retention of inorganic nitrogen in the forest soil, so that nitrate leaching in the runoff water increases (e.g. Emmet et al., 1991b; Futter et al., 2010; Gundersen et al., 2006; Kreutzweiser et al., 2008; Pardo et al., 1995; Rosén et al., 1996). The soil conditions of clear-cut areas are warmer and moister than in growing forests, due to decreased transpiration and increased sun exposure when the trees have been removed (Kreutzweiser et al., 2008). Moist, warm soil conditions are favourable for decomposition of organic material resulting in nitrate
http://dx.doi.org/10.1016/j.foreco.2015.07.009 0378-1127/Ó 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: Hellsten, S., et al. Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden. Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.009
2
S. Hellsten et al. / Forest Ecology and Management xxx (2015) xxx–xxx
formation after nitrification. Furthermore, most of the uptake of nitrogen by vegetation ceases in a clear-cut area, and the changed hydrological conditions (increased runoff and raised groundwater level) may further increase leaching of nitrate (Rosén et al., 1996; Sørensen et al., 2009). Increased nitrate leaching in runoff water occurs during a time period of three to six years in the south of Sweden, and up to 15 years in the north (Akselsson et al., 2004; Futter et al., 2010; Rosén et al., 1996; Wiklander et al., 1991). The Swedish Throughfall Monitoring Network, SWETHRO, is an environmental monitoring network that measures air concentrations of pollutants, deposition and soil water chemistry at forest sites in Sweden (Pihl Karlsson et al., 2011). Many of these sites were damaged to different extents due to the storm in January 2005. In most cases there existed long time-series of measurements of soil water chemistry prior to the storm. Hence, this provided a unique opportunity to investigate the relations between the extent of damage to the forests and the change in the soil water chemistry during the following years. The plots within SWETHRO are not generally positioned within small well-defined catchments, so it was not possible to directly relate the impacts on soil water chemistry to changes in the chemistry of the surface water runoff. The aim of this study was to investigate the effect of the different extents of windthrow on nitrate (NO3–N) concentrations in soil water in forests. The study focused on the soil water concentrations of nitrate since this is the main inorganic nitrogen compound in soil water that is affected by logging, and the effect on ammonium is small (Gundersen et al., 2006; Lepistö et al., 1995). The effect of logging on organic nitrogen is also small (Lepistö et al., 1995) and was therefore not considered. 2. Materials and methods 2.1. Sites The Swedish Throughfall Monitoring Network (SWETHRO) has been operated since 1985 (Pihl Karlsson et al., 2011). The monitoring network measures air concentrations of pollutants, deposition and soil water chemistry in forest ecosystems in Sweden. Monitoring plots (30 30 m) are positioned in closed, mature, managed forests with no major roads or other pollution sources in the vicinity. Currently SWETHRO includes measurements of soil water chemistry on 62 sites in Sweden, and the longest time series are 30 years (2015). Many of these sites, in southern Sweden, were damaged to different extents due to the storm Gudrun in 2005. Sites with short time series (<10 years) and sites with vague information about the extent of the storm damage were not included in the study and only coniferous sites were considered. In total 33 sites in the south of Sweden with Norway spruce and/or Scots pine forests were included in the study and their geographical positions are shown in Fig. 1. The 33 sites were divided into four different damage classes, depending on the extent of the storm damage (Table 1). As expected, the most damaged plots were all covered with Norway spruce. About 45 SWETHRO sites are co-located with observation plots operated by the Swedish Forest Agency. For these plots, information is available for changes over time regarding vegetation cover, crown thinning, stem diameter growth, needle chemistry and soil chemistry.
at 50 cm depth in the mineral soil. Based on a soil sampling campaign made in 2010–2011, the depth to the beginning of the undisturbed mineral soil (C-layer) was estimated to 40–60 cm for the plots included in this study. Hence, a 50 cm soil depth would be in the lower part of the B-layer or in the upper part of the C-layer. In general, five lysimeters were installed on each forest site, inside the canopy close to the throughfall collectors. However, some sites had fewer lysimeters due to difficulties with soil stoniness. Lysimeters were sampled by applying tension over a two day period. The water from the lysimeters was then combined into one composite sample for analysis. Only water samples containing >50 ml of water were included in the study, to eliminate uncertainties due to small sample volumes. Throughfall (TF) and bulk (BD) deposition were measured monthly throughout the years. TF measurements were carried out at all sites during the entire period analysed, whilst BD measurements were carried out during different time periods at different sites. Further details about deposition measurements and chemical analysis can be found in Pihl Karlsson et al. (2011). 2.3. Estimates of nitrogen deposition A considerable fraction of the atmospheric nitrogen deposition reaching the forests is taken up directly by the tree canopies (Adriaenssens et al., 2012; Eugster and Haeni, 2013; Ferm, 1993; Ferm and Hultberg, 1999). Hence, throughfall measurements cannot be used directly to estimate the total nitrogen deposition to forests. Therefore, bulk measurements, as well as throughfall measurements, were used in this study. Bulk deposition measurements mainly consist of wet deposition but there is also a small fraction of dry deposition to the samplers (Draaijers et al., 1996; Dämmgen et al., 2005). Information on bulk deposition was not available for all years and sites analysed in this study. Bulk deposition of inorganic nitrogen varies geographically across Sweden in a consistent pattern (Fig. 2, Karlsson et al., 2011), based on the geographical position expressed as the sum of latitude and longitude. Based on the relation in Fig. 2, the bulk deposition was estimated for the sites with moderate and large damage for the six-year period 1999–2004, prior to the occurrence of the storm (Table 3). The fraction of dry deposition contributing to the total nitrogen deposition declines towards northeast similarly to the decline of bulk deposition (Karlsson et al., 2011). Hence, it may be assumed that the relative difference regarding the estimated nitrogen deposition between the different sites in Table 3 is correct. 2.4. Weather data Official interpolated weather data can be obtained for any location in Sweden from the Swedish Meteorological and Hydrological Institute (SMHI), regarding air temperatures and precipitation (http://luftwebb.smhi.se/). Data for annual mean air temperature and annual precipitation were obtained for all sites. Differences in annual mean air temperatures and annual precipitation were analysed for the two time periods before and after the storm, 2000–2004 and 2005–2009 (see below) for all sites classified with moderate or large damage.
2.2. Measurements
2.5. Data analysis
Soil water below the root zone was sampled three times a year, to represent the conditions before (Feb–May), during (Jun–Sep) and after the vegetation period (Sep–Dec). Soil water samples were obtained using suction lysimeters with ceramic cups (P 80), placed
A database was constructed with information about the level of damage caused by the storm at the different sites, Table 2. Data on storm damage is mainly based on notes from field protocols and contacts with field personnel collecting the samples.
Please cite this article in press as: Hellsten, S., et al. Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden. Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.009
S. Hellsten et al. / Forest Ecology and Management xxx (2015) xxx–xxx
3
Fig. 1. The 33 sites included in the study, divided into four damage classes.
Table 1 Four damage classes, depending on level of storm damage. Damage class
Level of damage
Number of sites
1 2 3
No damage <10 trees within the site fell At least 10 trees, and up to 50% of the trees fell >50% of the trees fell
15 sites 9 sites 5 sites
4
No damage Small damage Moderate damage Large damage
4 sites
16
kg of N ha-1 yr -1
14 12
y=0.8733x 2 -126.88x+4614.6 R²=0.8733
10 8 6 4 2 0 69
70 71 72 73 Geographical position (Long+Lat)
74
Fig. 2. The correlation between the annual bulk deposition of inorganic nitrogen (NO3 + NH4) and the geographical position of the monitoring site expressed as the sum of longitude and latitude.
The database was complemented with information on soil solution chemistry (nitrate concentration and pH), nitrogen deposition (bulk deposition, estimated bulk deposition and throughfall
measurements), stand properties (stand age), climate data (temperature, precipitation) and other disturbances (e.g. damage from forestry machinery), based on field protocols and information from field personnel of the SWETHRO Network. The information regarding other disturbances has not been systematically applied in this study, but is valuable when looking at the data for each individual site, e.g. explaining potential storm effects at a ‘‘no damage site’’. Soil solution chemistry, nitrogen deposition and stand properties were derived from the database of the SWETHRO Network (Pihl Karlsson et al., 2011). Estimated annual bulk deposition of inorganic nitrogen was also included (see Section 2.3). Information about soil chemistry and ground vegetation was obtained from the Swedish Forest Agency observation plots (OBS-plots). Many new plots were initiated within SWETHRO during 1996, and many of these plots were also OBS-plots. The OBS-plots were sampled for soil chemistry in different soil layers in 1996. Over time, common SWETHRO and OBS-plots needed replacement after harvesting etc., but soil chemistry was not always sampled at the new plots. As a result, there are plots within SWETHRO with and without soil chemistry sampling. Of the 33 sites included in this study, information on soil chemistry for different soil layers in the year 1996 is available for 31 sites, however, not for two of the most damaged sites, Alandsryd and Knapanäs. Data for the C/N ratio in the humus layer were used for analyses of correlations with the increases of the soil water NO3–N concentrations before and after the storm. The 33 sites were divided into four different damage classes, depending on the extent of the storm damage (Table 1). Within each damage class, nitrate leaching and the correlations with other parameters such as deposition, C/N ratio, stand age and pH in soil water were investigated. For each of the sites, the average soil water nitrate concentration was calculated during two 5-year time periods, to represent
Please cite this article in press as: Hellsten, S., et al. Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden. Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.009
4
Damage class
Site
Extent of storm damage
Location
Tree species
No damage
Arlanda (A 92)
No storm damage
5.5
536
No storm damage
Norway spruce Scots pine
70
Attsjö (G 21)
85
6.1
665
Bergby (A 01)
No storm damage
Scots pine
75
5.5
558
Edeby (D 11)
No storm damage
5.6
552
No storm damage
105
5.5
583
Greckssundet (T 02)
No storm damage
57
4.3
794
Hensbacka (O 35)
No storm damage
86
6.5
873
Hjärtsjömåla (K 03)
No storm damage
Norway spruce Norway spruce Norway spruce Norway spruce Scots pine
75
Farstanäs (A 35)
70
6.2
699
Knutsta (D 14)
No storm damage
Scots pine
72
5.7
590
Marydn (L 15)
No storm damage
7.3
766
68
6.2
660
Solltorp (E 21)
No storm damage, except a broken treetop. No storm damage
Norway spruce Scots pine
46
Risebo (H 21)
5.8
610
No storm damage
74
6.3
599
Västra Torup 2 (L 07)
No storm damage
62
6.9
840
Örlingen (T 03)
No storm damage
Norway spruce Norway spruce Norway spruce Scots pine
70
Södra Averstad (S 05)
59°400 N, 17°580 E 56°540 N, 15°70 E 59°340 N, 18°40 E 58°570 N, 16°590 E 59°60 N, 17°380 E 59°350 N, 14°430 E 58°260 N, 11°440 E 56°210 N, 14°590 E 59°190 N, 16°70 E 55°370 N, 14°50 E 58°40 N, 16°70 E 58°90 N, 15°260 E 59°10 N, 13°70 E 56°80 N, 13°310 E 59°530 N, 14°260 E
59
4.3
795
Arkelstorp (L 05)
Norway spruce Scots pine
7.3
639
57
5.2
881
Höka (E 22)
One pine fell.
Scots pine
70
5.2
716
Sticklinge (A 05)
6.6
497
64
6.6
540
Söstared (N 01)
One pine fell.
Norway spruce Norway spruce Scots pine
99
Stora Ek (R 09)
One big spruce fell in the middle of the site. One tree fell.
82
6.9
869
Vång (K 13)
A few trees fell.
74
7
575
Värnvik (F 12)
2–3 spruces fell.
52
5.4
642
Åboland (O 01)
4 trees fell (2 spruces and 2 pines).
56°110 N, 14°150 E 57°460 N, 13°450 E 58°460 N, 15°90 E 59°230 N, 18°70 E 58°380 N, 13°470 E 57°310 N, 12°150 E 56°160 N, 15°270 E 57°500 N, 14°240 E 58°320 N, 11°440 E
50
Humlered (P 93)
2–3 trees fell in the monitoring site. One pine fell at the site.
58
6.1
900
A few trees also fell in close proximity to the site.
Angelstad (G 23)
About 35% of the trees fell.
6
810
The storm removed the trees north-east of the site.
10–12 spruce trees fell in, or into, the monitoring site.
Norway spruce Norway spruce
65
Bordsjö (F 22)
56°500 N, 13°450 E 57°500 N, 14°600 E
53
5.2
597
Small damage
Moderate damage
Norway spruce Norway spruce
Stand age (yr)a
Temp (°C)b
Prec. Other disturbances (mm)c
Many trees in close proximity to the site fell.
A few trees in close proximity to the site fell.
One tree outside the monitoring site fell into the site area.
A few trees in close proximity to the site fell. Damage from forestry machinery.
A few trees also fell in close proximity to the site.
Four trees fell outside the monitoring site showing a gap in the corner of the site. Forestry machinery have accessed the site.
An area 200 m north west of the site was badly damaged.
S. Hellsten et al. / Forest Ecology and Management xxx (2015) xxx–xxx
Please cite this article in press as: Hellsten, S., et al. Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden. Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.009
Table 2 The 33 sites from SWETHRO included in the study, divided into different damage classes.
c
Timrilt (N 13)
Stand age at the time of the storm, i.e. year 2005. Annual mean temperature, derived from the closest SMHI monitoring site (Swedish Meteorological and Hydrological Institute). Annual sum of precipitation, derived from the closest SMHI monitoring site (Swedish Meteorological and Hydrological Institute). a
b
6.4 50
1143
5.6
57°40 N, 14°230 E More than 50% of the trees fell. 56°470 N, 13°90 E Tagel (G 22)
About 50% of the trees fell.
Almost all trees fell. Knapanäs (G 09)
Norway spruce Norway spruce
80
707
The site was moved 10–20 m and two new lysimeters were installed in summer 2006 (three of the original lysimeters were left at the original site).The site was further damaged in a storm 2007. Damage from forestry machinery. The site was moved 2007 due to Bark Beetles. 6.3 51
593
5.5
Norway spruce Norway spruce All trees fell. Alandsryd (F 09) Large damage
13 spruce trees fell.
Mellby (F 18)
Vallåsen (N 17)
57°120 N, 13°590 E 56°410 N, 15°90 E
79
719
Gaps have been created in close proximity to the site, and forest edges are closer. Damage from forestry machinery. 7.3 68
1046
Many trees in close proximity to the site fell. 866 6
Norway spruce Norway spruce 57°90 N, 13°360 E 56°220 N, 13°60 E
51
1053 7 Norway spruce
15 spruce trees fell, showing a gap in the monitoring site. 15 spruce trees fell.
56°570 N, 12°430 E
67
3. Results
Tree species
Borgared (N 12)
5
the conditions before the storm (2000–2004), and after the storm (2005–2009). The nitrate response in soil water as an effect of the storm damage was compared between damage classes with ANOVA and t-test. All statistical analyses were made with the statistical tools available in Microsoft Excel.
Location Extent of storm damage Site Damage class
Table 2 (continued)
Stand age (yr)a
Temp (°C)b
Prec. Other disturbances (mm)c
S. Hellsten et al. / Forest Ecology and Management xxx (2015) xxx–xxx
The results in Fig. 3 show the nitrate (NO3–N) and ammonium (NH4–N) concentrations in soil water, 1997–2013 for the four sites most severely damaged by the storm. It can be seen that there were differences between the sites regarding the timing of the onset of the elevated nitrate concentrations. However, at all sites, except Tagel, there was a clear indication of increasing nitrate concentrations in the soil water before the end of 2007. Ammonium concentrations were generally low, however, at Timrilt two high ammonium values were recorded in 2008 (8.5 and 13.2 mg l1). This indicates that the storm event may have had an effect also on rates of N mineralization, at least at Timrilt. The average nitrate and ammonium concentrations before (2000–2004) and after (2005–2009) the storm are shown in Table 4 for the different damage classes. Fig. 4 shows the difference in soil water nitrate concentrations (NO3–N) in soil water between the two time periods. The areas with no or small damage have lower nitrate leaching after the storm Gudrun than the areas with moderate and large damage (p < 0.01, two-sampled t-tests assuming unequal variances). Due to the low number of plots with moderate and large damage, no statistically significant difference between these two damage classes regarding soil water nitrate concentrations after the storm could be demonstrated. The observed and estimated inorganic nitrogen (NO3 + NH4) deposition (bulk, estimated bulk and throughfall), pH in soil water (before the storm) and nitrate in soil water (before and after the storm) for the 9 sites within damage classes’ moderate and large damage are shown in Table 3. Note that the time period with observed values does not always overlap with the period with estimated values. Fig. 5A–C shows the difference in soil nitrate concentrations before and after the storm for the moderate and large damage classes plotted against the observed and estimated bulk inorganic nitrogen deposition as well as against the observed throughfall deposition. A regression analysis showed no statistically significant correlation between difference in NO3 concentration in soil solution and the measured bulk deposition, with the estimated bulk deposition or with the throughfall deposition, neither for the sites with large damage nor for the sites with moderate damage. A plot of the relation between the difference in soil water concentration of NO3 after and before the storm and the soil C/N ratio in the humus layer is shown in Fig. 6 for the sites classified with moderate and large damage. Unfortunately, information on soil chemistry was not available for two of the most damaged sites, Alandsryd and Knapanäs. No statistically significant correlation was found between the difference in soil water concentration of NO3 after and before the storm and the soil C/N ratio in the humus layer. However, it is interesting to note that the C/N ratio in the humus layer was high (34.9) at the site Tagel, which was severely damaged by the storm but did not show an increase in the soil water concentration of NO3 after and before the storm. Furthermore, neither stand age nor soil water pH were correlated with nitrogen leaching (data not shown). Differences in the weather conditions during the two time periods 2000–2004 and 2005–2009 were analysed based on weather data obtained from the Swedish Meteorological and Hydrological Institute (SMHI), regarding air temperatures and precipitation. The analysis was made for all sites classified with moderate or
Please cite this article in press as: Hellsten, S., et al. Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden. Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.009
6
S. Hellsten et al. / Forest Ecology and Management xxx (2015) xxx–xxx
(A)
(B)
(C)
(D)
Fig. 3. Nitrate – (NO3–N) concentrations in soil water 1997–2013 in the four sites with the highest level of windthrow (>50%): (A) Alandsryd, (B) Knapanäs, (C) Tagel and (D) Timrilt. Note that the monitoring site at Tagel was moved in the autumn of 2007 due to a Bark Beetle attack and that Knapanäs was moved 10–20 m after the storm and two new lysimeters were installed during summer 2006, whilst three original lysimeters were left at the original site. The Knapanäs site was further damaged in a storm in 2007. Measurements were terminated in 2010 at Knapanäs and 2012 at Alandsryd. Timrilt has an ammonium value of 13.2 mg l1 in April 2008, which exceeds the graph. The vertical dashed lines indicate the storm event.
large damage. There were only small differences between the periods. The air temperatures were on average 0.3 °C warmer and precipitation 3% higher for the 2005–2009 as compared with the 2000–2004 period.
4. Discussion 4.1. Variation in nitrate leaching with storm damage The current study showed increased levels of nitrate concentrations in soil water as a consequence of storm damaged forest (Figs. 3 and 4 and Table 4), in agreement with the studies of Ritter (2004) and Legout et al. (2009). However, the present study concerned forest ecosystems in southern Sweden with relatively lower levels of nitrogen deposition, as compared with the forests analysed in the previous studies. At three of the most severely damaged sites (Alandsryd, Knapanäs och Timrilt) the nitrate concentration in soil water reached maximum concentrations, 3–8 mg l1, during 2006–2009, i.e. 1–4 years after the storm. After 5–6 years, the levels were almost back to the levels before the storm event. These observations are in agreement with studies of the effects of clear-cut harvests on soil water nitrate concentrations. Measurements of nitrate concentrations in runoff water in southern Sweden have shown that elevated levels of inorganic nitrogen normally occurs during 5 years after logging and that the level from clear-cuts culminates after two or three years (Futter et al., 2010; Westling et al., 2004). Elevated nitrate concentrations can, both in the case of windthrow and in the case of clear-cut harvests, be explained by diminished tree demand for nitrogen and/or an increased
Fig. 4. The difference in soil water nitrate concentrations (NO3–N) in soil water between the time period after the storm (2005–2009) and the time period before the storm (2000–2004). The diagram shows the average value ± standard error for the different damage classes; no damage (15 sites), small damage (9 sites), moderate damage (5 sites) and large damage (4 sites). The sites with moderate and large damages had higher increase in nitrate concentrations than the sites with no or small damage (p < 0.01, two-sampled t-test assuming unequal variances).
mineralization in the soil due to increased oxygen supply and warmer soil conditions as the soil is more exposed to sun radiation, due to the removal of trees. However, this study also showed examples of damaged sites without any evident effects on soil water nitrate concentration, e.g. at the site Tagel (Fig. 3C), which was severely damaged by the storm. In the year 2000 the ground vegetation at Tagel consisted of a bottom layer (bryophytes and lichens) that covered 70% of the area and a field layer (trees and shrubs < 1 m, together with
Please cite this article in press as: Hellsten, S., et al. Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden. Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.009
0.53 0.014 0.47 0.83 3.52 2.01 1.51 0.021 2.93
0.53 0.007 0.46 0.82 1.58 2.00 1.51 0.019 2.73
other vascular plants) that covered 2%. That the extent of ground vegetation may be an important factor affecting the nitrate concentrations in soil water following a storm event was suggested by Mellert et al. (1998), who noted a strong negative correlation (r2 = 0.7) between vegetation cover and nitrate concentrations in soil water at 13 plots cleared after windthrow in Bavaria, Germany. Ritter (2004) came to the same conclusion following a windthrow study in Denmark, i.e. that nitrate concentrations in soil solution were correlated with ground vegetation cover. Other studies have illustrated the importance of N uptake by vegetation after logging, both regarding the magnitude and the duration of elevated nitrate concentrations (e.g. Emmet et al., 1991a,b; Nilsson and Örlander, 1999; Weis et al., 2001; Örlander et al., 1996). There was a significantly higher nitrate leaching at sites in the two classes with moderate or large damage, compared with the sites with no or small damage due to the storm (p < 0.01, Fig. 4). Due to the limited number of plots, it was not possible to distinguish differences in the level of nitrate leakage between the classes moderate and large damage. A majority of the sites classified as storm damaged showed elevated levels of nitrate. However some sites within ‘‘no damage’’ class also had increased levels of nitrate in soil water after the storm. This may indicate that the storm had an effect, although the damage was not recorded or obvious, due to root disturbance. On the other hand, within the ‘‘small damage’’ class it is likely that the storm damage in some cases was so modest that the effect of other factors was more important to nitrate levels in soil water, compared with the storm damage itself. Wilpert et al. (2000) and Ritter et al. (2005) showed that removal of trees in small gaps (<0.1 ha), can affect nitrate leaching. In contrast, Prescott et al. (2003) noted no effect on soil nitrate concentrations at removal of single trees. One explanation to the different results in these two studies could be the influence of nitrogen deposition or site quality. Prescott’s study was carried out in a low nitrogen deposition area compared with Wilpert’s study. Hence, windthrow of single trees could potentially have an effect on nitrate leaching in high deposition areas.
4.6 4.7 4.8 4.7 4.3 4.7 4.7 5.2 4.6 4.6 2.1 11.4 3.4 18.2 3.7 3.9 4.3 13.0
0.001 0.021 0.012 0.006 1.94 0.012 0.002 0.002 0.12
Diff NO3–N (mg l1) Soil water NO3–N (after) (05-09) mg l1 Soil water NO3–N (before) (00-04) mg l1 Soil water pH (00-04) mg l1 TF dep. (99-04) kg ha1 yr1
7
a
Not all years represented, different years for different sites.
9.8 6.1 13.3 9.2 14.9 9.3 6.6 7.3 12.5 (99-01) (99-01) (99-01) (99-00) (99-00) (99-00) (99-00) (99-01,04) (99-04) 10.0 8.2 14.1 9.6 14.9 9.2 10.9 8.0 14.1 70.58 72.83 69.76 70.74 69.46 70.71 71.83 71.44 69.93 Angelstad (G 23) Bordsjö (F 22) Borgared (N 12) Mellby (F 18) Vallåsen (N 17) Alandsryd (F 09) Knapanäs (G 09) Tagel (G 22) Timrilt (N 13)
Observed bulk dep. (99-04)a kg ha-1 yr1
Estimated bulk dep. (99-04) kg ha1 yr1
4.2. Variation in leaching with nitrogen deposition
Long + lat
Table 3 Inorganic nitrogen deposition (observed bulk deposition, estimated bulk deposition and throughfall), soil water pH before the storm and soil water nitrate concentrations (before and after the storm and the difference between the two) (data from SWETHRO). For the observed bulk deposition values, the time period of observations is given within brackets. TF dep. = throughfall deposition.
S. Hellsten et al. / Forest Ecology and Management xxx (2015) xxx–xxx
Previous studies have shown a correlation between the export of nitrate after clear-cut harvesting and the level of the inorganic nitrogen deposition preceding the harvest occasion, since the mineralization of nitrogen increases with increasing nitrogen availability (Akselsson et al., 2004; Lepistö et al., 1995). However, in this study it was not possible to demonstrate a significant correlation between increased nitrate concentrations in the soil water and the level of nitrogen deposition (Fig. 5A–C). The range of bulk nitrogen deposition levels for the sites analysed in this study was however limited. Timrilt has the highest nitrate and ammonium peak after the storm, and this site is located in an area with relatively high nitrogen deposition. Prior to the storm, partly elevated nitrate concentrations in the soil water at Timrilt indicated that this site was approaching nitrogen saturation (Fig. 3D). Most of these elevated nitrate concentrations occurred during spring, before the vegetation period, hence pointing to the importance of vegetation and growing season for nitrate recycling. Nitrate concentration in the soil water at Timrilt is still at a higher level than before the storm, and show seasonal fluctuation with the highest values outside the growing season (spring and autumn). 4.3. Effects on ground- and surface waters Elevated soil water nitrate concentrations due to storm events can be important at a local scale, and contribute to groundwater
Please cite this article in press as: Hellsten, S., et al. Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden. Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.009
8
S. Hellsten et al. / Forest Ecology and Management xxx (2015) xxx–xxx
(B)
(A)
(C)
Fig. 5. The difference in soil water NO3–N concentrations before (2000–2004) and after (2005–2009) the storm, plotted against inorganic nitrogen deposition. (A) Observed bulk deposition, (B) estimated bulk deposition and (C) throughfall deposition. No correlations were statistically significant using linear regression analysis.
Fig. 6. The difference in soil water NO3–N concentrations before (2000–2004) and after (2005–2009) the storm, plotted against the C/N ratio in the humus layer, obtained from soil sampling in 1996 within the Forest Agency monitoring network. No correlations were statistically significant using linear regression analysis.
contamination. However, it is important to evaluate the effect in relation to the area affected. If storm damage only affects a few percent of a forest area, and only single trees here and there, effects on surface water quality are expected to be minimal. However, if the storm disturbance is severe, with large areas affected by windthrow, the effect could be of significance. Although nitrate leaching from Swedish forests in general is small (Futter et al., 2010), leaching can be more important in the future if forest damage from windthrow increases. Normally, the annual proportion of new clear-cuts in southern Sweden is limited to approximately 1% of the productive forest area (Swedish Forestry Agency, 2014). However, due to the storm Gudrun in 2005, the areal extent of clear-cut areas increased 10-fold during a short time period in the south of Sweden. Furthermore, the damage may be even greater in windthrow areas,
Table 4 Average values for the nitrate concentration (NO3–N, mg l1) and ammonium concentration (NH4–N, mg l1) in soil water for the four damage classes before the storm (2000– 2004) and after the storm (2005–2009). Damage class (number of sites)
Before the storm (2000–2004)
After the storm (2005–2009)
Diff.
NO3–N (mg l1)
No damage (15) Small damage (9) Moderate damage (5) Large damage (4)
0.03 0.28 0.40 0.05
0.07 0.19 1.09 1.61
0.04 0.09 0.69 1.56
NH4–N (mg l1)
No damage (15) Small damage (9) Moderate damage (5) Large damage (4)
0.06 0.22 0.08 0.03
0.04 0.38 0.15 0.43
0.02 0.16 0.07 0.40
Please cite this article in press as: Hellsten, S., et al. Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden. Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.009
S. Hellsten et al. / Forest Ecology and Management xxx (2015) xxx–xxx
compared with conventional logging due to the clearing up operations. Normal precautions were not possible to maintain during the intensive work to clear the areas with extensive windthrow. In the area most severely affected by the storm, the drastic increase in deforested areas is therefore expected to have had an impact on nitrate leaching to run-off surface waters. 5. Conclusions Areas with the most severe damage (moderate and large damage) had in most cases increased nitrate concentrations in the soil water below the root zone following the storm. On average the increase was more than 1 mg l1 of NO3–N for these areas compared with well below 0.5 mg l1 NO3–N for the areas with no or small damage. No significant correlation between the size of the nitrogen deposition and increased nitrate concentrations in the soil water due to the storm was found in this study. The combined effect of the extent of storm damage, i.e. biomass removal, and site conditions, influences the nitrogen cycle, e.g. nitrogen mineralization and immobilization and nutrient uptake by vegetation and the period of disruption of the nitrogen cycle depends both on the intensity of the storm and on the resilience of the ecosystem. Acknowledgements This study was supported by two Grants from the Swedish Environmental Protection Agency – Sweden; Förhöjd kväveutlakning i skogsmark efter stormen Gudrun and the research programme CLEO (Climate Change and Environmental Objectives). Furthermore the authors are grateful for the contribution from the Swedish Throughfall Monitoring Network (SWETHRO). The authors would also like to thank numerous field staff for their work collecting the data. References Adriaenssens, S., Hansen, K., Staelens, J., Wuyts, K., De Schrijver, A., Baeten, L., Boeckx, P., Samson, R., Verheyen, K., 2012. Throughfall deposition and canopy exchange processes along a vertical gradient within the canopy of beech (Fagus sylvatica L.) and Norway spruce (Picea abies (L.) Karst). Sci. Total Environ. 420, 168–182. Akselsson, C., Westling, O., Örlander, G., 2004. Regional mapping of nitrogen leaching from clearcuts in southern Sweden. For. Ecol. Manage. 202, 235–243. Blennow, K., Andersson, M., Bergh, J., Sallnäs, O., Olofsson, E., 2010. Potential climate change impacts on the probability of wind damage in a south Swedish forest. Clim. Change 99, 261–278. Dämmgen, U., Eriskman, J.W., Cape, J.N., Grünhage, L., Fowler, D., 2005. Practical considerations for addressing uncertainties in monitoring bulk deposition. Environ. Pollut. 134, 535–548. Draaijers, G., Erisman, J., Spranger, T., Wyers, G., 1996. The application of throughfall measurements for atmospheric deposition monitoring. Atmos. Environ. 30, 3349–3361. Emmet, B.A., Anderson, J.M., Hornung, M., 1991a. Nitrogen sinks following two intensities of harvesting in a Sitka spruce forest (N. Wales) and the effect on the establishment of the next crop. For. Ecol. Manage. 41, 81–93. Emmet, B.A., Anderson, J.M., Hornung, M., 1991b. The controls on dissolved nitrogen losses following two intensities of harvesting in a Sitka spruce forest (N. Wales). For. Ecol. Manage. 41, 65–80. Eugster, W., Haeni, M., 2013. Nutrients or pollutants? Nitrogen deposition to european forests. Dev. Environ. Sci. 13. http://dx.doi.org/10.1016/B978-0-08098349-3.00003-7. Ferm, M., 1993. Throughfall measurements of nitrogen and sulphur compounds. Int. J. Anal. Chem. 50, 29–43. Ferm, M., Hultberg, H., 1999. Dry deposition and internal circulation of nitrogen, sulphur and base cations to a coniferous forest. Atmos. Environ. 33, 4421–4430. Futter, M.N., Ring, E., Högbom, L., Entenmann, S., Bishop, K.H., 2010. Consequences of nitrate leaching following stem-only harvesting of Swedish forests are dependent on spatial scale. Environ. Pollut. 158, 3552–3559. Galloway, J.N., Aber, J.A., Erisman, J.W., Seitzinger, S.P., Howarth, R.W., Cowling, E.B., Cosby, B.J., 2004. The nitrogen cascade. Bioscience 53, 341–356. Gregow, H., Peltola, H., Laapas, M., Saku, S., Venäläinen, A., 2011. Combined occurrence of wind, snow loading and soil frost with implications for risks to
9
forestry in Finland under the current and changing climatic conditions. Silva Fennica 45 (1), 35–54. Gundersen, P., Schmidt, I.K., Raulund-Rasmussen, K., 2006. Leaching of nitrate from temperate forests – effects of air pollution and forest management. Environ. Rev. 14, 1–57. Karlsson, P.E., Ferm, M., Hultberg, H., Hellsten, S., Akselsson, C., Pihl Karlsson, G. 2011. Total deposition of nitrogen to forests, IVL Report B 1952 (in Swedish). Kreutzweiser, D.P., Hazlett, P.W., Gunn, J.M., 2008. Logging impacts on the biogeochemistry of boreal forest soils and nutrient export to aquatic systems: a review. Environ. Rev. 16, 157–179. http://dx.doi.org/10.1139/A08-006. Legout, A., Nys, C., Picard, J.-F., Turpault, M.-P., Dambrine, E., 2009. Effects of storm Lothar (1999) on the chemical composition of soil solutions and on herbaceous cover, humus and soils (Fougères, France). For. Ecol. Manage. 257, 800–811. Lepistö, A., Andersson, L., Arheimer, B., Sundblad, K., 1995. Influence of catchment characteristics, forestry activities and deposition on nitrogen export from small forested catchments. Water Air Soil Pollut. 84, 81–102. Mellert, K.-H., Kölling, C., Rehfuess, K.E., 1998. Vegetationsentwicklung und Nitrataustrag auf 13 Sturmkahlflächen in Bayern. Forstarchiv 69, 3–11. Nilsson, U., Örlander, G., 1999. Vegetation management on grass dominated clearcuts planted with Norway spruce in southern Sweden. Can. J. For. Res. 29, 1015–1026. Nilsson, C., Stjernquist, I., Bärring, L., Schlyter, P., Jönsson, A.M., Samuelsson, H., 2004. Recorded storm damage in Swedish forests 1901–2000. For. Ecol. Manage. 199 (1), 165–173. Örlander, G., Nilsson, U., Hällgren, J.-E., 1996. Competition for water and nutrients between ground vegetation and planted Norway spruce. N. Z. J. For. Res. 26, 99– 117. Pardo, L.H., Driscoll, C.T., Likens, G.E., 1995. Patterns of nitrate loss from a chronosequence of clear-cut watersheds. Water Air Soil Pollut. 85, 1659–1664. Peltola, H., Kellomäki, S., Väisänen, H., 1999. Model computations of the impact of climatic change on the windthrow risk of trees. Clim. Change 41, 17–36. Pihl Karlsson, G., Akselsson, C., Hellsten, S., Karlsson, P.E., 2011. Reduced European emissions of S and N – effects on air concentrations, deposition and soil water chemistry in Swedish forests. Environ. Pollut. 159, 3571–3582. Prescott, C.E., Hope, G.D., Blevins, L.L., 2003. Effect of gap size on litter decomposition and soil nitrate concentrations in a high-elevation spruce-fir forest. Can. J. For. Res. 33, 2210–2220. Reuss, J.O., Johnson, D.W., 1986. Acid Deposition and the Acidification of Soils and Waters. Ecol. Stud., vol. 59. Springer-Verlag, New York, USA, pp. 119–123. Ritter, E., 2004. Studies of natural disturbances in Danish forest ecosystems – the effect of windthrow and gap formation on N availability and soil water chemistry. In: Sigurgeirsson, A., Jögiste, K. (Eds.), Natural disturbances dynamics as component of ecosystem management planning, abstracts and short papers from the workshop of the SNS network in Geysir, Iceland, 11–15 Oct, 2003. Ritter, E., Starr, M., Vesterdal, L., 2005. Losses of nitrate from gaps of different sizes in a managed beech (Fagus sylvatica) forest. Can. J. For. Res. 35, 308–319. Rosén, K., Aronson, J.-A., Eriksson, H.M., 1996. Effects of clear-cutting on streamwater quality in forest catchments in central Sweden. For. Ecol. Manage. 83, 237–384. Schelhaas, M.J., Nabuurs, G.J., Schuch, A., 2003. Natural disturbance in the European forests in the 19th and 20th centuries. Glob. Change Biol. 9, 1620–1633. Seidl, R., Blennow, K., 2012. Pervasive growth reduction in Norway Spruce forests following wind disturbance. PLoS ONE 7 (3), e33301. http://dx.doi.org/10.1371/ journal.pone.0033301. Seidl, R., Schelhaas, M.-J., Lexer, M.J., 2011. Unraveling the drivers of intensifying forest disturbance regimes in Europe. Glob. Change Biol. 17, 2842–2852. http:// dx.doi.org/10.1111/j.1365-2486.2011.02452.x. Smith, V.H., 2003. Eutrophication of freshwater and coastal marine ecosystems: a global problem. Environ. Sci. Pollut. Res. 10, 126–207. Smith, V.H., Schindler, D.W., 2009. Eutrophication science: where do we go from here? Trends Ecol. Evol. 24, 201–207. Sørensen, R., Ring, E., Meili, M., Högbom, L., Seibert, J., Grabs, T., Laudon, H., Bishop, K., 2009. Forest harvest increases runoff most during low flows on two boreal streams. Ambio 37, 357–363. Swedish Forestry Agency, 2006. Stormen 2005 – en skoglig analys. Meddelande 1, 2006. Jönköping, Sweden. ISSN 1100-0295. Swedish Forestry Agency (in Swedish). Swedish Forestry Agency, 2014. Swedish statistical yearbook of forestry 2013, ISSN 0491-7847, Skogsstyrelsen. Swedish Forestry Agency (in Swedish). Usbeck, T., Wohlgemuth, T., Dobbertin, M., Pfister, C., Bürgi, A., Rebetez, M., 2010. Increasing storm damage to forests in Switzerland from 1858 to 2007. Agric. For. Meteorol. 150 (1), 47–55. Weis, W., Huber, C., Göttlein, A., 2001. Regeneration of mature Norway spruce stands: early effects of selective cutting and clear cutting on seepage water quality and soil fertility. Sci. World J. 1, 493–499. Westling, O., Örlander, G., Andersson, I., 2004. Effekter på askåterföring till granplanteringar med ristäkt. IVL Rapport, B 1552 (in Swedish). Wiklander, O., Örlander, G., Andersson, R., 1991. Leaching of nitrogen from a forest catchment at Söderåsen in southern Sweden. Water Air Soil Pollut. 55, 263–282. Wilpert, v.K., Zirlewagen, D., Kohler, M., 2000. To what extent can silviculture enhance sustainability of forest sites under the immission regime in central Europe? Water Air Soil Pollut. 122, 105–120.
Please cite this article in press as: Hellsten, S., et al. Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden. Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.009
Contents lists available at ScienceDirect
Forest Ecology and Management journal homepage: www.elsevier.com/locate/foreco
Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden Sofie Hellsten a,⇑, Johanna Stadmark a, Gunilla Pihl Karlsson a, Per Erik Karlsson a, Cecilia Akselsson b a b
IVL Swedish Environmental Research Institute, Box 53021, SE-400 14 Gothenburg, Sweden Lund University, Department of Physical Geography and Ecosystem Science, Sölvegatan 12, SE-223 62 Lund, Sweden
a r t i c l e
i n f o
Article history: Received 19 April 2015 Received in revised form 2 July 2015 Accepted 10 July 2015 Available online xxxx Keywords: Storm felling Soil water Nitrate Nitrogen leakage Eutrophication SWETHRO
a b s t r a c t In January 2005, south-west Sweden was hit by a severe storm that caused large damage to the forests through massive windthrow. The aim of this study was to assess the effect of this windthrow on nitrate concentrations in the soil water below the root zone on 33 forest monitoring plots within the Swedish Throughfall Monitoring Network (SWETHRO). These sites were damaged to different extents by the storm. The analysis showed increased levels of nitrate concentrations in the soil water as a consequence of storm damaged forest. The present study concerned forest ecosystems with relatively low levels of nitrogen deposition, as compared with forests analysed in previous studies. The maximum soil water nitrate concentrations occurred 1–4 years after the storm. After 5–6 years, the concentrations were back at the same levels as before the storm event. It was not possible to demonstrate a significant correlation between increased nitrate concentrations in the soil water after the storm and the level of nitrogen deposition at the site. The potential impact on ground- and surface waters due to elevated nitrate concentrations in soil water after storm events is discussed. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction The frequency and intensity of natural disturbances, such as storm events, has increased in many forest ecosystems around the world (Schelhaas et al., 2003). During the last century damage due to storms in Swedish forests has increased (Nilsson et al., 2004). The main driving forces are expected to be a combination of climate change and changes in forest cover, tree species composition as well as changes in management regimes (Nilsson et al., 2004; Seidl et al., 2011; Usbeck et al., 2010). In Sweden, storm statistics have not shown clear indications of increased frequency and intensity of storms (Blennow et al., 2010). However, climate change may increase the probability of windthrow also due to a reduced resilience of forest ecosystems to wind damage. Finnish studies have shown that warmer winters in areas where ground frost occur increase the windthrow risk of trees at northern latitudes through reduced stability (Gregow et al., 2011; Peltola et al., 1999). In January 2005, south-west Sweden was hit by a severe storm with wind speeds up to 42 m s1 (Swedish Forest Agency, 2006). The storm caused large damage through massive windthrow, corresponding to 110% of the average annual harvest rates
⇑ Corresponding author. E-mail address: sofi[email protected] (S. Hellsten).
(1998–2004) in Sweden derived from only 16% of the country’s forest area (Seidl and Blennow, 2012). The drastic increase in deforested areas following the storm could be expected to result in increased levels of nitrate leaching, which may potentially contaminate groundwater and cause eutrophication of surface waters (Galloway et al., 2004; Gundersen et al., 2006; Kreutzweiser et al., 2008; Smith, 2003; Smith and Schindler, 2009). Furthermore, nitrate leaching also contributes to soil acidification, which causes potentially toxic metals to be released in soil and surface water (Reuss and Johnsson, 1986). Leaching of inorganic nitrogen (primarily nitrate) to the soil water increases as the nitrogen uptake to trees is disrupted. Disturbances in forest ecosystems, whether they are natural (e.g. storm events, insect attacks), or anthropogenic (forestry practices, such as clear-cut logging), can affect soil conditions and nutrient processes within the soil resulting in nutrient losses. Several studies carried out in boreal forests indicate that logging affects the retention of inorganic nitrogen in the forest soil, so that nitrate leaching in the runoff water increases (e.g. Emmet et al., 1991b; Futter et al., 2010; Gundersen et al., 2006; Kreutzweiser et al., 2008; Pardo et al., 1995; Rosén et al., 1996). The soil conditions of clear-cut areas are warmer and moister than in growing forests, due to decreased transpiration and increased sun exposure when the trees have been removed (Kreutzweiser et al., 2008). Moist, warm soil conditions are favourable for decomposition of organic material resulting in nitrate
http://dx.doi.org/10.1016/j.foreco.2015.07.009 0378-1127/Ó 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: Hellsten, S., et al. Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden. Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.009
2
S. Hellsten et al. / Forest Ecology and Management xxx (2015) xxx–xxx
formation after nitrification. Furthermore, most of the uptake of nitrogen by vegetation ceases in a clear-cut area, and the changed hydrological conditions (increased runoff and raised groundwater level) may further increase leaching of nitrate (Rosén et al., 1996; Sørensen et al., 2009). Increased nitrate leaching in runoff water occurs during a time period of three to six years in the south of Sweden, and up to 15 years in the north (Akselsson et al., 2004; Futter et al., 2010; Rosén et al., 1996; Wiklander et al., 1991). The Swedish Throughfall Monitoring Network, SWETHRO, is an environmental monitoring network that measures air concentrations of pollutants, deposition and soil water chemistry at forest sites in Sweden (Pihl Karlsson et al., 2011). Many of these sites were damaged to different extents due to the storm in January 2005. In most cases there existed long time-series of measurements of soil water chemistry prior to the storm. Hence, this provided a unique opportunity to investigate the relations between the extent of damage to the forests and the change in the soil water chemistry during the following years. The plots within SWETHRO are not generally positioned within small well-defined catchments, so it was not possible to directly relate the impacts on soil water chemistry to changes in the chemistry of the surface water runoff. The aim of this study was to investigate the effect of the different extents of windthrow on nitrate (NO3–N) concentrations in soil water in forests. The study focused on the soil water concentrations of nitrate since this is the main inorganic nitrogen compound in soil water that is affected by logging, and the effect on ammonium is small (Gundersen et al., 2006; Lepistö et al., 1995). The effect of logging on organic nitrogen is also small (Lepistö et al., 1995) and was therefore not considered. 2. Materials and methods 2.1. Sites The Swedish Throughfall Monitoring Network (SWETHRO) has been operated since 1985 (Pihl Karlsson et al., 2011). The monitoring network measures air concentrations of pollutants, deposition and soil water chemistry in forest ecosystems in Sweden. Monitoring plots (30 30 m) are positioned in closed, mature, managed forests with no major roads or other pollution sources in the vicinity. Currently SWETHRO includes measurements of soil water chemistry on 62 sites in Sweden, and the longest time series are 30 years (2015). Many of these sites, in southern Sweden, were damaged to different extents due to the storm Gudrun in 2005. Sites with short time series (<10 years) and sites with vague information about the extent of the storm damage were not included in the study and only coniferous sites were considered. In total 33 sites in the south of Sweden with Norway spruce and/or Scots pine forests were included in the study and their geographical positions are shown in Fig. 1. The 33 sites were divided into four different damage classes, depending on the extent of the storm damage (Table 1). As expected, the most damaged plots were all covered with Norway spruce. About 45 SWETHRO sites are co-located with observation plots operated by the Swedish Forest Agency. For these plots, information is available for changes over time regarding vegetation cover, crown thinning, stem diameter growth, needle chemistry and soil chemistry.
at 50 cm depth in the mineral soil. Based on a soil sampling campaign made in 2010–2011, the depth to the beginning of the undisturbed mineral soil (C-layer) was estimated to 40–60 cm for the plots included in this study. Hence, a 50 cm soil depth would be in the lower part of the B-layer or in the upper part of the C-layer. In general, five lysimeters were installed on each forest site, inside the canopy close to the throughfall collectors. However, some sites had fewer lysimeters due to difficulties with soil stoniness. Lysimeters were sampled by applying tension over a two day period. The water from the lysimeters was then combined into one composite sample for analysis. Only water samples containing >50 ml of water were included in the study, to eliminate uncertainties due to small sample volumes. Throughfall (TF) and bulk (BD) deposition were measured monthly throughout the years. TF measurements were carried out at all sites during the entire period analysed, whilst BD measurements were carried out during different time periods at different sites. Further details about deposition measurements and chemical analysis can be found in Pihl Karlsson et al. (2011). 2.3. Estimates of nitrogen deposition A considerable fraction of the atmospheric nitrogen deposition reaching the forests is taken up directly by the tree canopies (Adriaenssens et al., 2012; Eugster and Haeni, 2013; Ferm, 1993; Ferm and Hultberg, 1999). Hence, throughfall measurements cannot be used directly to estimate the total nitrogen deposition to forests. Therefore, bulk measurements, as well as throughfall measurements, were used in this study. Bulk deposition measurements mainly consist of wet deposition but there is also a small fraction of dry deposition to the samplers (Draaijers et al., 1996; Dämmgen et al., 2005). Information on bulk deposition was not available for all years and sites analysed in this study. Bulk deposition of inorganic nitrogen varies geographically across Sweden in a consistent pattern (Fig. 2, Karlsson et al., 2011), based on the geographical position expressed as the sum of latitude and longitude. Based on the relation in Fig. 2, the bulk deposition was estimated for the sites with moderate and large damage for the six-year period 1999–2004, prior to the occurrence of the storm (Table 3). The fraction of dry deposition contributing to the total nitrogen deposition declines towards northeast similarly to the decline of bulk deposition (Karlsson et al., 2011). Hence, it may be assumed that the relative difference regarding the estimated nitrogen deposition between the different sites in Table 3 is correct. 2.4. Weather data Official interpolated weather data can be obtained for any location in Sweden from the Swedish Meteorological and Hydrological Institute (SMHI), regarding air temperatures and precipitation (http://luftwebb.smhi.se/). Data for annual mean air temperature and annual precipitation were obtained for all sites. Differences in annual mean air temperatures and annual precipitation were analysed for the two time periods before and after the storm, 2000–2004 and 2005–2009 (see below) for all sites classified with moderate or large damage.
2.2. Measurements
2.5. Data analysis
Soil water below the root zone was sampled three times a year, to represent the conditions before (Feb–May), during (Jun–Sep) and after the vegetation period (Sep–Dec). Soil water samples were obtained using suction lysimeters with ceramic cups (P 80), placed
A database was constructed with information about the level of damage caused by the storm at the different sites, Table 2. Data on storm damage is mainly based on notes from field protocols and contacts with field personnel collecting the samples.
Please cite this article in press as: Hellsten, S., et al. Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden. Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.009
S. Hellsten et al. / Forest Ecology and Management xxx (2015) xxx–xxx
3
Fig. 1. The 33 sites included in the study, divided into four damage classes.
Table 1 Four damage classes, depending on level of storm damage. Damage class
Level of damage
Number of sites
1 2 3
No damage <10 trees within the site fell At least 10 trees, and up to 50% of the trees fell >50% of the trees fell
15 sites 9 sites 5 sites
4
No damage Small damage Moderate damage Large damage
4 sites
16
kg of N ha-1 yr -1
14 12
y=0.8733x 2 -126.88x+4614.6 R²=0.8733
10 8 6 4 2 0 69
70 71 72 73 Geographical position (Long+Lat)
74
Fig. 2. The correlation between the annual bulk deposition of inorganic nitrogen (NO3 + NH4) and the geographical position of the monitoring site expressed as the sum of longitude and latitude.
The database was complemented with information on soil solution chemistry (nitrate concentration and pH), nitrogen deposition (bulk deposition, estimated bulk deposition and throughfall
measurements), stand properties (stand age), climate data (temperature, precipitation) and other disturbances (e.g. damage from forestry machinery), based on field protocols and information from field personnel of the SWETHRO Network. The information regarding other disturbances has not been systematically applied in this study, but is valuable when looking at the data for each individual site, e.g. explaining potential storm effects at a ‘‘no damage site’’. Soil solution chemistry, nitrogen deposition and stand properties were derived from the database of the SWETHRO Network (Pihl Karlsson et al., 2011). Estimated annual bulk deposition of inorganic nitrogen was also included (see Section 2.3). Information about soil chemistry and ground vegetation was obtained from the Swedish Forest Agency observation plots (OBS-plots). Many new plots were initiated within SWETHRO during 1996, and many of these plots were also OBS-plots. The OBS-plots were sampled for soil chemistry in different soil layers in 1996. Over time, common SWETHRO and OBS-plots needed replacement after harvesting etc., but soil chemistry was not always sampled at the new plots. As a result, there are plots within SWETHRO with and without soil chemistry sampling. Of the 33 sites included in this study, information on soil chemistry for different soil layers in the year 1996 is available for 31 sites, however, not for two of the most damaged sites, Alandsryd and Knapanäs. Data for the C/N ratio in the humus layer were used for analyses of correlations with the increases of the soil water NO3–N concentrations before and after the storm. The 33 sites were divided into four different damage classes, depending on the extent of the storm damage (Table 1). Within each damage class, nitrate leaching and the correlations with other parameters such as deposition, C/N ratio, stand age and pH in soil water were investigated. For each of the sites, the average soil water nitrate concentration was calculated during two 5-year time periods, to represent
Please cite this article in press as: Hellsten, S., et al. Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden. Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.009
4
Damage class
Site
Extent of storm damage
Location
Tree species
No damage
Arlanda (A 92)
No storm damage
5.5
536
No storm damage
Norway spruce Scots pine
70
Attsjö (G 21)
85
6.1
665
Bergby (A 01)
No storm damage
Scots pine
75
5.5
558
Edeby (D 11)
No storm damage
5.6
552
No storm damage
105
5.5
583
Greckssundet (T 02)
No storm damage
57
4.3
794
Hensbacka (O 35)
No storm damage
86
6.5
873
Hjärtsjömåla (K 03)
No storm damage
Norway spruce Norway spruce Norway spruce Norway spruce Scots pine
75
Farstanäs (A 35)
70
6.2
699
Knutsta (D 14)
No storm damage
Scots pine
72
5.7
590
Marydn (L 15)
No storm damage
7.3
766
68
6.2
660
Solltorp (E 21)
No storm damage, except a broken treetop. No storm damage
Norway spruce Scots pine
46
Risebo (H 21)
5.8
610
No storm damage
74
6.3
599
Västra Torup 2 (L 07)
No storm damage
62
6.9
840
Örlingen (T 03)
No storm damage
Norway spruce Norway spruce Norway spruce Scots pine
70
Södra Averstad (S 05)
59°400 N, 17°580 E 56°540 N, 15°70 E 59°340 N, 18°40 E 58°570 N, 16°590 E 59°60 N, 17°380 E 59°350 N, 14°430 E 58°260 N, 11°440 E 56°210 N, 14°590 E 59°190 N, 16°70 E 55°370 N, 14°50 E 58°40 N, 16°70 E 58°90 N, 15°260 E 59°10 N, 13°70 E 56°80 N, 13°310 E 59°530 N, 14°260 E
59
4.3
795
Arkelstorp (L 05)
Norway spruce Scots pine
7.3
639
57
5.2
881
Höka (E 22)
One pine fell.
Scots pine
70
5.2
716
Sticklinge (A 05)
6.6
497
64
6.6
540
Söstared (N 01)
One pine fell.
Norway spruce Norway spruce Scots pine
99
Stora Ek (R 09)
One big spruce fell in the middle of the site. One tree fell.
82
6.9
869
Vång (K 13)
A few trees fell.
74
7
575
Värnvik (F 12)
2–3 spruces fell.
52
5.4
642
Åboland (O 01)
4 trees fell (2 spruces and 2 pines).
56°110 N, 14°150 E 57°460 N, 13°450 E 58°460 N, 15°90 E 59°230 N, 18°70 E 58°380 N, 13°470 E 57°310 N, 12°150 E 56°160 N, 15°270 E 57°500 N, 14°240 E 58°320 N, 11°440 E
50
Humlered (P 93)
2–3 trees fell in the monitoring site. One pine fell at the site.
58
6.1
900
A few trees also fell in close proximity to the site.
Angelstad (G 23)
About 35% of the trees fell.
6
810
The storm removed the trees north-east of the site.
10–12 spruce trees fell in, or into, the monitoring site.
Norway spruce Norway spruce
65
Bordsjö (F 22)
56°500 N, 13°450 E 57°500 N, 14°600 E
53
5.2
597
Small damage
Moderate damage
Norway spruce Norway spruce
Stand age (yr)a
Temp (°C)b
Prec. Other disturbances (mm)c
Many trees in close proximity to the site fell.
A few trees in close proximity to the site fell.
One tree outside the monitoring site fell into the site area.
A few trees in close proximity to the site fell. Damage from forestry machinery.
A few trees also fell in close proximity to the site.
Four trees fell outside the monitoring site showing a gap in the corner of the site. Forestry machinery have accessed the site.
An area 200 m north west of the site was badly damaged.
S. Hellsten et al. / Forest Ecology and Management xxx (2015) xxx–xxx
Please cite this article in press as: Hellsten, S., et al. Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden. Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.009
Table 2 The 33 sites from SWETHRO included in the study, divided into different damage classes.
c
Timrilt (N 13)
Stand age at the time of the storm, i.e. year 2005. Annual mean temperature, derived from the closest SMHI monitoring site (Swedish Meteorological and Hydrological Institute). Annual sum of precipitation, derived from the closest SMHI monitoring site (Swedish Meteorological and Hydrological Institute). a
b
6.4 50
1143
5.6
57°40 N, 14°230 E More than 50% of the trees fell. 56°470 N, 13°90 E Tagel (G 22)
About 50% of the trees fell.
Almost all trees fell. Knapanäs (G 09)
Norway spruce Norway spruce
80
707
The site was moved 10–20 m and two new lysimeters were installed in summer 2006 (three of the original lysimeters were left at the original site).The site was further damaged in a storm 2007. Damage from forestry machinery. The site was moved 2007 due to Bark Beetles. 6.3 51
593
5.5
Norway spruce Norway spruce All trees fell. Alandsryd (F 09) Large damage
13 spruce trees fell.
Mellby (F 18)
Vallåsen (N 17)
57°120 N, 13°590 E 56°410 N, 15°90 E
79
719
Gaps have been created in close proximity to the site, and forest edges are closer. Damage from forestry machinery. 7.3 68
1046
Many trees in close proximity to the site fell. 866 6
Norway spruce Norway spruce 57°90 N, 13°360 E 56°220 N, 13°60 E
51
1053 7 Norway spruce
15 spruce trees fell, showing a gap in the monitoring site. 15 spruce trees fell.
56°570 N, 12°430 E
67
3. Results
Tree species
Borgared (N 12)
5
the conditions before the storm (2000–2004), and after the storm (2005–2009). The nitrate response in soil water as an effect of the storm damage was compared between damage classes with ANOVA and t-test. All statistical analyses were made with the statistical tools available in Microsoft Excel.
Location Extent of storm damage Site Damage class
Table 2 (continued)
Stand age (yr)a
Temp (°C)b
Prec. Other disturbances (mm)c
S. Hellsten et al. / Forest Ecology and Management xxx (2015) xxx–xxx
The results in Fig. 3 show the nitrate (NO3–N) and ammonium (NH4–N) concentrations in soil water, 1997–2013 for the four sites most severely damaged by the storm. It can be seen that there were differences between the sites regarding the timing of the onset of the elevated nitrate concentrations. However, at all sites, except Tagel, there was a clear indication of increasing nitrate concentrations in the soil water before the end of 2007. Ammonium concentrations were generally low, however, at Timrilt two high ammonium values were recorded in 2008 (8.5 and 13.2 mg l1). This indicates that the storm event may have had an effect also on rates of N mineralization, at least at Timrilt. The average nitrate and ammonium concentrations before (2000–2004) and after (2005–2009) the storm are shown in Table 4 for the different damage classes. Fig. 4 shows the difference in soil water nitrate concentrations (NO3–N) in soil water between the two time periods. The areas with no or small damage have lower nitrate leaching after the storm Gudrun than the areas with moderate and large damage (p < 0.01, two-sampled t-tests assuming unequal variances). Due to the low number of plots with moderate and large damage, no statistically significant difference between these two damage classes regarding soil water nitrate concentrations after the storm could be demonstrated. The observed and estimated inorganic nitrogen (NO3 + NH4) deposition (bulk, estimated bulk and throughfall), pH in soil water (before the storm) and nitrate in soil water (before and after the storm) for the 9 sites within damage classes’ moderate and large damage are shown in Table 3. Note that the time period with observed values does not always overlap with the period with estimated values. Fig. 5A–C shows the difference in soil nitrate concentrations before and after the storm for the moderate and large damage classes plotted against the observed and estimated bulk inorganic nitrogen deposition as well as against the observed throughfall deposition. A regression analysis showed no statistically significant correlation between difference in NO3 concentration in soil solution and the measured bulk deposition, with the estimated bulk deposition or with the throughfall deposition, neither for the sites with large damage nor for the sites with moderate damage. A plot of the relation between the difference in soil water concentration of NO3 after and before the storm and the soil C/N ratio in the humus layer is shown in Fig. 6 for the sites classified with moderate and large damage. Unfortunately, information on soil chemistry was not available for two of the most damaged sites, Alandsryd and Knapanäs. No statistically significant correlation was found between the difference in soil water concentration of NO3 after and before the storm and the soil C/N ratio in the humus layer. However, it is interesting to note that the C/N ratio in the humus layer was high (34.9) at the site Tagel, which was severely damaged by the storm but did not show an increase in the soil water concentration of NO3 after and before the storm. Furthermore, neither stand age nor soil water pH were correlated with nitrogen leaching (data not shown). Differences in the weather conditions during the two time periods 2000–2004 and 2005–2009 were analysed based on weather data obtained from the Swedish Meteorological and Hydrological Institute (SMHI), regarding air temperatures and precipitation. The analysis was made for all sites classified with moderate or
Please cite this article in press as: Hellsten, S., et al. Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden. Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.009
6
S. Hellsten et al. / Forest Ecology and Management xxx (2015) xxx–xxx
(A)
(B)
(C)
(D)
Fig. 3. Nitrate – (NO3–N) concentrations in soil water 1997–2013 in the four sites with the highest level of windthrow (>50%): (A) Alandsryd, (B) Knapanäs, (C) Tagel and (D) Timrilt. Note that the monitoring site at Tagel was moved in the autumn of 2007 due to a Bark Beetle attack and that Knapanäs was moved 10–20 m after the storm and two new lysimeters were installed during summer 2006, whilst three original lysimeters were left at the original site. The Knapanäs site was further damaged in a storm in 2007. Measurements were terminated in 2010 at Knapanäs and 2012 at Alandsryd. Timrilt has an ammonium value of 13.2 mg l1 in April 2008, which exceeds the graph. The vertical dashed lines indicate the storm event.
large damage. There were only small differences between the periods. The air temperatures were on average 0.3 °C warmer and precipitation 3% higher for the 2005–2009 as compared with the 2000–2004 period.
4. Discussion 4.1. Variation in nitrate leaching with storm damage The current study showed increased levels of nitrate concentrations in soil water as a consequence of storm damaged forest (Figs. 3 and 4 and Table 4), in agreement with the studies of Ritter (2004) and Legout et al. (2009). However, the present study concerned forest ecosystems in southern Sweden with relatively lower levels of nitrogen deposition, as compared with the forests analysed in the previous studies. At three of the most severely damaged sites (Alandsryd, Knapanäs och Timrilt) the nitrate concentration in soil water reached maximum concentrations, 3–8 mg l1, during 2006–2009, i.e. 1–4 years after the storm. After 5–6 years, the levels were almost back to the levels before the storm event. These observations are in agreement with studies of the effects of clear-cut harvests on soil water nitrate concentrations. Measurements of nitrate concentrations in runoff water in southern Sweden have shown that elevated levels of inorganic nitrogen normally occurs during 5 years after logging and that the level from clear-cuts culminates after two or three years (Futter et al., 2010; Westling et al., 2004). Elevated nitrate concentrations can, both in the case of windthrow and in the case of clear-cut harvests, be explained by diminished tree demand for nitrogen and/or an increased
Fig. 4. The difference in soil water nitrate concentrations (NO3–N) in soil water between the time period after the storm (2005–2009) and the time period before the storm (2000–2004). The diagram shows the average value ± standard error for the different damage classes; no damage (15 sites), small damage (9 sites), moderate damage (5 sites) and large damage (4 sites). The sites with moderate and large damages had higher increase in nitrate concentrations than the sites with no or small damage (p < 0.01, two-sampled t-test assuming unequal variances).
mineralization in the soil due to increased oxygen supply and warmer soil conditions as the soil is more exposed to sun radiation, due to the removal of trees. However, this study also showed examples of damaged sites without any evident effects on soil water nitrate concentration, e.g. at the site Tagel (Fig. 3C), which was severely damaged by the storm. In the year 2000 the ground vegetation at Tagel consisted of a bottom layer (bryophytes and lichens) that covered 70% of the area and a field layer (trees and shrubs < 1 m, together with
Please cite this article in press as: Hellsten, S., et al. Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden. Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.009
0.53 0.014 0.47 0.83 3.52 2.01 1.51 0.021 2.93
0.53 0.007 0.46 0.82 1.58 2.00 1.51 0.019 2.73
other vascular plants) that covered 2%. That the extent of ground vegetation may be an important factor affecting the nitrate concentrations in soil water following a storm event was suggested by Mellert et al. (1998), who noted a strong negative correlation (r2 = 0.7) between vegetation cover and nitrate concentrations in soil water at 13 plots cleared after windthrow in Bavaria, Germany. Ritter (2004) came to the same conclusion following a windthrow study in Denmark, i.e. that nitrate concentrations in soil solution were correlated with ground vegetation cover. Other studies have illustrated the importance of N uptake by vegetation after logging, both regarding the magnitude and the duration of elevated nitrate concentrations (e.g. Emmet et al., 1991a,b; Nilsson and Örlander, 1999; Weis et al., 2001; Örlander et al., 1996). There was a significantly higher nitrate leaching at sites in the two classes with moderate or large damage, compared with the sites with no or small damage due to the storm (p < 0.01, Fig. 4). Due to the limited number of plots, it was not possible to distinguish differences in the level of nitrate leakage between the classes moderate and large damage. A majority of the sites classified as storm damaged showed elevated levels of nitrate. However some sites within ‘‘no damage’’ class also had increased levels of nitrate in soil water after the storm. This may indicate that the storm had an effect, although the damage was not recorded or obvious, due to root disturbance. On the other hand, within the ‘‘small damage’’ class it is likely that the storm damage in some cases was so modest that the effect of other factors was more important to nitrate levels in soil water, compared with the storm damage itself. Wilpert et al. (2000) and Ritter et al. (2005) showed that removal of trees in small gaps (<0.1 ha), can affect nitrate leaching. In contrast, Prescott et al. (2003) noted no effect on soil nitrate concentrations at removal of single trees. One explanation to the different results in these two studies could be the influence of nitrogen deposition or site quality. Prescott’s study was carried out in a low nitrogen deposition area compared with Wilpert’s study. Hence, windthrow of single trees could potentially have an effect on nitrate leaching in high deposition areas.
4.6 4.7 4.8 4.7 4.3 4.7 4.7 5.2 4.6 4.6 2.1 11.4 3.4 18.2 3.7 3.9 4.3 13.0
0.001 0.021 0.012 0.006 1.94 0.012 0.002 0.002 0.12
Diff NO3–N (mg l1) Soil water NO3–N (after) (05-09) mg l1 Soil water NO3–N (before) (00-04) mg l1 Soil water pH (00-04) mg l1 TF dep. (99-04) kg ha1 yr1
7
a
Not all years represented, different years for different sites.
9.8 6.1 13.3 9.2 14.9 9.3 6.6 7.3 12.5 (99-01) (99-01) (99-01) (99-00) (99-00) (99-00) (99-00) (99-01,04) (99-04) 10.0 8.2 14.1 9.6 14.9 9.2 10.9 8.0 14.1 70.58 72.83 69.76 70.74 69.46 70.71 71.83 71.44 69.93 Angelstad (G 23) Bordsjö (F 22) Borgared (N 12) Mellby (F 18) Vallåsen (N 17) Alandsryd (F 09) Knapanäs (G 09) Tagel (G 22) Timrilt (N 13)
Observed bulk dep. (99-04)a kg ha-1 yr1
Estimated bulk dep. (99-04) kg ha1 yr1
4.2. Variation in leaching with nitrogen deposition
Long + lat
Table 3 Inorganic nitrogen deposition (observed bulk deposition, estimated bulk deposition and throughfall), soil water pH before the storm and soil water nitrate concentrations (before and after the storm and the difference between the two) (data from SWETHRO). For the observed bulk deposition values, the time period of observations is given within brackets. TF dep. = throughfall deposition.
S. Hellsten et al. / Forest Ecology and Management xxx (2015) xxx–xxx
Previous studies have shown a correlation between the export of nitrate after clear-cut harvesting and the level of the inorganic nitrogen deposition preceding the harvest occasion, since the mineralization of nitrogen increases with increasing nitrogen availability (Akselsson et al., 2004; Lepistö et al., 1995). However, in this study it was not possible to demonstrate a significant correlation between increased nitrate concentrations in the soil water and the level of nitrogen deposition (Fig. 5A–C). The range of bulk nitrogen deposition levels for the sites analysed in this study was however limited. Timrilt has the highest nitrate and ammonium peak after the storm, and this site is located in an area with relatively high nitrogen deposition. Prior to the storm, partly elevated nitrate concentrations in the soil water at Timrilt indicated that this site was approaching nitrogen saturation (Fig. 3D). Most of these elevated nitrate concentrations occurred during spring, before the vegetation period, hence pointing to the importance of vegetation and growing season for nitrate recycling. Nitrate concentration in the soil water at Timrilt is still at a higher level than before the storm, and show seasonal fluctuation with the highest values outside the growing season (spring and autumn). 4.3. Effects on ground- and surface waters Elevated soil water nitrate concentrations due to storm events can be important at a local scale, and contribute to groundwater
Please cite this article in press as: Hellsten, S., et al. Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden. Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.009
8
S. Hellsten et al. / Forest Ecology and Management xxx (2015) xxx–xxx
(B)
(A)
(C)
Fig. 5. The difference in soil water NO3–N concentrations before (2000–2004) and after (2005–2009) the storm, plotted against inorganic nitrogen deposition. (A) Observed bulk deposition, (B) estimated bulk deposition and (C) throughfall deposition. No correlations were statistically significant using linear regression analysis.
Fig. 6. The difference in soil water NO3–N concentrations before (2000–2004) and after (2005–2009) the storm, plotted against the C/N ratio in the humus layer, obtained from soil sampling in 1996 within the Forest Agency monitoring network. No correlations were statistically significant using linear regression analysis.
contamination. However, it is important to evaluate the effect in relation to the area affected. If storm damage only affects a few percent of a forest area, and only single trees here and there, effects on surface water quality are expected to be minimal. However, if the storm disturbance is severe, with large areas affected by windthrow, the effect could be of significance. Although nitrate leaching from Swedish forests in general is small (Futter et al., 2010), leaching can be more important in the future if forest damage from windthrow increases. Normally, the annual proportion of new clear-cuts in southern Sweden is limited to approximately 1% of the productive forest area (Swedish Forestry Agency, 2014). However, due to the storm Gudrun in 2005, the areal extent of clear-cut areas increased 10-fold during a short time period in the south of Sweden. Furthermore, the damage may be even greater in windthrow areas,
Table 4 Average values for the nitrate concentration (NO3–N, mg l1) and ammonium concentration (NH4–N, mg l1) in soil water for the four damage classes before the storm (2000– 2004) and after the storm (2005–2009). Damage class (number of sites)
Before the storm (2000–2004)
After the storm (2005–2009)
Diff.
NO3–N (mg l1)
No damage (15) Small damage (9) Moderate damage (5) Large damage (4)
0.03 0.28 0.40 0.05
0.07 0.19 1.09 1.61
0.04 0.09 0.69 1.56
NH4–N (mg l1)
No damage (15) Small damage (9) Moderate damage (5) Large damage (4)
0.06 0.22 0.08 0.03
0.04 0.38 0.15 0.43
0.02 0.16 0.07 0.40
Please cite this article in press as: Hellsten, S., et al. Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden. Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.009
S. Hellsten et al. / Forest Ecology and Management xxx (2015) xxx–xxx
compared with conventional logging due to the clearing up operations. Normal precautions were not possible to maintain during the intensive work to clear the areas with extensive windthrow. In the area most severely affected by the storm, the drastic increase in deforested areas is therefore expected to have had an impact on nitrate leaching to run-off surface waters. 5. Conclusions Areas with the most severe damage (moderate and large damage) had in most cases increased nitrate concentrations in the soil water below the root zone following the storm. On average the increase was more than 1 mg l1 of NO3–N for these areas compared with well below 0.5 mg l1 NO3–N for the areas with no or small damage. No significant correlation between the size of the nitrogen deposition and increased nitrate concentrations in the soil water due to the storm was found in this study. The combined effect of the extent of storm damage, i.e. biomass removal, and site conditions, influences the nitrogen cycle, e.g. nitrogen mineralization and immobilization and nutrient uptake by vegetation and the period of disruption of the nitrogen cycle depends both on the intensity of the storm and on the resilience of the ecosystem. Acknowledgements This study was supported by two Grants from the Swedish Environmental Protection Agency – Sweden; Förhöjd kväveutlakning i skogsmark efter stormen Gudrun and the research programme CLEO (Climate Change and Environmental Objectives). Furthermore the authors are grateful for the contribution from the Swedish Throughfall Monitoring Network (SWETHRO). The authors would also like to thank numerous field staff for their work collecting the data. References Adriaenssens, S., Hansen, K., Staelens, J., Wuyts, K., De Schrijver, A., Baeten, L., Boeckx, P., Samson, R., Verheyen, K., 2012. Throughfall deposition and canopy exchange processes along a vertical gradient within the canopy of beech (Fagus sylvatica L.) and Norway spruce (Picea abies (L.) Karst). Sci. Total Environ. 420, 168–182. Akselsson, C., Westling, O., Örlander, G., 2004. Regional mapping of nitrogen leaching from clearcuts in southern Sweden. For. Ecol. Manage. 202, 235–243. Blennow, K., Andersson, M., Bergh, J., Sallnäs, O., Olofsson, E., 2010. Potential climate change impacts on the probability of wind damage in a south Swedish forest. Clim. Change 99, 261–278. Dämmgen, U., Eriskman, J.W., Cape, J.N., Grünhage, L., Fowler, D., 2005. Practical considerations for addressing uncertainties in monitoring bulk deposition. Environ. Pollut. 134, 535–548. Draaijers, G., Erisman, J., Spranger, T., Wyers, G., 1996. The application of throughfall measurements for atmospheric deposition monitoring. Atmos. Environ. 30, 3349–3361. Emmet, B.A., Anderson, J.M., Hornung, M., 1991a. Nitrogen sinks following two intensities of harvesting in a Sitka spruce forest (N. Wales) and the effect on the establishment of the next crop. For. Ecol. Manage. 41, 81–93. Emmet, B.A., Anderson, J.M., Hornung, M., 1991b. The controls on dissolved nitrogen losses following two intensities of harvesting in a Sitka spruce forest (N. Wales). For. Ecol. Manage. 41, 65–80. Eugster, W., Haeni, M., 2013. Nutrients or pollutants? Nitrogen deposition to european forests. Dev. Environ. Sci. 13. http://dx.doi.org/10.1016/B978-0-08098349-3.00003-7. Ferm, M., 1993. Throughfall measurements of nitrogen and sulphur compounds. Int. J. Anal. Chem. 50, 29–43. Ferm, M., Hultberg, H., 1999. Dry deposition and internal circulation of nitrogen, sulphur and base cations to a coniferous forest. Atmos. Environ. 33, 4421–4430. Futter, M.N., Ring, E., Högbom, L., Entenmann, S., Bishop, K.H., 2010. Consequences of nitrate leaching following stem-only harvesting of Swedish forests are dependent on spatial scale. Environ. Pollut. 158, 3552–3559. Galloway, J.N., Aber, J.A., Erisman, J.W., Seitzinger, S.P., Howarth, R.W., Cowling, E.B., Cosby, B.J., 2004. The nitrogen cascade. Bioscience 53, 341–356. Gregow, H., Peltola, H., Laapas, M., Saku, S., Venäläinen, A., 2011. Combined occurrence of wind, snow loading and soil frost with implications for risks to
9
forestry in Finland under the current and changing climatic conditions. Silva Fennica 45 (1), 35–54. Gundersen, P., Schmidt, I.K., Raulund-Rasmussen, K., 2006. Leaching of nitrate from temperate forests – effects of air pollution and forest management. Environ. Rev. 14, 1–57. Karlsson, P.E., Ferm, M., Hultberg, H., Hellsten, S., Akselsson, C., Pihl Karlsson, G. 2011. Total deposition of nitrogen to forests, IVL Report B 1952 (in Swedish). Kreutzweiser, D.P., Hazlett, P.W., Gunn, J.M., 2008. Logging impacts on the biogeochemistry of boreal forest soils and nutrient export to aquatic systems: a review. Environ. Rev. 16, 157–179. http://dx.doi.org/10.1139/A08-006. Legout, A., Nys, C., Picard, J.-F., Turpault, M.-P., Dambrine, E., 2009. Effects of storm Lothar (1999) on the chemical composition of soil solutions and on herbaceous cover, humus and soils (Fougères, France). For. Ecol. Manage. 257, 800–811. Lepistö, A., Andersson, L., Arheimer, B., Sundblad, K., 1995. Influence of catchment characteristics, forestry activities and deposition on nitrogen export from small forested catchments. Water Air Soil Pollut. 84, 81–102. Mellert, K.-H., Kölling, C., Rehfuess, K.E., 1998. Vegetationsentwicklung und Nitrataustrag auf 13 Sturmkahlflächen in Bayern. Forstarchiv 69, 3–11. Nilsson, U., Örlander, G., 1999. Vegetation management on grass dominated clearcuts planted with Norway spruce in southern Sweden. Can. J. For. Res. 29, 1015–1026. Nilsson, C., Stjernquist, I., Bärring, L., Schlyter, P., Jönsson, A.M., Samuelsson, H., 2004. Recorded storm damage in Swedish forests 1901–2000. For. Ecol. Manage. 199 (1), 165–173. Örlander, G., Nilsson, U., Hällgren, J.-E., 1996. Competition for water and nutrients between ground vegetation and planted Norway spruce. N. Z. J. For. Res. 26, 99– 117. Pardo, L.H., Driscoll, C.T., Likens, G.E., 1995. Patterns of nitrate loss from a chronosequence of clear-cut watersheds. Water Air Soil Pollut. 85, 1659–1664. Peltola, H., Kellomäki, S., Väisänen, H., 1999. Model computations of the impact of climatic change on the windthrow risk of trees. Clim. Change 41, 17–36. Pihl Karlsson, G., Akselsson, C., Hellsten, S., Karlsson, P.E., 2011. Reduced European emissions of S and N – effects on air concentrations, deposition and soil water chemistry in Swedish forests. Environ. Pollut. 159, 3571–3582. Prescott, C.E., Hope, G.D., Blevins, L.L., 2003. Effect of gap size on litter decomposition and soil nitrate concentrations in a high-elevation spruce-fir forest. Can. J. For. Res. 33, 2210–2220. Reuss, J.O., Johnson, D.W., 1986. Acid Deposition and the Acidification of Soils and Waters. Ecol. Stud., vol. 59. Springer-Verlag, New York, USA, pp. 119–123. Ritter, E., 2004. Studies of natural disturbances in Danish forest ecosystems – the effect of windthrow and gap formation on N availability and soil water chemistry. In: Sigurgeirsson, A., Jögiste, K. (Eds.), Natural disturbances dynamics as component of ecosystem management planning, abstracts and short papers from the workshop of the SNS network in Geysir, Iceland, 11–15 Oct, 2003. Ritter, E., Starr, M., Vesterdal, L., 2005. Losses of nitrate from gaps of different sizes in a managed beech (Fagus sylvatica) forest. Can. J. For. Res. 35, 308–319. Rosén, K., Aronson, J.-A., Eriksson, H.M., 1996. Effects of clear-cutting on streamwater quality in forest catchments in central Sweden. For. Ecol. Manage. 83, 237–384. Schelhaas, M.J., Nabuurs, G.J., Schuch, A., 2003. Natural disturbance in the European forests in the 19th and 20th centuries. Glob. Change Biol. 9, 1620–1633. Seidl, R., Blennow, K., 2012. Pervasive growth reduction in Norway Spruce forests following wind disturbance. PLoS ONE 7 (3), e33301. http://dx.doi.org/10.1371/ journal.pone.0033301. Seidl, R., Schelhaas, M.-J., Lexer, M.J., 2011. Unraveling the drivers of intensifying forest disturbance regimes in Europe. Glob. Change Biol. 17, 2842–2852. http:// dx.doi.org/10.1111/j.1365-2486.2011.02452.x. Smith, V.H., 2003. Eutrophication of freshwater and coastal marine ecosystems: a global problem. Environ. Sci. Pollut. Res. 10, 126–207. Smith, V.H., Schindler, D.W., 2009. Eutrophication science: where do we go from here? Trends Ecol. Evol. 24, 201–207. Sørensen, R., Ring, E., Meili, M., Högbom, L., Seibert, J., Grabs, T., Laudon, H., Bishop, K., 2009. Forest harvest increases runoff most during low flows on two boreal streams. Ambio 37, 357–363. Swedish Forestry Agency, 2006. Stormen 2005 – en skoglig analys. Meddelande 1, 2006. Jönköping, Sweden. ISSN 1100-0295. Swedish Forestry Agency (in Swedish). Swedish Forestry Agency, 2014. Swedish statistical yearbook of forestry 2013, ISSN 0491-7847, Skogsstyrelsen. Swedish Forestry Agency (in Swedish). Usbeck, T., Wohlgemuth, T., Dobbertin, M., Pfister, C., Bürgi, A., Rebetez, M., 2010. Increasing storm damage to forests in Switzerland from 1858 to 2007. Agric. For. Meteorol. 150 (1), 47–55. Weis, W., Huber, C., Göttlein, A., 2001. Regeneration of mature Norway spruce stands: early effects of selective cutting and clear cutting on seepage water quality and soil fertility. Sci. World J. 1, 493–499. Westling, O., Örlander, G., Andersson, I., 2004. Effekter på askåterföring till granplanteringar med ristäkt. IVL Rapport, B 1552 (in Swedish). Wiklander, O., Örlander, G., Andersson, R., 1991. Leaching of nitrogen from a forest catchment at Söderåsen in southern Sweden. Water Air Soil Pollut. 55, 263–282. Wilpert, v.K., Zirlewagen, D., Kohler, M., 2000. To what extent can silviculture enhance sustainability of forest sites under the immission regime in central Europe? Water Air Soil Pollut. 122, 105–120.
Please cite this article in press as: Hellsten, S., et al. Increased concentrations of nitrate in forest soil water after windthrow in southern Sweden. Forest Ecol. Manage. (2015), http://dx.doi.org/10.1016/j.foreco.2015.07.009