Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Effects of forest disturbance on seasonal soil temperature changes in the Tatra Mountains in southern Poland

Effects of forest disturbance on seasonal soil temperature changes in the Tatra Mountains in... The purpose of the study was to examine the effects of forest disturbance on seasonal changes in soil temperature in the Tatra Mountains (Poland). In the years 2015–2020 soil temperatures were measured at a depth of 20 cm on north- and south-facing mountain slopes in a catchment where forest was disturbed by hurricane-force winds in 2013 and in a control neighboring woodland catchment. The effect of forest disturbance was manifested first and foremost in an increase in the soil temperature during the summer months – average by 1.8 to 2.4 °C on a south-facing mountain slope – and by about 1 °C on a north-facing slope. The buffer effect of forest on soil temperature can be observed via lower coefficients of correlation between soil temperature and air temperature in a woodland catch - ment versus a disturbed catchment in the summer. In the winter, the effect of forest disturbance on soil temperature was less pronounced than in the summer. Small differences in soil temperature in the winter between the woodland catchment and the disturbed catchment were associated with the presence of snow cover and its capacity to yield thermal insulation. Good insulation of the soil from the atmosphere generated by snow cover yielded a very weak relationship between soil temperature and air temperature in the winter. In springtime the soil temperature increased the fastest on a south-facing slope in the disturbed catchment while in the autumn season, soil temperatures declined most rapidly on a slope facing north in the disturbed catchment. Key words: soil temperature; forest disturbance; windthrow; seasonal changes; Tatra Mountains Editor: Erika Gömöryová to an increase in evaporation and decline in transpira- 1. Introduction tion during low flow periods (Likens et al. 1970; Jewett The increase in air temperature observed across the Earth et al. 1995; Anderegg et al. 2012). However, according in the last few decades (Allen et al. 2018) is a cause of the to Gholami (2013), Hlásny et al. (2015) and Khaleghi rise in soil temperature. This pattern was examined in (2017), deforestation causes increase in peak discharge studies in the United States, where the soil temperature and runoff volume during rainfall events. Forest distur- at most of nearly 300 stations has shown a trend of warm- bance along with the soil warming leads to changes in ing in the last 35 years, with the average warming rate carbon cycling (Davidson & Janssens 2006; Allen 2009), at 0.38 °C per 10 years (Hu & Feng 2003). The increase as does nutrient cycling (Anderegg et al. 2012) and soil in soil temperature and the associated soil drought have respiration (Londo et al. 1999; Schlesinger & Andrews served as the cause of forest disturbance in many parts 2000; Ney et al. 2019). of the world over the last few decades (Allen et al. 2010; Most studies indicate that forest disturbance and Anderegg et al. 2012; Martinez-Vilalta 2012; Steinkamp deforestation cause a marked rise in soil temperature dur- et al. 2015). Forest disturbance as well as deforestation ing the summer months (Donnelly et al. 1991; Bhatti et drive an array of feedback mechanisms. For example, al. 2000; Hu et al. 2003; Moroni et al. 2009) and minor they generally lead to a continuous increase in the soil changes in the winter months (Donnelly et al. 1991; temperature primarily due to an increase in near-ground Hashimoto & Suzuki 2004; Moroni et al. 2009). How- solar radiation (Anderegg et al. 2012). The rise in soil ever, Whitson et al. (2005) measured distinctly higher temperature due to forest disturbance and deforestation triggers changes in the functioning of ecosystems. Water soil temperatures at harvested sites than at forest sites circulation patterns in catchments change mostly due during snowmelt. Iwahana et al. (2005) also argue that *Corresponding author. Joanna P. Siwek, e-mail: joanna.siwek@uj.edu.pl, phone: +48 126 645 277 J. P. Siwek / Cent. Eur. For. J. 67 (2021) 35–44 a neighboring undisturbed woodland catchment. In the clear-cutting enhances ground thawing and the depth of the soil active layer in areas underlain by permafrost, for course of the research a unique opportunity presented example in the eastern Russian region of Siberia. Snow itself. It became possible to directly examine the effects of forest disturbance on soil temperature in the same catch- cover also plays an important role in seasonal changes in ment. Spruce stands in a woodland catchment experi- soil temperature. The depth of snow together with snow thermal conductivity play a key role in snow thermal enced an attack of the bark beetle in 2018. In the following year the stands were partly overturned by high winds. insulation, as noted by Iwahana et al. (2005), Mellander et al. (2005) and Hu et al. (2013). Work by Mellander et al. (2005) showed that the nature of tree stands deter- mines the thickness of snow cover in a woodland area. For 2. Material and methods example, an open tree stand together with a low surface 2.1. Study area and meteorological conditions area covered by its canopy ultimately result in thicker layers of snow. The research study was conducted in two small sub- The purpose of the present study was to investigate catchments: (1) disturbed catchment, and (2) control the effect of forest disturbance triggered by natural causes woodland catchment (Fig. 1). Both sub-catchments are – such as a hurricane-force wind – on seasonal changes in located in the Kościeliski Potok catchment area in the soil temperature in a mountain region where tree stands Polish part of Western Tatra Mountains, which is pro- are dominated by spruce. An increase in soil temperature tected as Tatra National Park. The woodland catchment, after forest disturbance was expected; however, the key called the Kończysta Turnia, has a surface area of 14.1 ha, questions were the following: (i) What was the magni- while the disturbed catchment, called Pośrednia Kopka, tude of the increase? (ii) How does slope exposure affect has a surface area of 14.4 ha. The woodland catchment this magnitude? These are important questions today, as is located at elevations ranging from a minimum of 968 forest disturbance may increase soil warming caused by to a maximum of 1,264 m. The disturbed catchment is global climate warming. Despite the availability of sev- found at elevations in the range from 940 m to 1,200 m. eral global soil surface temperature databases based on The mean gradient of the woodland catchment is 30.1°, satellite products, according to Cassardo et al. (2018), and that of the disturbed catchment is 23.6° (Żelazny et there is still a lack of observational data on soil tempera- al. 2018). Both catchments are fragmented by narrow, ture profiles, especially for the root layer zone. deep V–shaped valleys, and are formed of sedimentary Initial research work focused on a comparison of soil rocks: limestone, sandstone, conglomerates. These par- temperature in a catchment where forest was disturbed ent material rocks are covered with Rendzic Leptosols by hurricane-force winds in 2013 and soil temperature in (Skeletic) and Haplic Cambisols (Eutric) (Skiba et al. Fig. 1. Study area in 2008 and 2014. 36 J. P. Siwek / Cent. Eur. For. J. 67 (2021) 35–44 2015). These soils are most often 40 to 60 cm deep. Soils in the study period occurred in late 2015 and early 2016 found in the woodland catchment are characterized by as well as late 2019 and early 2020 (Fig. 2a). According the occurrence of horizon O (1–5 cm), while soils found to Ustrnul et al. (2015), the mean annual atmospheric in the disturbed catchment lack this horizon (Żelazny et precipitation total in the study area ranges from 1,200 al. 2018). In both catchments the mineral horizons are mm at lower elevations to 1,600 mm at high elevations. formed of silt loam (USDA classification); however, the In the study period the highest atmospheric precipita- A horizon is characterized by a higher proportion of sand tion totals were recorded in the summer months (Fig. than lower horizons (Table 1). 2b). Snow cover appeared usually in October or Novem- Table 1. Texture and carbon (C) content in mineral horizons of the soils in the studied disturbed and woodland catchments – data source: Żelazny et al. (2018). Disturbed catchment Woodland catchment Depth C Sand Silt Clay Depth C Sand Silt Clay Soil horizon Texture* Soil horizon Texture* [%] [%] [cm] [cm] A 0–4 6.89 24 64 12 Silt loam A 0–6 6.66 29 60 11 Silt loam AB 4–19 3.98 8 74 19 Silt loam B 6–40 1.45 0 83 17 Silt loam BCa 19–40 2.19 0 82 18 Silt loam BCa 40–60 1.85 0 84 16 Silt loam *according to United States Department of Agriculture. ber and melted in April or May. The largest snow depth The study area is located in a temperate climate zone – was noted usually in December, January, and February, in a moderately cool climate zone (Hess 1965). According reaching usually more than 20 cm. Exceptionally small to Żmudzka et al. (2015) the mean annual air tempera- snow cover depths were noted in the winter of 2015/2016 ture ranges from 4 °C at lower elevations to 6 °C at high (Fig. 2c). elevations. In the study period (2015–2020) the warmest Originally, the study area was occupied by Dentario months of the year were June, July, and August when the glandulosae-Fagetum, dominated by beech (Fagus syl- mean monthly air temperature exceeded 10 °C (Fig. 2a). vatica L.), fir (Abies alba L.), and spruce (Picea abies The coolest months of the year when the mean monthly L.). The trees were felled at the end of the 19th century air temperature decreased below 0 °C were usually December, January, and February. The warmest winters and monocultures of spruce were planted at the begin- Fig. 2. Meteorological characteristics of study period: monthly air temperature (a), precipitation (b) and snow depth (c). 37 J. P. Siwek / Cent. Eur. For. J. 67 (2021) 35–44 ning of the 20th century (Szwagrzyk et al. 2019). The were installed on a south-facing slope in the woodland catchment (WS site). The probes experienced failures share of spruce in the studied stands exceeded 90% several times during the study period, which explains while the share of fir and beech did not exceed 10% gaps in the data set used in the study. (Bodziarczyk et al. 2019). Until December 2013 about Meteorologic data were obtained from weather sta- 93% of the Kończysta Turnia catchment was covered tions run by the Institute of Meteorology and Water Man- with forest, while almost 100% of the Pośrednia Kopka agement in Poland. The weather stations were located in catchment was forestland (Fig. 1). The age of the for- the vicinity of the study areas examined. Air temperature est ranged from 85 to 150 years. In December 2013 the data measured at a height of two meters above ground Pośrednia Kopka catchment experienced hurricane-force level were obtained from the Polana Chochołowska sta- winds felling 97% of its tree stands (Fig. 1). The maxi- tion, which is located about 6 km away from the stud- mum hourly average wind velocity during the windthrow ied catchments, and found at 1,147 meters of elevation. event was 29 m/s (Strzyżowski et al. 2016). In the same Atmospheric precipitation data and snow cover data were time period, only 13% of the forest in the Kończysta Tur- obtained from the Kiry-Kościelisko station located about nia catchment became disturbed (Żelazny et al. 2018). 1 km away from the studied catchments at an elevation Some of the fallen trees lying across the hillslopes in the of 933 meters. lower and central parts of the Pośrednia Kopka catch- Monthly soil temperature analyses were performed ment area were removed in 2014–2015. Fir and beech only for months with a complete set of data for particular seedlings were planted in a part of this area in 2015. study site. The coefficient of correlation for mean daily Research work performed in 2018 showed that 23% of soil temperature data for the 4 study sites with respect the disturbed area of the Pośrednia Kopka catchment was to mean air temperature data was calculated only for experiencing a succession of bush vegetation and juve- months with a complete data set from all the study sites. nile trees (Żelazny et al. 2018). Vaccinium myrtillus L., The Pearson correlation coefficient ( r) was computed. Rubus idaeus L., Athyrium distentifolium, Sorbus aucu- Mean daily air and soil temperatures were calculated paria L., Calamagrostis villosa, Deschampsia flexuosa using row data. Statistical analyses were performed in L., Homogyne alpina L., and Oxalis acetosella L. are the Statistica 1.3 software. most common species, which have emerged in the dis- turbed area (Szwagrzyk et al. 2019). The same study also showed that in 2018 almost 50% of the forested area in the Kończysta Turnia catchment experienced an attack of 3. results the bark beetle. In 2019 a large percentage of trees on the The largest differences between the highest and the low- south-facing slope of the woodland catchment became est values of the soil temperature over the course of the felled by high winds present in the area. year were noted on the south-facing slope in the disturbed catchment (DS site), while the smallest were noted on the north-facing slope in the woodland catchment (WN 2.2. Methods of measurements site) (Fig. 3). In summer months (June to August) the Soil temperature probes (Decagon ECH2O 5TM) were soil temperature at the DS site usually exceeded 15 °C – it ranges from about 12.5 °C up to 20 °C in summer 2017. installed at two sites in each of the studied catchments. Soil temperatures were lower during the same period of The WN and WS sites located in the studied woodland time on the south-facing slope in the woodland catch- catchment represent north-facing and south-facing ment (WS site), where usually did not exceed 15 °C. Soil woodland hillslopes, respectively. The studied DN and DS sites located in the disturbed catchment represent temperatures in the summer on north-facing slopes in the north-facing and south-facing disturbed hillslopes, woodland and disturbed catchments usually ranged from 10 to 15 °C – exceeding the latter only in selected cases. respectively. Soil temperature probes were installed in Soil temperatures decreased below 5 °C in all the studied the mineral soil horizons at a depth of 20 cm. The 20 catchments in wintertime, but did not normally fall below cm depth was chosen to ensure that the probes will be installed in the mineral and not the organic horizons of 0 °C. Only the soil temperature at the DS site decreased the soil. The probes were placed at an elevation of roughly below 0 °C in the winter of 2015. Soil temperatures in 900 m (Fig. 1). The technical specifications of the Deca- the spring and autumn months usually ranged from 5 to 10 °C (Fig. 3). gon ECH2O 5TM probes: range from −40 to 60 ºC, reso- The smallest monthly differences in soil temperature lution at 0.1 ºC, ±1 ºC. Soil temperature measurements were collected every 10 minutes. The first probes were occurred in all the studied sites in the wintertime, as evi- installed in October 2015 on a north-facing slope in the denced by the monthly interquartile range and monthly woodland catchment (WN site) and on a south-facing maximum-minimum range (Fig. 3). Larger monthly dif- slope in the disturbed catchment (DS site). In April 2017 ferences in the soil temperature in spring, summer, and probes were installed on a north-facing slope in the dis- autumn in both catchments occurred on south-facing turbed catchment (DN site), while in June 2017 probes slopes (vs. north-facing), with the largest monthly dif- 38 J. P. Siwek / Cent. Eur. For. J. 67 (2021) 35–44 Fig. 3. Statistical characteristics of soil temperature at four studied sites for each month of the studied period (November 2015 – February 2020). The above statistics are based on raw data. ferences in soil temperature noted at the DS site. The rela- mean monthly soil temperature data (Table 3). tionship between mean daily soil and air temperatures for a given month was stronger for south-facing slopes (DS Table 2. The Pearson correlation coefficient (p ≤ 0.05) be- and WS sites) than north-facing slopes (DN and WN tween mean daily soil temperature and mean daily air temper- sites). The strongest relationship was noted for the DS ature for four studied sites: DS – south-facing slope of the dis- site and weakest for the WN site (Table 2). This relation- turbed catchment, DN – north-facing slope of the disturbed catchment, WS – south-facing slope of the woodland catch- ship was strongest for all sites in the summer months and weakest in winter. ment, WN – north-facing slope of the woodland catchment (ns – not significant). A comparison of soil temperature measured at the Year / month DS DN WS WN same time on south-facing slopes in the disturbed and 2017 / June 0.804 0.614 0.833 0.645 woodland catchments (DS site vs. WS site) showed that 2017 / July 0.845 0.644 0.758 0.443 in 2017 and early 2018 (January to March, lack of data 2017 / August 0.874 0.618 0.748 0.487 2017 / September 0.774 0.628 0.707 0.572 for DS for subsequent months) the soil temperature at 2017 / November 0.565 0.516 0.657 0.502 the DS site was much higher than at the WS site. Mean 2017 / December ns -0.392 ns −0.395 2018 / January 0.479 0.375 0.444 ns monthly soil temperatures in summer (June to August 2018 / February 0.765 ns 0.475 0.773 2017) were from 1.8 to 2.4 °C higher at the DS site than 2018 / March 0.571 ns ns −0.389 the WS site (Table 3). These differences were smaller in 2019 / February ns ns ns ns 2019 / March 0.556 ns 0.490 ns the winter. This pattern changed in 2019 when soil tem- 2019 / April 0.788 0.492 0.737 0.441 peratures were often higher at the WS site than the DS 2019 / May 0.830 0.795 0.802 0.599 2019 / June 0.665 0.525 0.599 0.386 site, especially in the summer (Fig. 4), and mean monthly 2019 / July 0.798 0.730 0.722 0.602 soil temperatures at both study sites were comparable 2019 / August 0.736 0.465 0.694 ns (Table 3). A comparison of soil temperatures measured at 2019 / September 0.613 0.606 0.652 0.538 2019 / October 0.508 0.421 0.700 0.441 the same time in both catchments on north-facing slopes 2019 / November ns 0.453 0.508 ns (DN site versus WN site) showed that soil temperatures 2020 / January ns ns ns −0.364 2020 / February −0.405 ns ns ns in the summer were higher at the DN site versus the WN site, while in the winter the situation was just the opposite In the springtime, soil temperatures first begin to – higher soil temperature values were noted for the WN site versus the DN site. This pattern was evidenced by increase at the DS site, with temperatures higher than 39 J. P. Siwek / Cent. Eur. For. J. 67 (2021) 35–44 5 °C already in March. At other sites, soil temperatures catchment. After partial disturbance on the south-fac- begin to rise later, in April. The situation changed at the ing slope of the woodland catchment in 2019, the mean WS site in 2019 when the soil temperature at this site monthly soil temperature increased and was comparable began to rise as quickly as that at the DS site (Fig. 3). A to that in the disturbed catchment in 2013 (Table 3). The decrease in temperature in late autumn and winter below soil temperature difference on the north-facing slopes of the woodland and disturbed catchments in summertime 5 °C is r fi st observed on north-facing slopes in the studied was smaller at about 1 °C (Table 3). Higher soil temper- disturbed catchment. atures in disturbed areas versus woodland areas in the summertime have been documented by many researchers Table 3. Mean monthly soil temperature [°C] at four studied in other parts of the world. The differences in mean soil sites: DS – south-facing slope of the disturbed catchment, DN – north-facing slope of the disturbed catchment, WS – south- temperatures described by these other researchers for facing slope of the woodland catchment, WN – north-facing the summer months between woodland and disturbed slope of the woodland catchment. areas do vary mostly due to differences in the depth at Year / month DS DN WS WN which each measurement was performed. According to 2017 / June 14.3 12.0 11.9 10.9 Paul et al. (2004) soil depth is the most important factor 2017 / July 14.8 13.3 12.8 12.2 2017 / August 16.1 13.9 14.3 13.3 aside from other key factors such as air temperature and 2017 / September 12.2 10.3 10.5 10.3 the surface area covered by tree canopies controlling soil 2017 / November 4.5 3.0 3.8 4.1 2017 / December 2.4 1.7 1.9 2.5 temperature. Moroni et al. (2009) studied soil tempera- 2018 / January 1.8 1.2 1.5 1.9 ture at a depth of 10 cm in Newfoundland in Canada and 2018 / Fabruary 1.6 1.1 1.0 1.6 2018 / March 2.4 0.8 0.9 1.3 found that the difference in the soil temperature between 2019 / January 2.3 1.5 1.9 2.5 a harvested area and woodland area in the summer sea- 2019 / Fabruary 2.1 1.3 1.4 2.0 2019 / March 4.0 1.1 2.9 2.2 son equals about 2 °C. Bhatti et al. (2000) showed in 2019 / April 7.3 2.9 6.2 4.5 northeastern Ontario in Canada that soil temperature at a 2019 / May 8.5 7.2 7.8 6.6 2019 / June 14.6 13.7 14.5 12.4 depth of 5 cm increases in the summer after harvesting by 2019 / July 14.4 13.1 14.4 12.5 4 to 6 °C. According to research by Hu et al. (2013) con- 2019 / August 15.7 14.2 15.6 13.7 ducted in a steppe area in Central Asia, the soil tempera- 2019 / September 13.2 11.0 12.6 11.2 2019 / October 10.6 7.6 9.9 8.8 ture at a depth of 10 cm was higher by 4 to 8 °C relative 2019 / November 7.5 5.3 6.8 6.5 to that in a woodland area in the warm season. Londo et 2019 / December 4.3 2.5 3.4 3.9 2020 / January 3.3 1.9 2.1 2.7 al. (1999) studied soil temperature in Texas in the United 2020 / Fabruary 2.9 1.4 2.1 2.1 States and found a marked increase in soil temperature after deforestation – the measurements were performed at a depth of 10 cm in the summer season. The increase in 4. Discussion soil temperature was larger after clearcutting than after The effect of forest disturbance on the soil temperature partial clearcutting (Londo et al. 1999). at 20 cm depth in the study area is observable first and According to Hashimoto & Suzuki (2004) a marked foremost via a rise in the soil temperature in the sum- increase in soil temperature following forest clear-cutting mer months. The mean soil temperature in the summer is associated with an increase in absorbed solar radiation. months on south-facing slopes in the disturbed catch- Soils in woodland areas are protected from heating not only by trees but also by the organic forest floor situated ment is 1.8 to 2.4 °C higher than that in the woodland Fig. 4. Relationship between soil temperature (raw data) at DS site and WS site (A), and DN site and WN site (B) taking into account all measurements made in 2017 (red points), 2018 (grey points), and 2019 (blue points). 40 J. P. Siwek / Cent. Eur. For. J. 67 (2021) 35–44 atop the mineral soil horizons. The forest floor acts as due to increased solar radiation in disturbed areas versus a thermal insulator (Bhatti et al. 2000). The presented woodland areas, as shown by larger diurnal changes in research has shown that an area without trees and an the soil temperature on a south-facing slope in a dis- organic layer, such as a disturbed catchment, heats up turbed catchment versus all other measurement sites more during the summer than a wooded area with an (unpublished data). organic layer that is 1–5 centimeters thick, as would be Soil temperatures rise in spring (>5°C), as early the case in a woodland area. The buffer effect of forest as March, in the disturbed catchment on south-facing on soil temperature is confirmed by lower coefficients of slopes. At all other measurement sites, soil temperatures correlation between soil temperature and air temperature begin to increase only in April. Hu et al. (2013) studied in a woodland catchment versus a disturbed catchment. soil temperature in Central Asia and observed that the The effect of forest disturbance on soil temperature spring thawing of soils occurs two weeks earlier in tree- measured at a depth of 20 cm in the winter months in the less areas (i.e. steppe) versus woodland areas. On the catchments examined in the present study was weaker other hand, Whitson et al. (2005) studied soil tempera- than that in the summer. Similar conclusions were tures in southwestern Canada and learned that land cover reached by Donnelley et al. (1991) for the northeastern (harvested sites vs. forested sites) does not determine the United States, Hashimoto & Suzuki (2004) for Japan, timing of the spring soil thaw. Instead, slope exposure and Moroni et al. (2009) for Newfoundland (Canada). is the determining factor: the soil profiles of southerly Small differences in soil temperature in the winter aspect began thawing in mid-April, while those with between the woodland and disturbed areas examined northerly aspect in the latter half of May (Whitson et in the present study were undoubtedly associated with al. 2005). The presented research has shown that both the presence of snow cover and its capacity for thermal land cover and slope exposure determine the timing of insulation. The thickness of the snow cover in the study increases in soil temperature in spring. On south-facing period in the winter months usually reached into the tens slopes in disturbed areas, the snow cover disappears of centimeters (Fig. 2c). The soil temperature is above more rapidly than that on north-facing slopes as well as freezing in winter months when mean daily air temper- on woodland slopes. Soil on a disturbed, south-facing atures fall below −5 °C and sometimes below −20 °C. slope becomes devoid of snow earlier, on the one hand, Short-term decline in soil temperature below 0 °C on a and is not yet covered by young grass and bushes on the south-facing slope in the disturbed catchment could be other. The lack of thermal insulation leads to increase the due to the negligible thickness of snow cover, less than absorbed solar radiation triggering a rise in soil tempera- 10 cm. It is possible that the snow simply melted on the ture. Mellander et al. (2005) assert that in years with thin south-facing slope in the disturbed catchment. Meng et snow cover there was the largest variation in the timing al. (1995) observed that soil frost penetrates deeply into of soil warming (when soil temperature reaches 5 °C) in deforested wind-exposed soil, e.g. soil without snow spring between stands of different age. In autumn, soil cover. The relatively good insulation provided by snow temperatures examined in the present study fall the earli- cover to the soil separating it from the weather above est on north-facing slopes in the disturbed catchment due yields only a poor correlation between soil temperature to both less absorbed solar radiation than on south-facing and air temperature in the winter – unlike that in the sum- slopes and the absence of the buffer effect of forest. mer months. According to Mellander et al. (2005), snow The presented research represents a small step depth together with snow thermal conductivity play an towards a better understanding that forest disturbance important role in the timing of soil warming in winter may lead to dangerous feedback loops in the atmos- and spring. phere, pedosphere, and biosphere. The increase in soil Forest disturbance leads to an increase in the differ- temperature caused by forest disturbance will result in ence in soil temperature over the course of the year. On an increased release of greenhouse gases from the soil south-facing slopes, this is caused by a very large increase into the atmosphere. Soil temperature is a very impor- in soil temperature in the summer; in the winter months, tant factor shaping methane emissions (Mikkela et al. soil temperatures are generally somewhat higher than 1995) and soil respiration – soil CO efflux (Davidson et those in the woodland catchment. However, on north- al. 1998; Subke et al. 2003; Uvarov et al. 2006; Tang et facing slopes, this is caused by a small rise in the soil al. 2008). For example, Hick Pries et al. (2017) noted an temperature in summer and a small decline in winter. increase in CO loss of 34% to 37% from the r fi st meter of The presented research has confirmed the assertion by the top mineral soil horizon with a rise of 4ºC in soil tem- Hu et al. (2013) that the forest buffers the soil tempera- perature. The increase in greenhouse gas emissions will ture by lowering the soil temperature during the warm induce an increase in air temperature, which will result season and increasing it during the cold season; however, in a further increase in soil temperature and further tree this is only true of north-facing mountain slopes. What is mortality. Ney at al. (2019) note that forest management puzzling is the phenomenon of higher soil temperatures practices enacted following a natural forest disturbance on south-facing slopes in winter in the disturbed catch- are becoming an increasingly important part of climate ment versus the woodland catchment. This is most likely change mitigation strategies. 41 J. P. Siwek / Cent. Eur. For. J. 67 (2021) 35–44 intensify soil warming caused by climate warming. More- 5. Conclusions over, forest disturbance will lead to dangerous feedback The effect of forest disturbance triggered by hurricane- mechanisms. The increase in soil temperature caused by force winds on soil temperatures at a 20 cm depth in the forest disturbance will cause increased emissions of CO Western Tatras in Poland is marked by an increase in the from the soil to the atmosphere, which will contribute to difference in the soil temperature over the course of the an increase in air temperature. This will result in a further year. This is due r fi st and foremost to a marked increase in increase in soil temperature and the decline of some tem- soil temperature during the summer months. The mean perature sensitive tree stands. In order to minimize the soil temperature in the summer months on south-facing soil warming effect triggered by forest disturbance, tree slopes in the studied disturbed catchment was 1.8 to seedlings should be planted as soon as possible after for- 2.4 °C higher than that in the studied woodland catch- est disturbance. Moreover, this type of situation should ment. On north-facing slopes, the difference in the soil be used to plant stands whose tree species composition temperature between the woodland catchment and dis- corresponds to the natural habitat in a given climate zone turbed catchment in the summer months was smaller at in the mountains. Natural forest stands are believed to about 1 °C. The presented research has confirmed the be more resistant to natural disasters such as hurricane- presence of the buffer effect of forest on soil temperature force winds or bark beetle invasions than spruce mono- in summer, which is manifested by lower coefficients of cultures frequently found in Central Europe. correlation between soil temperature and air temperature in the woodland versus disturbed catchment. The effect of forest disturbance on soil temperature Acknowledgement in the winter months is less pronounced than in the The research was financed by the Jagiellonian University in summer months. This is associated with the presence of Kraków, Institute of Geography and Spatial Management. The snow cover that acts as a thermal insulator. In the winter authors wish to thank Grzegorz Zębik for his helpful advice and months (December-February) snow cover depth usually review of the English language of the manuscript. exceeds 20 cm in the studied area. Good insulation of the soil from the atmosphere by snow cover leads to a very weak relationship between soil temperature and air tem- references perature in winter. The soil temperature on disturbed, north-facing slopes in the winter is lower than that on Allen, C. D., 2009: Climate-induced forest dieback: woodland, north-facing slopes. This suggests the pres- a n e s c a l a t i n g g l o b a l p h e n o m e n o n ? U n a s y l v a , ence of a buffer effect of forest on the soil temperature 231/232:43–49. that increases the soil temperature during the colder Allen, C. D., Macalady, A. K., Chenchouni, H., Bachelet, season and decreases it in the warmer season, on north- D., McDowell, N., Vennetier, M. et al., 2010: A global facing slopes. However, it is puzzling that a south-facing overview of drought and heat-induced tree mortal- slope in a disturbed catchment would exhibit slightly ity reveals emerging climate change risks for forests. higher soil temperatures in winter than those in a wood- Forest Ecology and Management, 259:660–684. land catchment. This is likely due to increased absorbed Allen, M. R., Dube, O. P., Solecki, W., Aragón-Durand, solar radiation on south-facing slopes in disturbed areas F., Cramer, W., Humphreys, S. et al., 2018: Fram- versus woodland areas. However, this finding requires ing and Context. In: Masson-Delmotte, V., Zhai, P., further research. Pörtner, H-O., Roberts, D., Skea, J., Shukla, P.R. The research has shown that land cover (i.e. forest vs. et al. (eds): Global Warming of 1.5 °C. An IPCC lack of forest) and slope exposure (south versus north) Special Report on the impacts of global warming of determine the timing of increases in soil temperature in 1.5 °C above pre-industrial levels and related global spring and timing of soil temperature declines in autumn. greenhouse gas emission pathways, in the context In the spring season, soil temperatures increase the fast- of strengthening the global response to the threat of est (in March) on disturbed, south-facing slopes due to climate change, sustainable development, and efforts faster thawing of snow and absence of any vegetation to eradicate poverty. Intergovernmental Panel on Cli- such as young grasses or bushes. Given a low level of ther- mate Change. mal insulation, one may expect increased absorbed solar Anderegg, W. R. L., Kane, J. M., Anderegg, L. D. L., radiation that leads to a rapid increase in soil tempera- 2013: Consequences of widespread tree mortality triggered by drought and temperature stress. Nature ture. In the autumn season, temperatures on the studied Climate Change, 3:30–36. disturbed, north-facing slopes decrease at an earlier date Bhatti, J., Fleming, R. L., Foster, N. W., Meng, F-R., than elsewhere due to less absorbed solar radiation than Bourque, C. P. A., Arp, P., 2000: Simulations of on south-facing slopes as well as the lack of the buffer pre- and post-harvest soil temperature, soil mois- effect of forest cover. ture, and snowpack for jack pine: comparison with The present study shows that an increase in soil tem- e fi ld observations. Forest Ecology and Management, perature triggered by forest disturbance may signic fi antly 138:413–426. 42 J. P. Siwek / Cent. Eur. For. J. 67 (2021) 35–44 Bodziarczyk, J., Szwagrzyk, J., Zwijacz-Kozica, T., Zięba, Likens, G. E., Bormann, F. H., Johnson, N. M., Fisher, A., Szewczyk, J., Gazda A., 2019: The structure of D. W., Pierce, R. S., 1970: Effects of forest cutting forest stands in the Tatra National Park: The results and herbicide treatment on nutrient budgets in the of 2016–2017 inventory. Forest Research Papers, Hubbard Brook Watershed-Ecosystem. Ecological 80:13–21. Monographs, 40:23–47. Cassardo, C., Park, S. K., Sungmin, O., Galli, M., 2018: L o n d o , A . J . , M e s s i n a , M . G . , S c h o e n h o l t z , S . H . , Projected Changes in Soil Temperature and Surface 1999: Forest harvesting effects on soil temperature, Energy Budget Components over the Alps and North- moisture, and respiration in a bottomland hardwood ern Italy, Water, 10:1–18. forest. Soil Science Society of America Journal, 63:637–644. Davidson, E. A., Janssens, I. A., 2006: Temperature sen- Martínez-Vilalta, J., Lloret, F., Breshear, D. D., 2012: sitivity of soil carbon decomposition and feedbacks Drought-induced forest decline: causes, scope and to climate change. Nature, 440:165–173. implications. Biology Letters, 8:689–691. Donnelly, J. R., Shane, J. B., Yawney, H. W., 1991: Har- Mellander, P. E., Laudon, H., Bishop, K., 2005: Model- vesting causes only minor changes in physical prop- ling variability of snow depths and soil temperatures erties of upland Vermont soil. Northern Journal of in Scots pine stands. Agricultural and Forest Meteo- Applied Forestry, 8:33–36. rology, 133:109–118. Gholami, V., 2013: The inu fl ence of deforestation on run - Meng, F-R., Bourque, C-A., Jewett, K., Daugharty, off generation and soil erosion (Case study: Kasilian D., Arp, P. A., 1995: The Nashwaak Experimental Watershed). Journal of Forest Science, 59:272–278. Watershed Project: analysing effects of clearcutting Hashimoto, S., Suzuki, M., 2004: The impact of for- on soil temperature, soil moisture, snowpack, snow- est clear-cutting on soil temperature: a comparison melt and stream flow. Water, Air and Soil Pollution, between before and after cutting, and between clear- 82:363–374. cut and control sites. Journal of Forestry Research, Moroni, M., Carter, P., Ryan, D., 2009: Harvesting and 9:125–132. slash piling affect soil respiration, soil temperature, Hess, M., 1996: Klimat. In: Mirek, Z., Głowaciński, Z., and soil moisture regimes in Newfoundland boreal Klimek, K., Piękoś-Mirkowa, H. (eds.): Przyroda forests. Canadian Journal of Soil Science, 89:343– Tatrzańskiego Parku Narodowego, Tatry i Podta- trze 3. Tatrzański Park Narodowy. Kraków–Zako- Ney, P., Graf, A., Bogena, H., Diekkrüger, B., Drüe, C., pane, p. 53–68. Esser, O. et al., 2019: CO u fl xes before and after par - Hlásny, T., Kočický, D., Maretta, M., Sitková, Z., Barka, tial deforestation of a Central European spruce forest. I., Konôpka, M., Hlavatá, H., 2015: Effect of defor- Agricultural and Forest Meteorology, 274:61–74. estation on watershed water balance: hydrological Paul, K. I., Polglase, P. J., Smethurst, P. J., O’Connell, modelling-based approach. Central European For- A. M., Carlyle, C. J., Khanna, P. K., 2004: Soil tem- estry Journal, 61: 89–100. perature under forests: a simple model for predict- Hu, G., Liu, H., Anenkhonov, O. A., Korolyuk, A. Y., ing soil temperature under a range of forest types. Sandanov, D. V., Guo, D., 2013: Forest buffers soil Agricultural and Forest Meteorology, 121:67–182. temperature and postpones soil thaw as indicated by Schlesinger, W. H., Andrews, J. A., 2000: Soil respira- a three-year large-scale soil temperature monitoring tion and the global carbon cycle. Biogeochemistry, in the forest-steppe ecotone in Inner Asia. Global and 48:7–20. Planetary Change, 104:1–6. Skiba, S., Koreň, M., Drewnik, M., Kukla, J., 2015: Soils. Hu, Q., Feng, S., 2003: A daily soil temperature data- In: Dąbrowska, K., Guzik, M. (eds.): Atlas of the Tatra set and soil temperature climatology of the contigu- Mts. – Abiotic Nature. Tatrzański Park Narodowy. ous United States. Journal of Applied Meteorology, Zakopane, plate number: VI.1. 42:1139–1156. Steinkamp, J., Hickler, T., 2015: Is drought-induced for- Iwahana, G., Machimura, T., Kobayashi, Y., Fedorov, est dieback globally increasing? Journal of Ecology, A., Konstantinov, P., Fukuda, M., 2005: Influence of 103:31–43. forest clear-cutting on the thermal and hydrological Strzyżowski, D., Fidelus-Orzechowska, J., Żelazny, M., regime of the active layer near Yakutsk, Eastern Sibe- 2016: Geomorphological changes within a hillslope ria. Journal of Geophysical Research, 110:G02004. caused by a windthrow event in the Tatra Mounta- Jewett, K., Daugharty, D., Krause, H., Arp, P., 1995: ins, Southern Poland. Geogras fi ka Annaler. Series A, Watershed responses to clear-cutting: effects on Physical Geography, 98:347–360. soil solutions and stream water discharge in central Szwagrzyk, J., Bodziarczyk, J., Pielech, R., Zieba, A., New Brunswick. Canadian Journal of Soil Science, Zwijacz-Kozica, T., 2019: The impact of wind and 75:475–490. bark beetle outbreaks and protective measures on Khaleghi, M. R., 2017: The influence of deforestation the vegetation of spruce forests in the Tatra National and anthropogenic activities on runoff generation. Park. Parki Narodowe i Rezerwaty Przyrody, 38: Journal of Forest Science, 63:245–253. 57–68. 43 J. P. Siwek / Cent. Eur. For. J. 67 (2021) 35–44 Ustrnul, Z., Walawender, E., Czekierda, D., Šťastný, P., Żelazny, M., Adamska, M., Ciemborowicz, M., Fidelus- Lapin, M., Mikulová, K. 2015: Precipitation and snow Orzechowska, J., Gus, M., Jaśkowiec, B. et al., 2018: cover. In: Dąbrowska, K., Guzik, M. (eds.): Atlas of Ocena wpływu wielkoobszarowych wiatrołomów the Tatra Mts. – Abiotic Nature. Tatrzański Park Nar- n a r e ż i m h y d r o c h e m i c z n y i d e n u d a c j ę z l e w n i odowy. Zakopane, plate number: II.3. położonych w obszarach leśnych na terenach gór- Whitson, I., Chanasyk, D., Prepas, E., 2005: Effect of skich (Tatry Polskie) – etap 4. Kraków, Raport forest harvest on soil temperature and water storage K/KDU/000494, 395 p. and movement patterns on boreal plain hillslopes. Żmudzka, E., Nejedlík, P., Mikulová, K., 2015: Tem- Journal of Environmental Engineering and Science, perature, thermal indices. In: Dąbrowska, K., Guzik, 4:429–439. M. (eds.): Atlas of the Tatra Mts. – Abiotic Nature. Tatrzański Park Narodowy. Zakopane, plate num- ber: II.2. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Forestry Journal de Gruyter

Effects of forest disturbance on seasonal soil temperature changes in the Tatra Mountains in southern Poland

Forestry Journal , Volume 67 (1): 10 – Mar 1, 2021

Loading next page...
 
/lp/de-gruyter/effects-of-forest-disturbance-on-seasonal-soil-temperature-changes-in-aK9ljOvEiy

References (20)

Publisher
de Gruyter
Copyright
© 2021 Joanna Paulina Siwek, published by Sciendo
ISSN
0323-1046
eISSN
0323-1046
DOI
10.2478/forj-2021-0003
Publisher site
See Article on Publisher Site

Abstract

The purpose of the study was to examine the effects of forest disturbance on seasonal changes in soil temperature in the Tatra Mountains (Poland). In the years 2015–2020 soil temperatures were measured at a depth of 20 cm on north- and south-facing mountain slopes in a catchment where forest was disturbed by hurricane-force winds in 2013 and in a control neighboring woodland catchment. The effect of forest disturbance was manifested first and foremost in an increase in the soil temperature during the summer months – average by 1.8 to 2.4 °C on a south-facing mountain slope – and by about 1 °C on a north-facing slope. The buffer effect of forest on soil temperature can be observed via lower coefficients of correlation between soil temperature and air temperature in a woodland catch - ment versus a disturbed catchment in the summer. In the winter, the effect of forest disturbance on soil temperature was less pronounced than in the summer. Small differences in soil temperature in the winter between the woodland catchment and the disturbed catchment were associated with the presence of snow cover and its capacity to yield thermal insulation. Good insulation of the soil from the atmosphere generated by snow cover yielded a very weak relationship between soil temperature and air temperature in the winter. In springtime the soil temperature increased the fastest on a south-facing slope in the disturbed catchment while in the autumn season, soil temperatures declined most rapidly on a slope facing north in the disturbed catchment. Key words: soil temperature; forest disturbance; windthrow; seasonal changes; Tatra Mountains Editor: Erika Gömöryová to an increase in evaporation and decline in transpira- 1. Introduction tion during low flow periods (Likens et al. 1970; Jewett The increase in air temperature observed across the Earth et al. 1995; Anderegg et al. 2012). However, according in the last few decades (Allen et al. 2018) is a cause of the to Gholami (2013), Hlásny et al. (2015) and Khaleghi rise in soil temperature. This pattern was examined in (2017), deforestation causes increase in peak discharge studies in the United States, where the soil temperature and runoff volume during rainfall events. Forest distur- at most of nearly 300 stations has shown a trend of warm- bance along with the soil warming leads to changes in ing in the last 35 years, with the average warming rate carbon cycling (Davidson & Janssens 2006; Allen 2009), at 0.38 °C per 10 years (Hu & Feng 2003). The increase as does nutrient cycling (Anderegg et al. 2012) and soil in soil temperature and the associated soil drought have respiration (Londo et al. 1999; Schlesinger & Andrews served as the cause of forest disturbance in many parts 2000; Ney et al. 2019). of the world over the last few decades (Allen et al. 2010; Most studies indicate that forest disturbance and Anderegg et al. 2012; Martinez-Vilalta 2012; Steinkamp deforestation cause a marked rise in soil temperature dur- et al. 2015). Forest disturbance as well as deforestation ing the summer months (Donnelly et al. 1991; Bhatti et drive an array of feedback mechanisms. For example, al. 2000; Hu et al. 2003; Moroni et al. 2009) and minor they generally lead to a continuous increase in the soil changes in the winter months (Donnelly et al. 1991; temperature primarily due to an increase in near-ground Hashimoto & Suzuki 2004; Moroni et al. 2009). How- solar radiation (Anderegg et al. 2012). The rise in soil ever, Whitson et al. (2005) measured distinctly higher temperature due to forest disturbance and deforestation triggers changes in the functioning of ecosystems. Water soil temperatures at harvested sites than at forest sites circulation patterns in catchments change mostly due during snowmelt. Iwahana et al. (2005) also argue that *Corresponding author. Joanna P. Siwek, e-mail: joanna.siwek@uj.edu.pl, phone: +48 126 645 277 J. P. Siwek / Cent. Eur. For. J. 67 (2021) 35–44 a neighboring undisturbed woodland catchment. In the clear-cutting enhances ground thawing and the depth of the soil active layer in areas underlain by permafrost, for course of the research a unique opportunity presented example in the eastern Russian region of Siberia. Snow itself. It became possible to directly examine the effects of forest disturbance on soil temperature in the same catch- cover also plays an important role in seasonal changes in ment. Spruce stands in a woodland catchment experi- soil temperature. The depth of snow together with snow thermal conductivity play a key role in snow thermal enced an attack of the bark beetle in 2018. In the following year the stands were partly overturned by high winds. insulation, as noted by Iwahana et al. (2005), Mellander et al. (2005) and Hu et al. (2013). Work by Mellander et al. (2005) showed that the nature of tree stands deter- mines the thickness of snow cover in a woodland area. For 2. Material and methods example, an open tree stand together with a low surface 2.1. Study area and meteorological conditions area covered by its canopy ultimately result in thicker layers of snow. The research study was conducted in two small sub- The purpose of the present study was to investigate catchments: (1) disturbed catchment, and (2) control the effect of forest disturbance triggered by natural causes woodland catchment (Fig. 1). Both sub-catchments are – such as a hurricane-force wind – on seasonal changes in located in the Kościeliski Potok catchment area in the soil temperature in a mountain region where tree stands Polish part of Western Tatra Mountains, which is pro- are dominated by spruce. An increase in soil temperature tected as Tatra National Park. The woodland catchment, after forest disturbance was expected; however, the key called the Kończysta Turnia, has a surface area of 14.1 ha, questions were the following: (i) What was the magni- while the disturbed catchment, called Pośrednia Kopka, tude of the increase? (ii) How does slope exposure affect has a surface area of 14.4 ha. The woodland catchment this magnitude? These are important questions today, as is located at elevations ranging from a minimum of 968 forest disturbance may increase soil warming caused by to a maximum of 1,264 m. The disturbed catchment is global climate warming. Despite the availability of sev- found at elevations in the range from 940 m to 1,200 m. eral global soil surface temperature databases based on The mean gradient of the woodland catchment is 30.1°, satellite products, according to Cassardo et al. (2018), and that of the disturbed catchment is 23.6° (Żelazny et there is still a lack of observational data on soil tempera- al. 2018). Both catchments are fragmented by narrow, ture profiles, especially for the root layer zone. deep V–shaped valleys, and are formed of sedimentary Initial research work focused on a comparison of soil rocks: limestone, sandstone, conglomerates. These par- temperature in a catchment where forest was disturbed ent material rocks are covered with Rendzic Leptosols by hurricane-force winds in 2013 and soil temperature in (Skeletic) and Haplic Cambisols (Eutric) (Skiba et al. Fig. 1. Study area in 2008 and 2014. 36 J. P. Siwek / Cent. Eur. For. J. 67 (2021) 35–44 2015). These soils are most often 40 to 60 cm deep. Soils in the study period occurred in late 2015 and early 2016 found in the woodland catchment are characterized by as well as late 2019 and early 2020 (Fig. 2a). According the occurrence of horizon O (1–5 cm), while soils found to Ustrnul et al. (2015), the mean annual atmospheric in the disturbed catchment lack this horizon (Żelazny et precipitation total in the study area ranges from 1,200 al. 2018). In both catchments the mineral horizons are mm at lower elevations to 1,600 mm at high elevations. formed of silt loam (USDA classification); however, the In the study period the highest atmospheric precipita- A horizon is characterized by a higher proportion of sand tion totals were recorded in the summer months (Fig. than lower horizons (Table 1). 2b). Snow cover appeared usually in October or Novem- Table 1. Texture and carbon (C) content in mineral horizons of the soils in the studied disturbed and woodland catchments – data source: Żelazny et al. (2018). Disturbed catchment Woodland catchment Depth C Sand Silt Clay Depth C Sand Silt Clay Soil horizon Texture* Soil horizon Texture* [%] [%] [cm] [cm] A 0–4 6.89 24 64 12 Silt loam A 0–6 6.66 29 60 11 Silt loam AB 4–19 3.98 8 74 19 Silt loam B 6–40 1.45 0 83 17 Silt loam BCa 19–40 2.19 0 82 18 Silt loam BCa 40–60 1.85 0 84 16 Silt loam *according to United States Department of Agriculture. ber and melted in April or May. The largest snow depth The study area is located in a temperate climate zone – was noted usually in December, January, and February, in a moderately cool climate zone (Hess 1965). According reaching usually more than 20 cm. Exceptionally small to Żmudzka et al. (2015) the mean annual air tempera- snow cover depths were noted in the winter of 2015/2016 ture ranges from 4 °C at lower elevations to 6 °C at high (Fig. 2c). elevations. In the study period (2015–2020) the warmest Originally, the study area was occupied by Dentario months of the year were June, July, and August when the glandulosae-Fagetum, dominated by beech (Fagus syl- mean monthly air temperature exceeded 10 °C (Fig. 2a). vatica L.), fir (Abies alba L.), and spruce (Picea abies The coolest months of the year when the mean monthly L.). The trees were felled at the end of the 19th century air temperature decreased below 0 °C were usually December, January, and February. The warmest winters and monocultures of spruce were planted at the begin- Fig. 2. Meteorological characteristics of study period: monthly air temperature (a), precipitation (b) and snow depth (c). 37 J. P. Siwek / Cent. Eur. For. J. 67 (2021) 35–44 ning of the 20th century (Szwagrzyk et al. 2019). The were installed on a south-facing slope in the woodland catchment (WS site). The probes experienced failures share of spruce in the studied stands exceeded 90% several times during the study period, which explains while the share of fir and beech did not exceed 10% gaps in the data set used in the study. (Bodziarczyk et al. 2019). Until December 2013 about Meteorologic data were obtained from weather sta- 93% of the Kończysta Turnia catchment was covered tions run by the Institute of Meteorology and Water Man- with forest, while almost 100% of the Pośrednia Kopka agement in Poland. The weather stations were located in catchment was forestland (Fig. 1). The age of the for- the vicinity of the study areas examined. Air temperature est ranged from 85 to 150 years. In December 2013 the data measured at a height of two meters above ground Pośrednia Kopka catchment experienced hurricane-force level were obtained from the Polana Chochołowska sta- winds felling 97% of its tree stands (Fig. 1). The maxi- tion, which is located about 6 km away from the stud- mum hourly average wind velocity during the windthrow ied catchments, and found at 1,147 meters of elevation. event was 29 m/s (Strzyżowski et al. 2016). In the same Atmospheric precipitation data and snow cover data were time period, only 13% of the forest in the Kończysta Tur- obtained from the Kiry-Kościelisko station located about nia catchment became disturbed (Żelazny et al. 2018). 1 km away from the studied catchments at an elevation Some of the fallen trees lying across the hillslopes in the of 933 meters. lower and central parts of the Pośrednia Kopka catch- Monthly soil temperature analyses were performed ment area were removed in 2014–2015. Fir and beech only for months with a complete set of data for particular seedlings were planted in a part of this area in 2015. study site. The coefficient of correlation for mean daily Research work performed in 2018 showed that 23% of soil temperature data for the 4 study sites with respect the disturbed area of the Pośrednia Kopka catchment was to mean air temperature data was calculated only for experiencing a succession of bush vegetation and juve- months with a complete data set from all the study sites. nile trees (Żelazny et al. 2018). Vaccinium myrtillus L., The Pearson correlation coefficient ( r) was computed. Rubus idaeus L., Athyrium distentifolium, Sorbus aucu- Mean daily air and soil temperatures were calculated paria L., Calamagrostis villosa, Deschampsia flexuosa using row data. Statistical analyses were performed in L., Homogyne alpina L., and Oxalis acetosella L. are the Statistica 1.3 software. most common species, which have emerged in the dis- turbed area (Szwagrzyk et al. 2019). The same study also showed that in 2018 almost 50% of the forested area in the Kończysta Turnia catchment experienced an attack of 3. results the bark beetle. In 2019 a large percentage of trees on the The largest differences between the highest and the low- south-facing slope of the woodland catchment became est values of the soil temperature over the course of the felled by high winds present in the area. year were noted on the south-facing slope in the disturbed catchment (DS site), while the smallest were noted on the north-facing slope in the woodland catchment (WN 2.2. Methods of measurements site) (Fig. 3). In summer months (June to August) the Soil temperature probes (Decagon ECH2O 5TM) were soil temperature at the DS site usually exceeded 15 °C – it ranges from about 12.5 °C up to 20 °C in summer 2017. installed at two sites in each of the studied catchments. Soil temperatures were lower during the same period of The WN and WS sites located in the studied woodland time on the south-facing slope in the woodland catch- catchment represent north-facing and south-facing ment (WS site), where usually did not exceed 15 °C. Soil woodland hillslopes, respectively. The studied DN and DS sites located in the disturbed catchment represent temperatures in the summer on north-facing slopes in the north-facing and south-facing disturbed hillslopes, woodland and disturbed catchments usually ranged from 10 to 15 °C – exceeding the latter only in selected cases. respectively. Soil temperature probes were installed in Soil temperatures decreased below 5 °C in all the studied the mineral soil horizons at a depth of 20 cm. The 20 catchments in wintertime, but did not normally fall below cm depth was chosen to ensure that the probes will be installed in the mineral and not the organic horizons of 0 °C. Only the soil temperature at the DS site decreased the soil. The probes were placed at an elevation of roughly below 0 °C in the winter of 2015. Soil temperatures in 900 m (Fig. 1). The technical specifications of the Deca- the spring and autumn months usually ranged from 5 to 10 °C (Fig. 3). gon ECH2O 5TM probes: range from −40 to 60 ºC, reso- The smallest monthly differences in soil temperature lution at 0.1 ºC, ±1 ºC. Soil temperature measurements were collected every 10 minutes. The first probes were occurred in all the studied sites in the wintertime, as evi- installed in October 2015 on a north-facing slope in the denced by the monthly interquartile range and monthly woodland catchment (WN site) and on a south-facing maximum-minimum range (Fig. 3). Larger monthly dif- slope in the disturbed catchment (DS site). In April 2017 ferences in the soil temperature in spring, summer, and probes were installed on a north-facing slope in the dis- autumn in both catchments occurred on south-facing turbed catchment (DN site), while in June 2017 probes slopes (vs. north-facing), with the largest monthly dif- 38 J. P. Siwek / Cent. Eur. For. J. 67 (2021) 35–44 Fig. 3. Statistical characteristics of soil temperature at four studied sites for each month of the studied period (November 2015 – February 2020). The above statistics are based on raw data. ferences in soil temperature noted at the DS site. The rela- mean monthly soil temperature data (Table 3). tionship between mean daily soil and air temperatures for a given month was stronger for south-facing slopes (DS Table 2. The Pearson correlation coefficient (p ≤ 0.05) be- and WS sites) than north-facing slopes (DN and WN tween mean daily soil temperature and mean daily air temper- sites). The strongest relationship was noted for the DS ature for four studied sites: DS – south-facing slope of the dis- site and weakest for the WN site (Table 2). This relation- turbed catchment, DN – north-facing slope of the disturbed catchment, WS – south-facing slope of the woodland catch- ship was strongest for all sites in the summer months and weakest in winter. ment, WN – north-facing slope of the woodland catchment (ns – not significant). A comparison of soil temperature measured at the Year / month DS DN WS WN same time on south-facing slopes in the disturbed and 2017 / June 0.804 0.614 0.833 0.645 woodland catchments (DS site vs. WS site) showed that 2017 / July 0.845 0.644 0.758 0.443 in 2017 and early 2018 (January to March, lack of data 2017 / August 0.874 0.618 0.748 0.487 2017 / September 0.774 0.628 0.707 0.572 for DS for subsequent months) the soil temperature at 2017 / November 0.565 0.516 0.657 0.502 the DS site was much higher than at the WS site. Mean 2017 / December ns -0.392 ns −0.395 2018 / January 0.479 0.375 0.444 ns monthly soil temperatures in summer (June to August 2018 / February 0.765 ns 0.475 0.773 2017) were from 1.8 to 2.4 °C higher at the DS site than 2018 / March 0.571 ns ns −0.389 the WS site (Table 3). These differences were smaller in 2019 / February ns ns ns ns 2019 / March 0.556 ns 0.490 ns the winter. This pattern changed in 2019 when soil tem- 2019 / April 0.788 0.492 0.737 0.441 peratures were often higher at the WS site than the DS 2019 / May 0.830 0.795 0.802 0.599 2019 / June 0.665 0.525 0.599 0.386 site, especially in the summer (Fig. 4), and mean monthly 2019 / July 0.798 0.730 0.722 0.602 soil temperatures at both study sites were comparable 2019 / August 0.736 0.465 0.694 ns (Table 3). A comparison of soil temperatures measured at 2019 / September 0.613 0.606 0.652 0.538 2019 / October 0.508 0.421 0.700 0.441 the same time in both catchments on north-facing slopes 2019 / November ns 0.453 0.508 ns (DN site versus WN site) showed that soil temperatures 2020 / January ns ns ns −0.364 2020 / February −0.405 ns ns ns in the summer were higher at the DN site versus the WN site, while in the winter the situation was just the opposite In the springtime, soil temperatures first begin to – higher soil temperature values were noted for the WN site versus the DN site. This pattern was evidenced by increase at the DS site, with temperatures higher than 39 J. P. Siwek / Cent. Eur. For. J. 67 (2021) 35–44 5 °C already in March. At other sites, soil temperatures catchment. After partial disturbance on the south-fac- begin to rise later, in April. The situation changed at the ing slope of the woodland catchment in 2019, the mean WS site in 2019 when the soil temperature at this site monthly soil temperature increased and was comparable began to rise as quickly as that at the DS site (Fig. 3). A to that in the disturbed catchment in 2013 (Table 3). The decrease in temperature in late autumn and winter below soil temperature difference on the north-facing slopes of the woodland and disturbed catchments in summertime 5 °C is r fi st observed on north-facing slopes in the studied was smaller at about 1 °C (Table 3). Higher soil temper- disturbed catchment. atures in disturbed areas versus woodland areas in the summertime have been documented by many researchers Table 3. Mean monthly soil temperature [°C] at four studied in other parts of the world. The differences in mean soil sites: DS – south-facing slope of the disturbed catchment, DN – north-facing slope of the disturbed catchment, WS – south- temperatures described by these other researchers for facing slope of the woodland catchment, WN – north-facing the summer months between woodland and disturbed slope of the woodland catchment. areas do vary mostly due to differences in the depth at Year / month DS DN WS WN which each measurement was performed. According to 2017 / June 14.3 12.0 11.9 10.9 Paul et al. (2004) soil depth is the most important factor 2017 / July 14.8 13.3 12.8 12.2 2017 / August 16.1 13.9 14.3 13.3 aside from other key factors such as air temperature and 2017 / September 12.2 10.3 10.5 10.3 the surface area covered by tree canopies controlling soil 2017 / November 4.5 3.0 3.8 4.1 2017 / December 2.4 1.7 1.9 2.5 temperature. Moroni et al. (2009) studied soil tempera- 2018 / January 1.8 1.2 1.5 1.9 ture at a depth of 10 cm in Newfoundland in Canada and 2018 / Fabruary 1.6 1.1 1.0 1.6 2018 / March 2.4 0.8 0.9 1.3 found that the difference in the soil temperature between 2019 / January 2.3 1.5 1.9 2.5 a harvested area and woodland area in the summer sea- 2019 / Fabruary 2.1 1.3 1.4 2.0 2019 / March 4.0 1.1 2.9 2.2 son equals about 2 °C. Bhatti et al. (2000) showed in 2019 / April 7.3 2.9 6.2 4.5 northeastern Ontario in Canada that soil temperature at a 2019 / May 8.5 7.2 7.8 6.6 2019 / June 14.6 13.7 14.5 12.4 depth of 5 cm increases in the summer after harvesting by 2019 / July 14.4 13.1 14.4 12.5 4 to 6 °C. According to research by Hu et al. (2013) con- 2019 / August 15.7 14.2 15.6 13.7 ducted in a steppe area in Central Asia, the soil tempera- 2019 / September 13.2 11.0 12.6 11.2 2019 / October 10.6 7.6 9.9 8.8 ture at a depth of 10 cm was higher by 4 to 8 °C relative 2019 / November 7.5 5.3 6.8 6.5 to that in a woodland area in the warm season. Londo et 2019 / December 4.3 2.5 3.4 3.9 2020 / January 3.3 1.9 2.1 2.7 al. (1999) studied soil temperature in Texas in the United 2020 / Fabruary 2.9 1.4 2.1 2.1 States and found a marked increase in soil temperature after deforestation – the measurements were performed at a depth of 10 cm in the summer season. The increase in 4. Discussion soil temperature was larger after clearcutting than after The effect of forest disturbance on the soil temperature partial clearcutting (Londo et al. 1999). at 20 cm depth in the study area is observable first and According to Hashimoto & Suzuki (2004) a marked foremost via a rise in the soil temperature in the sum- increase in soil temperature following forest clear-cutting mer months. The mean soil temperature in the summer is associated with an increase in absorbed solar radiation. months on south-facing slopes in the disturbed catch- Soils in woodland areas are protected from heating not only by trees but also by the organic forest floor situated ment is 1.8 to 2.4 °C higher than that in the woodland Fig. 4. Relationship between soil temperature (raw data) at DS site and WS site (A), and DN site and WN site (B) taking into account all measurements made in 2017 (red points), 2018 (grey points), and 2019 (blue points). 40 J. P. Siwek / Cent. Eur. For. J. 67 (2021) 35–44 atop the mineral soil horizons. The forest floor acts as due to increased solar radiation in disturbed areas versus a thermal insulator (Bhatti et al. 2000). The presented woodland areas, as shown by larger diurnal changes in research has shown that an area without trees and an the soil temperature on a south-facing slope in a dis- organic layer, such as a disturbed catchment, heats up turbed catchment versus all other measurement sites more during the summer than a wooded area with an (unpublished data). organic layer that is 1–5 centimeters thick, as would be Soil temperatures rise in spring (>5°C), as early the case in a woodland area. The buffer effect of forest as March, in the disturbed catchment on south-facing on soil temperature is confirmed by lower coefficients of slopes. At all other measurement sites, soil temperatures correlation between soil temperature and air temperature begin to increase only in April. Hu et al. (2013) studied in a woodland catchment versus a disturbed catchment. soil temperature in Central Asia and observed that the The effect of forest disturbance on soil temperature spring thawing of soils occurs two weeks earlier in tree- measured at a depth of 20 cm in the winter months in the less areas (i.e. steppe) versus woodland areas. On the catchments examined in the present study was weaker other hand, Whitson et al. (2005) studied soil tempera- than that in the summer. Similar conclusions were tures in southwestern Canada and learned that land cover reached by Donnelley et al. (1991) for the northeastern (harvested sites vs. forested sites) does not determine the United States, Hashimoto & Suzuki (2004) for Japan, timing of the spring soil thaw. Instead, slope exposure and Moroni et al. (2009) for Newfoundland (Canada). is the determining factor: the soil profiles of southerly Small differences in soil temperature in the winter aspect began thawing in mid-April, while those with between the woodland and disturbed areas examined northerly aspect in the latter half of May (Whitson et in the present study were undoubtedly associated with al. 2005). The presented research has shown that both the presence of snow cover and its capacity for thermal land cover and slope exposure determine the timing of insulation. The thickness of the snow cover in the study increases in soil temperature in spring. On south-facing period in the winter months usually reached into the tens slopes in disturbed areas, the snow cover disappears of centimeters (Fig. 2c). The soil temperature is above more rapidly than that on north-facing slopes as well as freezing in winter months when mean daily air temper- on woodland slopes. Soil on a disturbed, south-facing atures fall below −5 °C and sometimes below −20 °C. slope becomes devoid of snow earlier, on the one hand, Short-term decline in soil temperature below 0 °C on a and is not yet covered by young grass and bushes on the south-facing slope in the disturbed catchment could be other. The lack of thermal insulation leads to increase the due to the negligible thickness of snow cover, less than absorbed solar radiation triggering a rise in soil tempera- 10 cm. It is possible that the snow simply melted on the ture. Mellander et al. (2005) assert that in years with thin south-facing slope in the disturbed catchment. Meng et snow cover there was the largest variation in the timing al. (1995) observed that soil frost penetrates deeply into of soil warming (when soil temperature reaches 5 °C) in deforested wind-exposed soil, e.g. soil without snow spring between stands of different age. In autumn, soil cover. The relatively good insulation provided by snow temperatures examined in the present study fall the earli- cover to the soil separating it from the weather above est on north-facing slopes in the disturbed catchment due yields only a poor correlation between soil temperature to both less absorbed solar radiation than on south-facing and air temperature in the winter – unlike that in the sum- slopes and the absence of the buffer effect of forest. mer months. According to Mellander et al. (2005), snow The presented research represents a small step depth together with snow thermal conductivity play an towards a better understanding that forest disturbance important role in the timing of soil warming in winter may lead to dangerous feedback loops in the atmos- and spring. phere, pedosphere, and biosphere. The increase in soil Forest disturbance leads to an increase in the differ- temperature caused by forest disturbance will result in ence in soil temperature over the course of the year. On an increased release of greenhouse gases from the soil south-facing slopes, this is caused by a very large increase into the atmosphere. Soil temperature is a very impor- in soil temperature in the summer; in the winter months, tant factor shaping methane emissions (Mikkela et al. soil temperatures are generally somewhat higher than 1995) and soil respiration – soil CO efflux (Davidson et those in the woodland catchment. However, on north- al. 1998; Subke et al. 2003; Uvarov et al. 2006; Tang et facing slopes, this is caused by a small rise in the soil al. 2008). For example, Hick Pries et al. (2017) noted an temperature in summer and a small decline in winter. increase in CO loss of 34% to 37% from the r fi st meter of The presented research has confirmed the assertion by the top mineral soil horizon with a rise of 4ºC in soil tem- Hu et al. (2013) that the forest buffers the soil tempera- perature. The increase in greenhouse gas emissions will ture by lowering the soil temperature during the warm induce an increase in air temperature, which will result season and increasing it during the cold season; however, in a further increase in soil temperature and further tree this is only true of north-facing mountain slopes. What is mortality. Ney at al. (2019) note that forest management puzzling is the phenomenon of higher soil temperatures practices enacted following a natural forest disturbance on south-facing slopes in winter in the disturbed catch- are becoming an increasingly important part of climate ment versus the woodland catchment. This is most likely change mitigation strategies. 41 J. P. Siwek / Cent. Eur. For. J. 67 (2021) 35–44 intensify soil warming caused by climate warming. More- 5. Conclusions over, forest disturbance will lead to dangerous feedback The effect of forest disturbance triggered by hurricane- mechanisms. The increase in soil temperature caused by force winds on soil temperatures at a 20 cm depth in the forest disturbance will cause increased emissions of CO Western Tatras in Poland is marked by an increase in the from the soil to the atmosphere, which will contribute to difference in the soil temperature over the course of the an increase in air temperature. This will result in a further year. This is due r fi st and foremost to a marked increase in increase in soil temperature and the decline of some tem- soil temperature during the summer months. The mean perature sensitive tree stands. In order to minimize the soil temperature in the summer months on south-facing soil warming effect triggered by forest disturbance, tree slopes in the studied disturbed catchment was 1.8 to seedlings should be planted as soon as possible after for- 2.4 °C higher than that in the studied woodland catch- est disturbance. Moreover, this type of situation should ment. On north-facing slopes, the difference in the soil be used to plant stands whose tree species composition temperature between the woodland catchment and dis- corresponds to the natural habitat in a given climate zone turbed catchment in the summer months was smaller at in the mountains. Natural forest stands are believed to about 1 °C. The presented research has confirmed the be more resistant to natural disasters such as hurricane- presence of the buffer effect of forest on soil temperature force winds or bark beetle invasions than spruce mono- in summer, which is manifested by lower coefficients of cultures frequently found in Central Europe. correlation between soil temperature and air temperature in the woodland versus disturbed catchment. The effect of forest disturbance on soil temperature Acknowledgement in the winter months is less pronounced than in the The research was financed by the Jagiellonian University in summer months. This is associated with the presence of Kraków, Institute of Geography and Spatial Management. The snow cover that acts as a thermal insulator. In the winter authors wish to thank Grzegorz Zębik for his helpful advice and months (December-February) snow cover depth usually review of the English language of the manuscript. exceeds 20 cm in the studied area. Good insulation of the soil from the atmosphere by snow cover leads to a very weak relationship between soil temperature and air tem- references perature in winter. The soil temperature on disturbed, north-facing slopes in the winter is lower than that on Allen, C. D., 2009: Climate-induced forest dieback: woodland, north-facing slopes. This suggests the pres- a n e s c a l a t i n g g l o b a l p h e n o m e n o n ? U n a s y l v a , ence of a buffer effect of forest on the soil temperature 231/232:43–49. that increases the soil temperature during the colder Allen, C. D., Macalady, A. K., Chenchouni, H., Bachelet, season and decreases it in the warmer season, on north- D., McDowell, N., Vennetier, M. et al., 2010: A global facing slopes. However, it is puzzling that a south-facing overview of drought and heat-induced tree mortal- slope in a disturbed catchment would exhibit slightly ity reveals emerging climate change risks for forests. higher soil temperatures in winter than those in a wood- Forest Ecology and Management, 259:660–684. land catchment. This is likely due to increased absorbed Allen, M. R., Dube, O. P., Solecki, W., Aragón-Durand, solar radiation on south-facing slopes in disturbed areas F., Cramer, W., Humphreys, S. et al., 2018: Fram- versus woodland areas. However, this finding requires ing and Context. In: Masson-Delmotte, V., Zhai, P., further research. Pörtner, H-O., Roberts, D., Skea, J., Shukla, P.R. The research has shown that land cover (i.e. forest vs. et al. (eds): Global Warming of 1.5 °C. An IPCC lack of forest) and slope exposure (south versus north) Special Report on the impacts of global warming of determine the timing of increases in soil temperature in 1.5 °C above pre-industrial levels and related global spring and timing of soil temperature declines in autumn. greenhouse gas emission pathways, in the context In the spring season, soil temperatures increase the fast- of strengthening the global response to the threat of est (in March) on disturbed, south-facing slopes due to climate change, sustainable development, and efforts faster thawing of snow and absence of any vegetation to eradicate poverty. Intergovernmental Panel on Cli- such as young grasses or bushes. Given a low level of ther- mate Change. mal insulation, one may expect increased absorbed solar Anderegg, W. R. L., Kane, J. M., Anderegg, L. D. L., radiation that leads to a rapid increase in soil tempera- 2013: Consequences of widespread tree mortality triggered by drought and temperature stress. Nature ture. In the autumn season, temperatures on the studied Climate Change, 3:30–36. disturbed, north-facing slopes decrease at an earlier date Bhatti, J., Fleming, R. L., Foster, N. W., Meng, F-R., than elsewhere due to less absorbed solar radiation than Bourque, C. P. A., Arp, P., 2000: Simulations of on south-facing slopes as well as the lack of the buffer pre- and post-harvest soil temperature, soil mois- effect of forest cover. ture, and snowpack for jack pine: comparison with The present study shows that an increase in soil tem- e fi ld observations. Forest Ecology and Management, perature triggered by forest disturbance may signic fi antly 138:413–426. 42 J. P. Siwek / Cent. Eur. For. J. 67 (2021) 35–44 Bodziarczyk, J., Szwagrzyk, J., Zwijacz-Kozica, T., Zięba, Likens, G. E., Bormann, F. H., Johnson, N. M., Fisher, A., Szewczyk, J., Gazda A., 2019: The structure of D. W., Pierce, R. S., 1970: Effects of forest cutting forest stands in the Tatra National Park: The results and herbicide treatment on nutrient budgets in the of 2016–2017 inventory. Forest Research Papers, Hubbard Brook Watershed-Ecosystem. Ecological 80:13–21. Monographs, 40:23–47. Cassardo, C., Park, S. K., Sungmin, O., Galli, M., 2018: L o n d o , A . J . , M e s s i n a , M . G . , S c h o e n h o l t z , S . H . , Projected Changes in Soil Temperature and Surface 1999: Forest harvesting effects on soil temperature, Energy Budget Components over the Alps and North- moisture, and respiration in a bottomland hardwood ern Italy, Water, 10:1–18. forest. Soil Science Society of America Journal, 63:637–644. Davidson, E. A., Janssens, I. A., 2006: Temperature sen- Martínez-Vilalta, J., Lloret, F., Breshear, D. D., 2012: sitivity of soil carbon decomposition and feedbacks Drought-induced forest decline: causes, scope and to climate change. Nature, 440:165–173. implications. Biology Letters, 8:689–691. Donnelly, J. R., Shane, J. B., Yawney, H. W., 1991: Har- Mellander, P. E., Laudon, H., Bishop, K., 2005: Model- vesting causes only minor changes in physical prop- ling variability of snow depths and soil temperatures erties of upland Vermont soil. Northern Journal of in Scots pine stands. Agricultural and Forest Meteo- Applied Forestry, 8:33–36. rology, 133:109–118. Gholami, V., 2013: The inu fl ence of deforestation on run - Meng, F-R., Bourque, C-A., Jewett, K., Daugharty, off generation and soil erosion (Case study: Kasilian D., Arp, P. A., 1995: The Nashwaak Experimental Watershed). Journal of Forest Science, 59:272–278. Watershed Project: analysing effects of clearcutting Hashimoto, S., Suzuki, M., 2004: The impact of for- on soil temperature, soil moisture, snowpack, snow- est clear-cutting on soil temperature: a comparison melt and stream flow. Water, Air and Soil Pollution, between before and after cutting, and between clear- 82:363–374. cut and control sites. Journal of Forestry Research, Moroni, M., Carter, P., Ryan, D., 2009: Harvesting and 9:125–132. slash piling affect soil respiration, soil temperature, Hess, M., 1996: Klimat. In: Mirek, Z., Głowaciński, Z., and soil moisture regimes in Newfoundland boreal Klimek, K., Piękoś-Mirkowa, H. (eds.): Przyroda forests. Canadian Journal of Soil Science, 89:343– Tatrzańskiego Parku Narodowego, Tatry i Podta- trze 3. Tatrzański Park Narodowy. Kraków–Zako- Ney, P., Graf, A., Bogena, H., Diekkrüger, B., Drüe, C., pane, p. 53–68. Esser, O. et al., 2019: CO u fl xes before and after par - Hlásny, T., Kočický, D., Maretta, M., Sitková, Z., Barka, tial deforestation of a Central European spruce forest. I., Konôpka, M., Hlavatá, H., 2015: Effect of defor- Agricultural and Forest Meteorology, 274:61–74. estation on watershed water balance: hydrological Paul, K. I., Polglase, P. J., Smethurst, P. J., O’Connell, modelling-based approach. Central European For- A. M., Carlyle, C. J., Khanna, P. K., 2004: Soil tem- estry Journal, 61: 89–100. perature under forests: a simple model for predict- Hu, G., Liu, H., Anenkhonov, O. A., Korolyuk, A. Y., ing soil temperature under a range of forest types. Sandanov, D. V., Guo, D., 2013: Forest buffers soil Agricultural and Forest Meteorology, 121:67–182. temperature and postpones soil thaw as indicated by Schlesinger, W. H., Andrews, J. A., 2000: Soil respira- a three-year large-scale soil temperature monitoring tion and the global carbon cycle. Biogeochemistry, in the forest-steppe ecotone in Inner Asia. Global and 48:7–20. Planetary Change, 104:1–6. Skiba, S., Koreň, M., Drewnik, M., Kukla, J., 2015: Soils. Hu, Q., Feng, S., 2003: A daily soil temperature data- In: Dąbrowska, K., Guzik, M. (eds.): Atlas of the Tatra set and soil temperature climatology of the contigu- Mts. – Abiotic Nature. Tatrzański Park Narodowy. ous United States. Journal of Applied Meteorology, Zakopane, plate number: VI.1. 42:1139–1156. Steinkamp, J., Hickler, T., 2015: Is drought-induced for- Iwahana, G., Machimura, T., Kobayashi, Y., Fedorov, est dieback globally increasing? Journal of Ecology, A., Konstantinov, P., Fukuda, M., 2005: Influence of 103:31–43. forest clear-cutting on the thermal and hydrological Strzyżowski, D., Fidelus-Orzechowska, J., Żelazny, M., regime of the active layer near Yakutsk, Eastern Sibe- 2016: Geomorphological changes within a hillslope ria. Journal of Geophysical Research, 110:G02004. caused by a windthrow event in the Tatra Mounta- Jewett, K., Daugharty, D., Krause, H., Arp, P., 1995: ins, Southern Poland. Geogras fi ka Annaler. Series A, Watershed responses to clear-cutting: effects on Physical Geography, 98:347–360. soil solutions and stream water discharge in central Szwagrzyk, J., Bodziarczyk, J., Pielech, R., Zieba, A., New Brunswick. Canadian Journal of Soil Science, Zwijacz-Kozica, T., 2019: The impact of wind and 75:475–490. bark beetle outbreaks and protective measures on Khaleghi, M. R., 2017: The influence of deforestation the vegetation of spruce forests in the Tatra National and anthropogenic activities on runoff generation. Park. Parki Narodowe i Rezerwaty Przyrody, 38: Journal of Forest Science, 63:245–253. 57–68. 43 J. P. Siwek / Cent. Eur. For. J. 67 (2021) 35–44 Ustrnul, Z., Walawender, E., Czekierda, D., Šťastný, P., Żelazny, M., Adamska, M., Ciemborowicz, M., Fidelus- Lapin, M., Mikulová, K. 2015: Precipitation and snow Orzechowska, J., Gus, M., Jaśkowiec, B. et al., 2018: cover. In: Dąbrowska, K., Guzik, M. (eds.): Atlas of Ocena wpływu wielkoobszarowych wiatrołomów the Tatra Mts. – Abiotic Nature. Tatrzański Park Nar- n a r e ż i m h y d r o c h e m i c z n y i d e n u d a c j ę z l e w n i odowy. Zakopane, plate number: II.3. położonych w obszarach leśnych na terenach gór- Whitson, I., Chanasyk, D., Prepas, E., 2005: Effect of skich (Tatry Polskie) – etap 4. Kraków, Raport forest harvest on soil temperature and water storage K/KDU/000494, 395 p. and movement patterns on boreal plain hillslopes. Żmudzka, E., Nejedlík, P., Mikulová, K., 2015: Tem- Journal of Environmental Engineering and Science, perature, thermal indices. In: Dąbrowska, K., Guzik, 4:429–439. M. (eds.): Atlas of the Tatra Mts. – Abiotic Nature. Tatrzański Park Narodowy. Zakopane, plate num- ber: II.2.

Journal

Forestry Journalde Gruyter

Published: Mar 1, 2021

Keywords: soil temperature; forest disturbance; windthrow; seasonal changes; Tatra Mountains

There are no references for this article.