Abstract

Shoot water relations were studied in Siberian larch (Larix sibirica Ledeb.) trees growing at the borderline between taiga and steppe in northern Mongolia. Larix sibirica is the main tree species in these forests covering 80% of Mongolia’s forested area. Minimum shoot water potentials (Ψm) close to the point of zero turgor (Ψ0) repeatedly recorded throughout the growing season suggest that the water relations in L. sibirica were often critical. The Ψm varied in close relation to the atmospheric vapor pressure deficit, whereas Ψ0 was correlated with monthly precipitation. Young larch trees growing at the forest line to the steppe were more susceptible to drought than mature trees at the same sites. Furthermore, isolated trees growing on the steppe exhibited lower Ψm and recovered to a lower degree from drought overnight than the trees at the forest line. Indications of drought stress in L. sibirica were obtained in two study areas in Mongolia’s forest-steppe ecotone: one in the mountain taiga of the western Khentey in northernmost Mongolia, the other in the forest-steppe at the southern distribution limit of L. sibirica on Mt. Bogd Uul, southern Khentey. Larix sibirica growing in riverine taiga with contact to the groundwater table was better water-supplied than the larch trees growing at the forest line to the steppe. Larch trees from the interior of light taiga forests on north-facing slopes, however, exhibited more critical water relations than the trees at the forest line. Frequent drought stress in mature trees and even more in young larch trees at the forest-steppe borderline suggests that L. sibirica does not have the potential to encroach on the steppe under the present climate, except in a sequence of exceptionally moist and cool years. A regression of the present borderline between forest and steppe is likely to occur, as average temperatures are increasing everywhere and precipitation is decreasing regionally in Mongolia’s taiga forest region. Higher stomatal conductance concomitant to lower Ψm in trees of northern-slope forests compared to trees from the forest line to the steppe may be the result of a recent increase in drought intensity that affects better drought-adapted trees at the forest edge less than the trees in the forest interior. We conclude that drought is a key factor explaining the forest-steppe borderline in northern Mongolia. The proportion of forests within the present vegetation pattern of forests on north-facing slopes and the grasslands on south-facing slopes in Mongolia’s forest-steppe ecotone is not likely to increase under the present climate, but may decrease with increasing aridity due to global warming.

Introduction

Northern Eurasia is covered by the huge belt of boreal coniferous forests. In Eurasia’s oceanic parts, the boreal forest gives way to broadleaved forests in the south, whereas steppe grasslands are located south of it in continental Eurasia, i.e., eastern Siberia, Kazakhstan and Mongolia (Lavrenko and Karamysheva 1993). While the transition from the boreal forest to the temperate deciduous forests in oceanic areas with high precipitation is caused by the inferiority of conifers to the competition of broadleaved trees with increasing length of the growing season, drought limits the boreal forest at its southern distribution limit in the center of the Eurasian landmasses (Walter and Breckle 1994). The ecotone between the northern coniferous forests and the steppe is an aspect-dependent vegetation pattern of forests on north-facing slopes and grasslands on south-facing slopes (Wallis de Vries et al. 1996, Dulamsuren et al. 2005a, Chytrý et al. 2008). The southern distribution limit of the individual tree species occurring in this ecotone and, with it, the entire forest-steppe border migrates depending on the variations of precipitation and temperature (Dinesman et al. 1989, Gunin et al. 1999, Miehe et al. 2007). Also, anthropo-zoogenic influences affect the position of the forest-steppe borderline (Hilbig 1995, Rösch et al. 2005, Sankey et al. 2006).

The temperature increase during the past decades at the forest-steppe transition in southern Siberia, Kazakhstan and Mongolia is far beyond the global average (Batima et al. 2005, Henderson et al. 2006), whereas precipitation trends differ between regions (Aizen et al. 2001, Giese and Moßig 2004). In parts of northern Mongolia, precipitation has significantly decreased in the late 20th century (Batima et al. 2005), and a further decline is predicted for the 21st century (Sato et al. 2007). Since forests already today occur at their drought limit in such areas with decreasing precipitation and increasing temperature, it is reasonable to assume that the increasing aridity will in future reduce the area of forests in the present forest-steppe mosaics.

Ecophysiological studies examining the current extent of drought stress in trees at forest-steppe ecotones are rare. This is also true for Mongolia, which harbors one of the world’s largest and the least disturbed forest-steppe transition zones (Vostokova and Gunin 2005). In the forest-steppe of northeastern Mongolia, Li et al. (2005a) had investigated the carbon balance in a stand of Siberian larch (Larix sibirica Ledeb.), which is Mongolia’s most common tree species (Gunin et al. 1999), in comparison to the nearby steppe grassland (Li et al. 2005b). Li et al. (2006, 2007a) applied a stable isotope approach for identifying water sources used by L. sibirica in Mongolia’s forest-steppe ecotone. Direct study, as to how L. sibirica or any other tree species occurring in Mongolia’s forest-steppe ecotone respond to the seasonal shortages of water, is limited to a single investigation of δ13C signatures in a L. sibirica stand in northeastern Mongolia (Li et al. 2007b). Therefore, we investigated the seasonal and diurnal dynamics of shoot water relations of L. sibirica trees growing at the forest-steppe borderline of northern Mongolia and supplemented these analyses with punctual measurements of CO2/H2O gas exchange and chlorophyll fluorescence characteristics. The aim was to assess the potential of the larch trees to encroach onto grasslands in front of the present forest lines under the current climate and to withstand increasing aridity in the future.

Special emphasis was given to the hypotheses that (1) trees at the forest line to the steppe suffer from drought stress during summer and (2) young larch trees at the forest edge are more susceptible to drought stress than mature trees. Other research questions addressed included a comparison of the water relations in larch trees growing at the forest line to the steppe and in the interior of the taiga forest occupying the north-facing slopes. Water relation in L. sibirica growing in a riverine forest was also punctually studied. While most parts of the study were carried out in the mountain taiga of the western Khentey Mountains in northernmost Mongolia, some comparative measurements of the shoot water potential were also carried out in the forest-steppe of the southern Khentey Mountains immediately at the southern distribution limit of the Eurosiberian taiga to the open steppe.

Materials and methods

Study sites

Investigations were conducted in two regions of northern Mongolia. The first study area, where most measurements were carried out, was located near Khonin Nuga Research Station in the western Khentey Mountains (49°04′48″ N and 107°17′15″ E), 130 km north of the Mongolian capital Ulan Bator. The second study area was selected in the southern Khentey at Mt. Bogd Uul, 12 km southeast of Ulan Bator. The study areas differ in their latitudinal position and vegetation. Khonin Nuga is located in the northernmost part of north-eastern Mongolia close to the Russian border, whereas the Bogd Uul mountain range forms the southernmost distribution limit of forests in north-eastern Mongolia. Khonin Nuga belongs to the mountain taiga, where steppe occurs as insular outposts on south-facing slopes surrounded by woodlands on north-facing slopes and in valleys (Dulamsuren et al. 2005a). Mt. Bogd Uul is located in the forest-steppe belt, which is principally dominated by grasslands with islands of light taiga forest limited to north-facing slopes (Hilbig et al. 2004).

Long-term climate data from the vicinities of the study areas are available from the weather station Eroo, ca. 70 km northwest Khonin Nuga Research Station, and from the airport in Ulan Bator, ca. 25 km west northwest of the sample plots on Mt. Bogd Uul (Table 1). Microclimate measurements from nine weather stations near Khonin Nuga Research Stations from spring 2005 to fall 2007 suggest similar temperatures as recorded at the weather station Eroo (Dulamsuren and Hauck 2009). However, precipitation is too variable in Mongolia to deduce long-term annual means from the weather data recorded from 2005 to 2007. The forest vegetation at Khonin Nuga suggests that the annual precipitation is higher than that at the weather stations Eroo and Ulan Bator, as the latter are located in the forest-steppe which is strongly dominated by the drought-tolerant L. sibirica, whereas the forests of Khonin Nuga, located in the mountain taiga, are more diverse and include moisture-demanding tree species, like Abies sibirica Ledeb. and Sorbus sibirica Hedl. (Dulamsuren et al. 2005a). The maximum precipitation is received in summer.

Table 1.

Climate data from the weather stations Eroo (ca. 70 km northwest of Khonin Nuga) and Ulan Bator (ca. 25 km west northwest of Tor Khurkh, Mt. Bogd Uul). Arithmetic mean ± SE; ranges of annual mean temperature or annual precipitation in brackets.

 Eroo (ca. 900 m) Ulan Bator (ca. 1365 m) 
Study period 1961–2004 1950–2006 
Annual mean temperature (°C) −2.1 ± 0.2 −1.8 ± 0.2 
Mean January temperature (°C) −27.4 ± 0.5 (−4.0 to 3.2) −23.9 ± 0.4 (−31.8 to −18.3) 
Mean July temperature (°C) 18.7 ± 1.3 (16.4 to 22.3) 17.2 ± 0.2 (14.3 to 21.9) 
Precipitation (mm) 277 ± 11 (157 to 459) 259 ± 10 (62 to 404) 
January precipitation (mm) 3 ± 1 (0 to 19) 2 ± 0 (0 to 8) 
July precipitation (mm) 78 ± 5 (30 to 160) 71 ± 5 (7 to 197) 
 Eroo (ca. 900 m) Ulan Bator (ca. 1365 m) 
Study period 1961–2004 1950–2006 
Annual mean temperature (°C) −2.1 ± 0.2 −1.8 ± 0.2 
Mean January temperature (°C) −27.4 ± 0.5 (−4.0 to 3.2) −23.9 ± 0.4 (−31.8 to −18.3) 
Mean July temperature (°C) 18.7 ± 1.3 (16.4 to 22.3) 17.2 ± 0.2 (14.3 to 21.9) 
Precipitation (mm) 277 ± 11 (157 to 459) 259 ± 10 (62 to 404) 
January precipitation (mm) 3 ± 1 (0 to 19) 2 ± 0 (0 to 8) 
July precipitation (mm) 78 ± 5 (30 to 160) 71 ± 5 (7 to 197) 

Studies in the western Khentey were carried out on Mt. Bayantogol (1020–1180 m, 49°5′ N and 107°17′ E) north of Khonin Nuga Research Station. The south-facing slope of Mt. Bayantogol rises directly from the northern shore of the River Eroo and has a size of ca. 1.5 km × 250 m (parallel × perpendicular to the contour lines). This south-facing slope is covered with different communities of meadow steppe and mountain steppe as well as small savanna-like woodlands of Ulmus pumila L. (Dulamsuren et al. 2005a, 2005b). On its north-facing slope, Mt. Bayantogol is covered with a light taiga forest of Betula platyphylla Sukaczev and L. sibirica with a canopy cover between 20% and 40%. The borderline between the forest and the grasslands at the mountain ridge is almost exclusively built by L. sibirica.

Shoot water potential measurements on Mt. Bogd Uul were conducted in the north-eastern part of the mountain system that is northwest of the Tor Khurkh Valley on a sample plot of 1950 m altitude (Bayan Tokhom; 47°43′ N and 107°5′ E). Meadow steppe with single larch trees growing in strips of ca. 50 m from the forest line occurred on southeast-facing slopes. The light taiga forests of Mt. Bogd Uul solely consist of L. sibirica. The forested slope studied was exposed to the east. The bases of the studied slopes are high valleys within the Bogd Uul mountain system.

Measurements of the shoot water potential

The plant water status was assessed by conducting diurnal and seasonal measurements of the shoot water potential (Ψ) on rainless days with clear sky using a Model 1000 Pressure Chamber Instrument (PMS Instrument Company, Albany, Oregon) applying the method of Scholander et al. (1964). Measurements included the determination of the predawn water potential (Ψp), which represents the daily maximum of Ψ, and of the midday water potential (Ψm), which equals the daily minimum of Ψ, in the field. These measurements were conducted with freshly cut twigs of 10 cm length from sun-exposed branches at a height of 2 m. Five replicates were studied per tree at a specific point in time. In addition, pressure–volume curves were recorded from fully water-saturated twigs, sampled the day (≤ 12 h) before the measurement to determine the point of zero turgor (Ψ0) (Roberts et al. 1980). The Ψ0 is the maximum value of Ψ, from which on irreversible damage due to cytorrhysis is possible. However, the cells can withstand values of Ψ below Ψ0 if the stability of the cell wall allows negative turgor values (Rhizopoulou 1997). Twigs used for the determination of Ψ0 were re-cut underwater and saturated with water in a glass wrapped in plastic foil for 12 h. The Ψ0 is less affected by the length of the saturation phase before the measurements (Parker and Pallardy 1987). The Ψ0 was deduced from plots of −1/Ψ versus the relative water content (RWC) of the sample. The RWC was calculated following RWC (%) = [(fresh mass − dry mass)/(turgid mass − dry mass)] × 100. Three twigs per tree were sampled as pseudoreplicates for the determination of Ψ0.

Water potential measurements on Mt. Bayantogol included monthly measurements of Ψp, Ψm and Ψ0 throughout two subsequent growing seasons from May to September 2005 and June to September 2006. The Ψp and Ψm were determined in five mature trees (ca. 60–80-years old, 26 ± 5 cm in diameter at breast height) and three young trees (stem diameter 3.5 ± 0.3 cm, stem height between 60 and 100 cm). The 60–80-year-old trees represented the most common age class in the forest (Dulamsuren and Hauck unpublished). The five mature larch trees stood 15, 35, 45 or 50 m in front of or directly at the forest line. The three young trees grew 25, 35 or 40 m in front of the forest line on the montane meadow steppe. The Ψ0 was determined each in one mature and young tree growing on the meadow steppe at a distance of 15–30 m from the forest line.

In addition to the monthly measurements of Ψp and Ψm, the diurnal variation of Ψ was measured on selected days in the middle of the growing season. These measurements included mature and young trees at the forest edge as well as a comparison between mature larch trees growing on the steppe and in the neighboring floodplain of a small creek. They were carried out at Mt. Baziin Am at 1050 m (49°2′ N and 107°15′ E), as the valley bottom below Mt. Bayantogol is deforested. During the measurements involving the floodplain, an ML-2x Theta soil moisture sensor (Umweltanalytische Produkte, Ibbenbüren, Germany) was used together with an HH2 portable logger to measure the volumetric soil water content in the upper 6 cm of the soil covering a soil volume of 75 cm3. Furthermore, Ψm was determined in each of the five mature larch trees growing at the forest line or in the forest interior (120–140 m behind the forest line) on Mt. Bayantogol in August 2006.

In the Bogd Uul mountain range, Ψ was measured in the two warmest months of the growing season in July and August 2007. Diurnal courses, including Ψp and Ψm, were recorded from five trees growing in the steppe up to 50 m in front of the forest line and five trees growing in the forest up to 50 m behind the forest line. The Ψ0 was determined in the laboratory for four to five trees per month on the steppe and in the forest.

Determination of δ13C signatures

The δ13C signature of needles of mature trees of L. sibirica was determined in samples from the western Khentey Mountains. Samples were taken from each of the five trees growing in the meadow steppe, 15–45 m in front of the forest line, or growing in the forest interior, 120–140 m behind the forest line. Five additional samples were taken from trees growing at the lower part of a north-facing slope in the valley of the River Eroo, opposite Mt. Bayantogol (970 m; 49°5′ N and 107°18′ E). These trees were sampled as an additional reference, as the site is only a little above the level of the floodplain, which was thought to be better water-supplied than that in the stand on Mt. Bayantogol. Sun-exposed needles were collected at 2 m height above the ground from the southward side of the trees in August 2006. The samples were dried at 105 °C for 24 h, ground to a fine powder, and weighed out in tin capsules. The analyses were conducted with a Delta V Advantage isotope ratio mass spectrometer (Thermo Fisher Scientific, Waltham, MA), which was combined with an NA 1500 C/N Elementar Analyzer (Carlo Erba Strumentazione, Milan, Italy) and Conflo III interface (Thermo Fisher Scientific). Acetanilide was used as an internal standard. Using this internal standard, the δ13C signature was related to the Peedee belemnite limestone standard using δ13C (‰) = ([Rsample/Rstandard] −1) × 1000, with R = 12C/13C. The enrichment of 13C indicating drought stress results in high (less negative) values of δ13C.

Gas exchange measurements

Diurnal variation of CO2/H2O gas exchange of L. sibirica shoots was studied on Mt. Bayantogol in August 2006. Net photosynthesis, transpiration and stomatal conductance were measured under ambient conditions with an LCi Ultra Compact Photosynthesis System equipped with a standard conifer cuvette (ADC Bioscientific, Herts, UK). Gas exchange was measured in each three mature larch trees growing on the meadow steppe (15, 35 or 45 m) in front of the forest line, and in the forest interior, 120–140 m behind the forest line. Measurements were performed on 2-m high, sun-exposed branches. The measurements could not start earlier than 10:00 A.M., as needles were covered with a water film from nocturnal fog before.

Measurement and calculation of chlorophyll fluorescence parameters

The effective quantum yield (Φ2) of photochemical energy conversion in photosystem II (PSII) was measured with a Mini-PAM chlorophyll fluorometer (Walz Mess- und Regeltechnik, Effeltrich, Germany). Diurnal variation of Φ2 in the field was recorded on Mt. Bayantogol in August 2006 from five mature trees at the forest-steppe growing 15–45 m in front of or directly at the forest line, five young trees growing 25–40 m in front of the forest line and five mature trees growing in the interior of the forest 120–140 m behind the forest line. Five pseudoreplicate measurements were made per tree. In addition to Φ2, the photosynthetically active radiation (PAR) and the leaf temperature were measured with sensors integrated in the leaf clip holder of the Mini-PAM. The PAR data were used to calculate the electron transport rate (ETR) using ETR = Φ2 × PAR × 0.42. The non-photochemical quenching (NPQ) was calculated following the equation NPQ = (FM − FM′)/FM′ with FM being the absolute maximum fluorescence in the dark-adapted state (measured at predawn) and FM′ being the absolute maximum fluorescence in the light-adapted state (Roháček 2002).

Weather conditions during the study period

Data from an HOBO weather station (Onset, Bourne, MA) located in the meadow steppe along the forest edge on Mt. Bayantogol at an elevation of 1060 m are available for the growing seasons covered by the present study (Dulamsuren and Hauck 2009). Direct weather information from Mt. Bogd Uul is lacking, but the data from Mt. Bayantogol can be used to assess the general trends for interannual variation during the study years. The weather station was equipped with an H21 data logger and a set of HOBO Smart Sensors including instruments for measuring air temperature and relative air humidity at 150 cm above soil level (temperature/RH sensor S-THA), soil temperature at 1 and 15 cm depth (8 Bit temperature sensor S-TMA), volumetric soil water content at 10 cm depth (soil moisture sensor S-SMA) and PAR in horizontal position at 2 m above the ground (PAR sensor S-LIA). Precipitation at 1 m (rain gauge sensor S-RGA) was sampled at another HOBO weather station 3 km southwest of Mt. Bayantogol in the floodplain of the River Eroo at 910 m altitude. Data were recorded every 10 min and are the averages of 10 consecutive measurements. Snow was sampled in five 10-l buckets near the weather station in the floodplain meadow.

To characterize the weather conditions in the study periods, weather data of July 2005 and July 2006 are compiled in Table 2. In the western Khentey, the summer was warmer and drier in the first study year (2005) than in the second one (2006). The weather conditions during our measurements in the southern Khentey in summer 2007 were characterized by a dry July (36 mm of precipitation at Ulan Bator) and a humid August (72 mm).

Table 2.

July weather data for Mt. Bayantogol, western Khentey for 2005–2007. Arithmetic mean ± SE; absolute maxima or minima in brackets.

 2005 2006 2007 
Air temperature, mean (°C) 18.4 ± 0.4 17.0 ± 0.5 18.7 ± 0.7 
Air temperature, max. (°C) 27.5 ± 0.7 (37.4) 24.0 ± 0.7 (31.1) 27.8 ± 1.1 (37.9) 
Soil temperature, 1 cm, mean (°C) 24.4 ± 0.4 21.8 ± 0.5 23.1 ± 0.6 
Soil temperature, 1 cm, max. (°C) 38.1 ± 0.8 (45.4) 31.9 ± 1.1 (44.9) 35.0 ± 1.4 (49.6) 
Soil temperature, 15 cm, mean (°C) 21.4 ± 0.2 19.7 ± 0.2 20.7 ± 0.3 
Soil temperature, 15 cm, max. (°C) 23.5 ± 0.2 (25.2) 21.9 ± 0.3 (24.0) 22.9 ± 0.3 (26.0) 
Relative humidity, mean (%) 69.7 ± 1.8 81.1 ± 1.7 69.8 ± 2.6 
Relative humidity, min. (%) 35.2 ± 2.1 (15.8) 49.1 ± 2.5 (25.8) 35.9 ± 3.2 (8.8) 
Soil water content (% vol.) 6.2 ± 0.3 6.9 ± 0.2 7.1 ± 0.5 
Precipitation, sum (mm) 30 76 – 
Precipitation, frequency (days) 10 15 – 
PAR, sum (mol m−22.26 1.98 2.21 
 2005 2006 2007 
Air temperature, mean (°C) 18.4 ± 0.4 17.0 ± 0.5 18.7 ± 0.7 
Air temperature, max. (°C) 27.5 ± 0.7 (37.4) 24.0 ± 0.7 (31.1) 27.8 ± 1.1 (37.9) 
Soil temperature, 1 cm, mean (°C) 24.4 ± 0.4 21.8 ± 0.5 23.1 ± 0.6 
Soil temperature, 1 cm, max. (°C) 38.1 ± 0.8 (45.4) 31.9 ± 1.1 (44.9) 35.0 ± 1.4 (49.6) 
Soil temperature, 15 cm, mean (°C) 21.4 ± 0.2 19.7 ± 0.2 20.7 ± 0.3 
Soil temperature, 15 cm, max. (°C) 23.5 ± 0.2 (25.2) 21.9 ± 0.3 (24.0) 22.9 ± 0.3 (26.0) 
Relative humidity, mean (%) 69.7 ± 1.8 81.1 ± 1.7 69.8 ± 2.6 
Relative humidity, min. (%) 35.2 ± 2.1 (15.8) 49.1 ± 2.5 (25.8) 35.9 ± 3.2 (8.8) 
Soil water content (% vol.) 6.2 ± 0.3 6.9 ± 0.2 7.1 ± 0.5 
Precipitation, sum (mm) 30 76 – 
Precipitation, frequency (days) 10 15 – 
PAR, sum (mol m−22.26 1.98 2.21 

Statistics

Arithmetic mean ± SE is given throughout the paper. Data were tested for normal distribution with the Shapiro–Wilk test. Pairwise comparisons of means were made with Student’s t test. For repeated pairwise comparisons in the same data set (Ψm versus Ψ0), the t test was Bonferroni-corrected. Duncan’s multiple range test was used for multiple comparisons of the δ13C data. Statistical analyses were computed with SAS Version 6.04 software (SAS Institute Inc., Cary, NC).

Results

Mountain taiga of the western Khentey

Both in mature and in young larch trees growing at the forest line to the steppe, Ψm decreased throughout the growing season (Figure 1A, B). The Ψp decreased from June to August and recovered in September 2005, but did not decline before September in the cooler and moister year 2006. In general, field values of Ψ varied with the atmospheric vapor pressure deficit (VPD) (Figure 1C). Except for the early growing season, Ψ0 was correlated with the amount of precipitation in the current month in mature trees (Figure 1D). Linear regression of Ψ0 versus monthly precipitation for the months June to September in 2005 and 2006 yielded a significant relationship with r = 0.81 (P ≤ 0.001). The Ψ0 of the young trees, however, did not recover after drought in July 2005 with increasing precipitation in August and September 2005 (Figure 1D). Therefore, linear regression of Ψ0 in the young trees on monthly precipitation did not yield a significant relationship (r = 0.43, P = 0.14). The adjustment of Ψ0 to water availability was sufficient in mature trees throughout the growing season in the moist year 2006, but not in 2005 when Ψ0 was not significantly different from Ψm in June, August and September (Figure 1A). In the young trees growing at the forest-steppe borderline, Ψm dropped to the level of Ψ0 throughout the dry growing season 2005 and in midsummer of the wetter year 2006 (Figure 1B).

Figure 1.

Seasonal and diurnal variation of the shoot water potential in L. sibirica growing at the forest line to the steppe at Mt. Bayantogol, western Khentey. (A, B) Minimum (Ψm) and predawn (Ψp) shoot water potentials as well as points of zero turgor Ψ0 in (A) mature (N = 5) and (B) young trees (N = 3) in the growing seasons 2005 and 2006. Asterisks indicate significant difference between Ψm and Ψ0 (P ≤ 0.05, Bonferroni-corrected t test). (C) Diurnal variation of the shoot water potential in a mature tree and VPD on August 12, 2005. (D) Variation of the point of zero turgor (Ψ0) with monthly precipitation in mature and young L. sibirica.

Figure 1.

Seasonal and diurnal variation of the shoot water potential in L. sibirica growing at the forest line to the steppe at Mt. Bayantogol, western Khentey. (A, B) Minimum (Ψm) and predawn (Ψp) shoot water potentials as well as points of zero turgor Ψ0 in (A) mature (N = 5) and (B) young trees (N = 3) in the growing seasons 2005 and 2006. Asterisks indicate significant difference between Ψm and Ψ0 (P ≤ 0.05, Bonferroni-corrected t test). (C) Diurnal variation of the shoot water potential in a mature tree and VPD on August 12, 2005. (D) Variation of the point of zero turgor (Ψ0) with monthly precipitation in mature and young L. sibirica.

Drought stress in trees forming the frontier to the steppe increased along with an increasing distance from the closed forest, though this relationship was only statistically significant in the drier growing season 2005 (Figure 2). In July 2005, Ψm in the tree advancing farthest into the steppe (at 50 m distance from the forest line) differed by −1 MPa from Ψm in a larch tree growing immediately at the forest line (Figure 2C). Furthermore, nocturnal recovery of water relations is increasingly hampered along with the increasing distance from the forest edge, as indicated by the Ψp values (Figure 2A). In July 2006, when twice as much precipitation was received as in the year before (Table 2), an insignificant trend for lower Ψm with increasing distance from the forest line occurred (Figure 2D), whereas Ψp was nearly independent of the position of the tree along the forest-steppe gradient (Figure 2D). As the young trees were more susceptible to drought than the mature ones (Figure 1B), the regression lines in Figure 2 refer only to the mature trees. The results in Figure 2 agree with the measurements of the diurnal variation of Ψ showing more strongly decreasing values of Ψ during daytime in mature and young L. sibirica growing in the meadow steppe, 30–45 m in front of the forest line, than in mature larch growing at the forest edge, only up to 15 m in front of the closed forest (Figure 3A).

Figure 2.

July values of predawn (Ψp) and minimum (Ψm) leaf water potentials in L. sibirica growing in the meadow steppe at different distances from the forest line on Mt. Bayantogol, western Khentey. (A) Ψp, 2005; (B) Ψp, 2006; (C) Ψm, 2005; (D) Ψm, 2006. Linear regression lines were calculated for mature trees only; dotted lines refer to non-significant relationships: (A) r = −0.86, P = 0.03; (B) r = −0.43, P = 0.24; (C) r = −0.92, P = 0.01; (D) r = −0.73, P = 0.08. Different scales for Ψp and Ψm on the ordinates should be noted.

Figure 2.

July values of predawn (Ψp) and minimum (Ψm) leaf water potentials in L. sibirica growing in the meadow steppe at different distances from the forest line on Mt. Bayantogol, western Khentey. (A) Ψp, 2005; (B) Ψp, 2006; (C) Ψm, 2005; (D) Ψm, 2006. Linear regression lines were calculated for mature trees only; dotted lines refer to non-significant relationships: (A) r = −0.86, P = 0.03; (B) r = −0.43, P = 0.24; (C) r = −0.92, P = 0.01; (D) r = −0.73, P = 0.08. Different scales for Ψp and Ψm on the ordinates should be noted.

Figure 3.

Diurnal variation of the shoot water potential in L. sibirica in the western Khentey. (A) Mature trees on the meadow steppe 0–15 m (N = 3) or 40–45 m (N = 2) in front of the forest line and young trees 30–40 m (N = 3) in front of the forest line on July 31, 2006. (B, C) Mature trees (N = 5) on the meadow steppe and in riverine forest on August 13, 2005 (B) and August 4, 2006 (C).

Figure 3.

Diurnal variation of the shoot water potential in L. sibirica in the western Khentey. (A) Mature trees on the meadow steppe 0–15 m (N = 3) or 40–45 m (N = 2) in front of the forest line and young trees 30–40 m (N = 3) in front of the forest line on July 31, 2006. (B, C) Mature trees (N = 5) on the meadow steppe and in riverine forest on August 13, 2005 (B) and August 4, 2006 (C).

Depending on the water supply of the forest stand, the shoot water status of larch trees growing in the interior of the light taiga forest was either better or worse than that of the trees growing on the steppe in front of the forest line. A comparison with L. sibirica growing in the floodplain of a small creek, where the roots most likely reached the groundwater table, showed higher values of Ψm in the forest than on the steppe (Figure 3B, C). Hourly measurements of soil moisture synchronously with the water potential measurements shown in Figure 3C exhibited soil water contents that were 10 times higher in the riverine forest than at the forest line to the steppe (24 ± 1% vol. versus 2.3 ± 0.2% vol.). Larch trees from the north-facing slope of Mt. Bayantogol had significantly lower values of Ψm (−1.75 ± 0.07 MPa) than trees at the neighboring forest edge (−1.27 ± 0.09 MPa; N = 5, t test, P ≤ 0.01). Stomatal conductance (Figure 4A) and transpiration (not shown) concurrently measured with the shoot water potential were higher in the forest interior on the north-facing slopes than at the forest line to the steppe. Despite the lower Ψm and the higher stomatal conductance in the larch trees of the forest interior, the rate of net photosynthesis under ambient conditions was not different between trees from the forest interior and the forest line (Figure 4B). The δ13C signatures were not significantly different (Duncan’s multiple range test, P ≤ 0.05) between larch trees from the forest edge (−29.08 ± 1.25‰ at Mt. Bayantogol) and the interior of forests from north-facing slopes (−27.35 ± 0.48‰ at Mt. Bayantogol and −26.56 ± 0.79‰ in the Eroo Valley), though there was an insignificant trend for higher δ13C values in the latter.

Figure 4.

Diurnal variation of (A) stomatal conductance and (B) net CO2 assimilation rate (N = 3) in L. sibirica growing at the forest line to the steppe and in the forest interior at Mt. Bayantogol, western Khentey on August 23, 2006.

Figure 4.

Diurnal variation of (A) stomatal conductance and (B) net CO2 assimilation rate (N = 3) in L. sibirica growing at the forest line to the steppe and in the forest interior at Mt. Bayantogol, western Khentey on August 23, 2006.

Chlorophyll fluorescence measured simultaneously with the photosynthetic gas exchange showed that the rapid and steep rise in PAR at the forest edge induced strong light quenching (NPQ) in the PSII antennae and thus caused Φ2 to deteriorate dramatically (Figure 5). Under high-light conditions prevailing most of the day, Φ2 of forest edge trees remained down-regulated at low values around 0.3 resulting in a tendentiously lower ETR compared to the trees from the forest interior that received less illumination and thus had to implement appreciably less NPQ (Figure 5). No difference was observed in the chlorophyll fluorescence characteristics between mature and young trees growing at the forest edge (Figure 5). When comparing fluorescence with gas exchange data, it is remarkable that between 10:00 and 17:00 h the rather stable ETR contrasts with the steadily declining CO2 assimilation rate (Figures 4C and 5D).

Figure 5.

Diurnal variation of chlorophyll fluorescence and climatic parameters in L. sibirica growing in the forest interior and at the forest line to the steppe (N = 5) on August 23, 2006. (A) Photosynthetically active photon flux density (PPFD), (B) leaf temperature, (C) Φ2, (D) ETR, (E) NPQ.

Figure 5.

Diurnal variation of chlorophyll fluorescence and climatic parameters in L. sibirica growing in the forest interior and at the forest line to the steppe (N = 5) on August 23, 2006. (A) Photosynthetically active photon flux density (PPFD), (B) leaf temperature, (C) Φ2, (D) ETR, (E) NPQ.

Forest-steppe of the southern Khentey

Diurnal courses of Ψ on Mt. Bogd Uul showed no significant difference between larch trees growing in the light taiga forest and in the grasslands at the forest edge toward the steppe (not shown). The Ψm was always close to Ψ0 (Table 3). In July, Ψm was even below Ψ0; in August, when water potentials were generally higher, Ψ0 was significantly above Ψm (Table 3).

Table 3.

Points of zero turgor (Ψ0) as well as minimum (Ψm) and maximum (Ψp) shoot water potentials (Ψm) in L. sibirica growing at the forest edge on the steppe or in the interior of light taiga forests at Mt. Bogd Uul, southern Khentey. Asterisks indicate significant differences between Ψ0 and Ψm (Bonferroni-corrected t test, P ≤ 0.05, d.f. = 5–6).

MPa Month Forest edge Forest interior 
Ψp July −0.99 ± 0.03 −0.98 ± 0.03 
August −0.52 ± 0.02 −0.55 ± 0.02 
Ψm July −2.29 ± 0.03 −2.26 ± 0.01 
August −1.53 ± 0.03 −1.52 ± 0.04 
Ψ0 July −1.90 ± 0.13* −2.09 ± 0.04* 
August −1.87 ± 0.06* −1.88 ± 0.07* 
MPa Month Forest edge Forest interior 
Ψp July −0.99 ± 0.03 −0.98 ± 0.03 
August −0.52 ± 0.02 −0.55 ± 0.02 
Ψm July −2.29 ± 0.03 −2.26 ± 0.01 
August −1.53 ± 0.03 −1.52 ± 0.04 
Ψ0 July −1.90 ± 0.13* −2.09 ± 0.04* 
August −1.87 ± 0.06* −1.88 ± 0.07* 

Discussion

Our studies of the water relations in L. sibirica showed that the trees at the forest-steppe borderline in northern Mongolia frequently suffer from drought stress. The water supply of trees growing on mountain ridges at the forest line to the steppe strongly depends on the current precipitation, as the roots are not in contact with the groundwater table (Li et al. 2007a). The results of Li et al. (2007a) match with the correlation of Ψ0 with the monthly precipitation observed in our study (Figure 1D). Velisevich and Kozlov (2006) found a dependence of wood formation in L. sibirica on the precipitation received in the early growing season. Permafrost is a source of water for trees growing in closed forests on north-facing slopes in northern Mongolia and southern Siberia, but not for trees on the south-facing slopes covered by grasslands (Sugimoto et al. 2002, Böhner and Lehmkuhl 2005, Etzelmüller et al. 2006).

Daily values of Ψm in the range of Ψ0 or even < Ψ0 were frequently encountered during dry weather (Figure 1A, B; Table 3). Such low in situ values of Ψ close to Ψ0 show that the water supply of L. sibirica at the forest line to the steppe is often critical. The values of Ψm which are significantly below Ψ0 suggest the occurrence of negative turgor values in L. sibirica, which are regularly found in plants with rigid cell walls (Kreeb 1961, Rhizopoulou 1997). The repeated occurrence of Ψm values around Ψ0 implies that the larch trees growing in the forest-steppe transition are likely to suffer from extended drought periods where needle growth must be inhibited and the risk of xylem cavitation may be high. However, no vulnerability curves of shoot embolism exist for the larch trees studied. The North American Larix occidentalis Nutt. is known to be resistant to drought stress-induced cavitation (Piñol and Sala 2000). Similar results were obtained in Larix laricina (Du Roi) K.Koch, for which Wang (2005) showed only a slight decrease of specific hydraulic conductivity during the growing season in British Columbia.

The tense water relations inferred from the proximity of Ψm and Ψ0 suggest that L. sibirica is currently unable to encroach from the forest onto the steppe, except in a sequence of several particularly moist and cool years. This conclusion is supported by the fact that Ψm equaled Ψ0 even more often in saplings than in mature trees (Figure 1B). While mature larch trees can efficiently regulate the concentration of solutes in the vacuole and, with it, the osmotic potential (Badalotti et al. 2000) along with the availability of water (Figure 1D), young trees on the steppe did not respond to increasing precipitation after drought (Figure 1D). Seedlings planted in two subsequent years on steppe slopes of the western Khentey near Khonin Nuga corroborated that drought exacerbates the establishment of L. sibirica (Dulamsuren et al. 2008, Hauck et al. 2008). In combination with insect and small mammal herbivory, water shortage completely destroyed all planted seedlings within one growing season. Sowing experiments showed that low soil moisture, high soil temperatures and granivory make the emergence of L. sibirica from seeds in the steppes bordering on the north-facing taiga forests an improbable event (Dulamsuren et al. 2008).

The present results combined with the results of Dulamsuren et al. (2008) and Hauck et al. (2008) suggest that a spread of L. sibirica on the sun-exposed steppe slopes is currently unlikely, irrespective of the origin of the present vegetation pattern with forests on north-facing slopes and grasslands on south-facing slopes. Increasing temperatures concomitant to decreasing precipitation in parts of the Mongolian forest-steppe ecotone as predicted in scenarios (Sato and Kimura 2006, Sato et al. 2007) further reduce the probability of a future encroachment of L. sibirica onto the steppe. Rather, the current drought stress indicates that the steppe might expand to the disadvantage of forests with increasing aridity in future. Hereby, our study area differs from southern Siberia, where increased precipitation concomitant to increased temperature leads to improved growth conditions for forest trees (Lapenis et al. 2005, Tchebakova et al. 2005). It also differs from alpine forest lines in Mongolia, where tree growth is promoted by increased temperatures (Jacoby et al. 1996). Shifts of forest-steppe borderlines even on large spatial scales along with the changes in climate happened repeatedly in Mongolia throughout the Holocene (Gunin et al. 1999, Miehe et al. 2007). Spreads of L. sibirica into the steppe during the 20th century were correlated both to favorable moist and cool climate and to low grazing pressure by livestock (Treter 2000, Sankey et al. 2006).

Lower Ψm and Ψp values in larch trees growing as single trees in front of the forest line than in L. sibirica from the forest line itself (Figures 2 and 3A) reflect the warmer and drier microclimate on the open steppe slopes than at the forest edge. The shoot water status of larch trees in the forest strongly depends on the site. The L. sibirica growing in riverine forest exhibited higher Ψm than that at the forest edge (Figure 3B), as the water supply in floodplains is ensured by both groundwater and the current precipitation (Li et al. 2007a). In contrast to our expectations, larch trees growing on north-facing slopes had lower values of Ψm than the trees at the forest edge. The trend for higher δ13C values in larch trees from the forest interior than from the forest line matches with these Ψm values (Peuke et al. 2006), though the difference in the δ13C values was not large enough to be statistically significant. The more critical water relations in trees from the forest interior could theoretically be due to a higher competition for water as a result of the higher stand density in the forest interior than at the forest edge. However, higher stomatal conductance in the forest interior than at the forest line (Figure 4A) suggest that the larch trees on the northern slopes are less adapted to drought than the trees at the forest line. Conifers adapt to microclimate already during seed formation (Greenwood and Hutchinson 1996, Rehfeldt et al. 1999). Further, the strong selection pressure on the germinating seedlings at the forest edge is likely to promote those genotypes that are most efficiently adapted to drought. The higher drought stress observed in larch trees on north-facing slopes than at the forest edge to the steppe in the present study might be the result of a recent increase in aridity due to the global late 20th century warming.

Low δ13C values in L. sibirica can also be caused by herbivores, as Li et al. (2007b) found lower values in newly emerged needles after an herbivore attack than in needles formed during foliation at the beginning of the growing season. The larch forest of the present study was subject to heavy infestation by gypsy moth (Lymantria dispar L.) in 2005 and to a minor invasion in 2006. Moreover, gypsy moth caused significantly more damage at the forest line to the steppe than in the forest interior (Hauck et al. 2008). However, the difference found for Ψm and stomatal conductance between the forest line and the forest interior in our study indicates that water supply, rather than the gypsy moth attack, was the main cause of the trend for more negative δ13C values at the forest edge than in the forest interior.

Our data from the mountain taiga in the western Khentey and from the forest-steppe belt in the southern Khentey show that L. sibirica often suffers from drought stress both at its southernmost distribution limit (Table 3) and in the less arid mountain sites of northernmost Mongolia (Figure 1A, B). Whether trees in larch forests occurring as outposts within grasslands at the southern distribution limit of L. sibirica (Hilbig et al. 2004) are subject to more severe drought stress than trees growing in the mountain taiga, where steppe only occurs as islands inside the forest (Dulamsuren et al. 2005a, 2005b), cannot be reliably inferred from our data. Though trees growing both in the forest and on the steppe reached Ψm values significantly < Ψ0 in the southern, but not in the western Khentey, any comparison between different regions would require simultaneous long-term measurements of Ψ because of the considerable interannual variation of precipitation and temperature in the highly continental Mongolia (Nandintsetseg et al. 2007).

The studied trees of L. sibirica were well able to deal with the high-light conditions in the forest-steppe ecotone. The capacity to dissipate surplus light energy as heat in the PSII antennae (NPQ) together with the energy quenching through alternative electron sinks was sufficient even in the larch trees growing under the high-light conditions of the forest edge to prevent photoinhibition or photodestructive damage (Figure 5E). This was obvious from the rapid and complete recovery of the PSII efficiency (Φ2) to values > 0.8 in all trees after sunset (Figure 5C). However, the strong mismatch between stable ETR (Figure 5D) and steadily declining CO2 uptake (Figure 5C) suggests intense dissipation of excess electrons through either photorespiration or the water–water cycle or both in addition to NPQ (Niyogi 2000, Ort and Baker 2002). The photoprotective role of the photorespiratory pathway was extensively demonstrated on transgenic tobacco plants with increased and reduced capacity for photorespiration as well as on wildtype soybean plants (Kozaki and Takeba 1996, Jiang et al. 2006), whereas Asada (1999, 2000) clearly showed the close association of the water–water cycle with light protection.

Though our data on the present water relations in L. sibirica suggest that the species does not have the potential to colonize new sites in Mongolia’s forest-steppe ecotone, which are presently covered with grasslands, our results do not allow us to conclude as to whether the present vegetation pattern with forests on north-facing slopes and steppes on south-facing slopes is natural or was originally created by anthropogenic activities. Studies on the vegetation history of Mongolia suggest that the significance of anthropo-zoogenic factors for the formation of the present landscape differs between regions. Pollen analyses of our particular study area at Khonin Nuga in the western Khentey showed that the present vegetation pattern existed throughout the late Holocene, and this pattern probably reflects the spatial variation of microclimate (Schlütz et al. 2008). In Khonin Nuga, this conclusion is supported by convincing evidence that the area is avoided by pastoral nomads (Schlütz et al. 2008). In other more densely populated areas of Mongolia, which are more attractive for livestock breeding than Khonin Nuga, human impact is thought to have modified the natural, climate-dependent pattern of forests and grasslands to the disadvantage of the former (Hilbig 1995, Rösch et al. 2005, Miehe et al. 2007).

Conclusions

The results of the present study show that the trees of L. sibirica growing at the forest line to the steppe at the southern distribution limit of the northern Eurasian coniferous forest belt regularly suffer from drought stress. Comparison of Ψm and Ψ0 values revealed that the water relations in L. sibirica were often critical, though the study years were no particular drought years. This suggests that L. sibirica is currently unable to encroach on the steppe at the studied forest-steppe ecotones, but that the drought years could readily cause a retreat of the forest line. The low potential of L. sibirica to colonize areas currently covered with grasslands is supported by the observation that young individuals were particularly susceptible to drought stress (Figure 1B). Strongly increasing temperatures in the forested areas of Mongolia in recent times and regionally decreasing precipitation (Batima et al. 2005, Sato and Kimura 2006, Sato et al. 2007) suggest that the potential of L. sibirica to encroach onto the steppe in the forest-steppe ecotone of Mongolia will deteriorate in future. Lower shoot water potentials and higher stomatal conductance in trees from the interior of larch forests on north-facing slopes than in trees growing at the forest line to the steppe could be due to a recent increase in aridity during the lifetime of the trees and thus forewarn of future declines of L. sibirica at these sites. A retreat of L. sibirica from the forest-steppe ecotones would equal a decline of forests per se, as L. sibirica covers around 80% of Mongolia’s forested area (Savin et al. 1978, Gunin et al. 1999).

This study was supported by grants of the German Science Foundation (Deutsche Forschungsgemeinschaft) to Ch. Dulamsuren (Du 1145/1–1, 1–2). The authors are thankful to Prof. Dr. Michael Mühlenberg (Center of Nature Conservation, University of Göttingen) for making the facilities of Khonin Nuga Research Station (a biological field station jointly run by the University of Göttingen and the National University of Mongolia, Ulan Bator) available to the authors. Dr. Doris Möller, Denise Schwerdtner (University of Göttingen), Dr. S. Byambasuren (Ulan Bator, †) and T. Sain-Erdene (National University of Mongolia, Ulan Bator) helped with the measurements of the shoot water potentials. Prof. Dr. Milan Chytrý (Masaryk University, Brno, Czech Republic) and an unknown reviewer are thanked for their helpful comments.

References

Aizen
E.M.
Aizen
V.B.
Melack
J.M.
Nakamura
T.
Otha
T.
Precipitation and atmospheric circulation patterns at mid-latitudes of Asia
Int. J. Climatol.
 , 
2001
, vol. 
21
 (pg. 
535
-
556
)
Asada
K.
The water–water cycle in chloroplasts: scavenging of active oxygens and dissipation of excess photons
Annu. Rev. Plant Physiol. Plant Mol. Biol.
 , 
1999
, vol. 
50
 (pg. 
601
-
639
)
Asada
K.
The water–water cycle as alternative photon and electron sink
Philos. Trans. R. Soc. Lon. B Biol. Sci.
 , 
2000
, vol. 
355
 (pg. 
1419
-
1430
)
Badalotti
A.
Anfodillo
T.
Grace
J.
Evidence of osmoregulation in Larix decidua at alpine forest line and comparative responses to water availability of two co-occurring evergreen species
Ann. For. Sci.
 , 
2000
, vol. 
57
 (pg. 
632
-
633
)
Batima
P.
Natsagdorj
L.
Gombluudev
P.
Erdenetsetseg
B.
Observed climate change in Mongolia
Assess. Impacts Adapt. Climate Change Work. Pap.
 , 
2005
, vol. 
12
 (pg. 
1
-
26
)
Böhner
J.
Lehmkuhl
F.
Environmental change modelling for Central and High Asia: Pleistocene, present and future scenarios
Boreas
 , 
2005
, vol. 
34
 (pg. 
220
-
231
)
Chytrý
M.
Danihelka
J.
Kubešová
S.
, et al.  . 
Diversity of forest vegetation across a strong gradient of climatic continentality: western Sayan Mountains, southern Siberia
Plant Ecol.
 , 
2008
, vol. 
196
 (pg. 
61
-
83
)
Dinesman
L.G.
Knyazev
A.V.
Kiseleva
N.K.
History of the steppe ecosystems of the Mongolian People’s Republic
 , 
1989
Moscow
Nauka
215
 
(in Russian)
Dulamsuren
Ch.
Hauck
M.
Spatial and seasonal variation of climate on steppe slopes of the northern Mongolian mountain taiga
Grassl. Sci.
 , 
2009
 
(in press)
Dulamsuren
Ch.
Hauck
M.
Mühlenberg
M.
Vegetation at the taiga forest-steppe borderline in the western Khentej Mountains, northern Mongolia
Ann. Bot. Fenn.
 , 
2005
, vol. 
42
 (pg. 
411
-
426
)
Dulamsuren
Ch.
Hauck
M.
Mühlenberg
M.
Ground vegetation in the Mongolian taiga forest-steppe ecotone does not offer evidence for the human origin of grasslands
Appl. Veg. Sci.
 , 
2005
, vol. 
8
 (pg. 
149
-
154
)
Dulamsuren
Ch.
Hauck
M.
Mühlenberg
M.
Insect and small mammal herbivores limit tree establishment in northern Mongolian steppe
Plant Ecol.
 , 
2008
, vol. 
195
 (pg. 
143
-
156
)
Etzelmüller
B.
Heggem
E.S.F.
Sharkhuu
N.
Frauenfelder
R.
Kääb
A.
Goulden
C.
Mountain permafrost distribution modelling using a multi-criteria approach in the Hövsgöl area, northern Mongolia
Permafrost Periglac. Process
 , 
2006
, vol. 
17
 (pg. 
91
-
104
)
Giese
E.
Moßig
I.
Klimawandel in Zentralasien
Schriftenr. Zentr Int. Entw. Umweltforsch.
 , 
2004
, vol. 
17
 (pg. 
1
-
70
)
Greenwood
M.S.
Hutchinson
K.W.
Hom
J.
Birdsey
R.
O’Brian
K.
Genetic aftereffects of increased temperature in Larix
Proceedings 1995 Meeting of the Northern Global Change Program, 14–16 March 1995
 , 
1996
Pittsburgh, PA
USDA Forest Service
(pg. 
56
-
62
)
Gunin
P.D.
Vostokova
E.A.
Dorofeyuk
N.I.
Tarasov
P.E.
Black
C.C.
Vegetation dynamics of Mongolia
 , 
1999
Dordrecht
Kluwer
233
Hauck
M.
Dulamsuren
Ch.
Heimes
C.
Effects of insect herbivory on the performance of Larix sibirica in a forest-steppe ecotone
Environ. Exp. Bot.
 , 
2008
, vol. 
62
 (pg. 
351
-
356
)
Henderson
K.
Laube
A.
Gäggeler
H.W.
Olivier
S.
Papina
T.
Schwikowski
M.
Temporal variations of accumulation and temperature during the past two centuries from Belukha ice core, Siberian Altai
J. Geophys. Res.
 , 
2006
, vol. 
111
 pg. 
D03104
  
doi:10.1029/2005JD005819
Hilbig
W.
The vegetation of Mongolia
 , 
1995
Amsterdam
SPB
258
Hilbig
W.
Jäger
E.J.
Knapp
H.D.
Die Vegetation des Bogd-uul bei Ulaanbaatar (Mongolei) – Standortsbindung und pflanzengeographische Stellung
Feddes Repert.
 , 
2004
, vol. 
115
 (pg. 
265
-
342
)
Jacoby
G.C.
D’Arrigo
R.D.
Davaajamts
T.
Mongolian tree-rings and 20th-century warming
Science
 , 
1996
, vol. 
273
 (pg. 
771
-
773
)
Jiang
C.-D.
Gao
H.-Y.
Zou
Q.
Jiang
G.-M.
Li
L.-H.
Leaf orientation, photorespiration and xanthophyll cycle protect young soybean leaves against high irradiance in the field
Environ. Exp. Bot.
 , 
2006
, vol. 
55
 (pg. 
87
-
96
)
Kozaki
A.
Takeba
G.
Photorespiration protects C3 plants from photooxidation
Nature
 , 
1996
, vol. 
384
 (pg. 
57
-
560
)
Kreeb
K.
Zur Frage des negative Turgors bei mediterranen Hartlaubpflanzen unter natürlichen Bedingungen
Planta
 , 
1961
, vol. 
56
 (pg. 
479
-
489
)
Lapenis
A.
Shvidenkow
A.
Shepaschenkoz
D.
Nilsson
S.
Aiyyer
A.
Acclimation of Russian forests to recent changes in climate
Glob. Change Biol.
 , 
2005
, vol. 
11
 (pg. 
2090
-
2102
)
Lavrenko
E.M.
Karamysheva
Z.V.
Coupl
R.T.
Steppes of the former Soviet Union and Mongolia
Natural Grasslands. Eastern Hemisphere and Résumé
 , 
1993
Amsterdam
Elsevier
(pg. 
3
-
59
)
Li
S.-G.
Asanuma
J.
Kotani
A.
Eugster
W.
Davaa
G.
Oyunbaatar
D.
Sugita
M.
Year-round measurements of net ecosystem CO2 flux over a montane larch forest in Mongolia
J. Geophys. Res.
 , 
2005
, vol. 
110
 pg. 
D09303
  
doi:10.1029/2004JD005453
Li
S.-G.
Asanuma
J.
Kotani
A.
Eugster
W.
Davaa
G.
Oyunbaatar
D.
Sugita
M.
Net ecosystem carbon dioxide exchange over grazed steppe in central Mongolia
Glob. Change Biol.
 , 
2005
, vol. 
11
 (pg. 
1941
-
1955
)
Li
S.-G.
Tsujimura
M.
Sugimoto
A.
Sasaki
L.
Yamanaka
T.
Davaa
G.
Oyunbaatar
D.
Sugita
M.
Seasonal variation in oxygen isotope composition of waters for a montane larch forest in Mongolia
Trees
 , 
2006
, vol. 
20
 (pg. 
122
-
130
)
Li
S.-G.
Romero-Saltos
H.
Tsujimura
M.
Sugimoto
A.
Sasaki
L.
Davaa
G.
Oyunbaatar
D.
Plant water sources in the cold semiarid ecosystem of the upper Kherlen river catchment in Mongolia: a stable isotope approach
J. Hydrol.
 , 
2007
, vol. 
333
 (pg. 
109
-
117
)
Li
S.-G.
Tsujimura
M.
Sugimoto
A.
Davaa
G.
Oyunbaatar
D.
Sugita
M.
Temporal variation of δ13C of larch leaves from a montane boreal forest in Mongolia
Trees
 , 
2007
, vol. 
21
 (pg. 
479
-
490
)
Miehe
G.
Schlütz
F.
Miehe
S.
Opgenoorth
L.
Cermak
J.
Samiya
R.
Jäger
E.J.
Wesche
K.
Mountain forest islands and Holocene environmental changes in Central Asia: a case study from the southern Gobi Altay, Mongolia
Palaeogeogr. Palaeoclimatol. Palaeoecol.
 , 
2007
, vol. 
250
 (pg. 
150
-
166
)
Nandintsetseg
B.
Greene
J.S.
Goulden
C.E.
Trends in extreme daily precipitation and temperature near Lake Hövsgöl, Mongolia
Int. J. Climatol.
 , 
2007
, vol. 
27
 (pg. 
341
-
347
)
Niyogi
K.K.
Safety valves for photosynthesis
Curr. Opin. Plant Biol.
 , 
2000
, vol. 
3
 (pg. 
455
-
460
)
Ort
D.R.
Baker
N.R.
A photoprotective role for O2 as an alternative electron sink in photosynthesis?
Curr. Opin. Plant Biol.
 , 
2002
, vol. 
5
 (pg. 
193
-
198
)
Parker
W.C.
Pallardy
S.G.
The influence of resaturation method and tissue type on pressure–volume analysis of Quercus alba L. seedlings
J. Exp. Bot.
 , 
1987
, vol. 
28
 (pg. 
535
-
549
)
Peuke
A.D.
Gessler
A.
Rennenberg
H.
The effect of drought on C and N stable isotopes in different fractions of leaves, stems and roots of sensitive and tolerant beech ecotypes
Plant Cell Environ.
 , 
2006
, vol. 
29
 (pg. 
823
-
835
)
Piñol
J.
Sala
A.
Ecological implications of xylem cavitation for several Pinaceae in the Pacific Northern USA
Funct. Ecol.
 , 
2000
, vol. 
14
 (pg. 
538
-
545
)
Rehfeldt
G.E.
Ying
C.C.
Spittlehouse
D.L.
Hamilton
D.A.
Genetic responses to climate in Pinus contorta: niche breadth, climate change, and reforestation
Ecol. Monogr.
 , 
1999
, vol. 
69
 (pg. 
375
-
407
)
Rhizopoulou
S.
Is negative turgor fallacious?
Physiol. Plant.
 , 
1997
, vol. 
99
 (pg. 
505
-
510
)
Roberts
S.W.
Strain
B.R.
Knoerr
K.R.
Seasonal patterns of leaf water relations in four co-occurring forest tree species: parameters from pressure–volume curves
Oecologia
 , 
1980
, vol. 
46
 (pg. 
330
-
337
)
Roháček
K.
Chlorophyll fluorescence parameters: the definitions, photosynthetic meaning, and mutual relationships
Photosynthetica
 , 
2002
, vol. 
40
 (pg. 
13
-
29
)
Rösch
M.
Fischer
E.
Märkle
T.
Human diet and land use in the time of the Khans – archaeobotanical research in the capital of the Mongolian Empire, Qara Qorum, Mongolia
Veg. Hist. Archaeobot.
 , 
2005
, vol. 
14
 (pg. 
485
-
492
)
Sankey
T.T.
Montagne
C.
Graumlich
L.
Lawrence
R.
Nielsen
J.
Lower forest-grassland ecotones and 20th century livestock herbivory effects in northern Mongolia
For. Ecol. Manag.
 , 
2006
, vol. 
233
 (pg. 
36
-
44
)
Sato
T.
Kimura
F.
Regional climate simulations to diagnose environmental changes in Mongolia
Bull. Terr. Environ. Res. Cent. Univ. Tsukuba
 , 
2006
, vol. 
7
 (pg. 
59
-
69
)
Sato
T.
Kimura
F.
Kitoh
A.
Projection of global warming onto regional precipitation over Mongolia using a regional climate model
J. Hydrol.
 , 
2007
, vol. 
333
 (pg. 
144
-
154
)
Savin
E.N.
Ogorodnikov
A.V.
Semechkin
I.V.
Dugarzhov
Ch.
Korotkov
I.A.
Forests of the Mongolian People’s Republic
 , 
1978
Moscow
Nauka
127
 
(in Russian)
Schlütz
F.
Dulamsuren
Ch.
Wieckowska
M.
Mühlenberg
M.
Hauck
M.
Late Holocene vegetation history suggests natural origin of steppes in the northern Mongolian mountain taiga
Palaeogeogr. Palaeoclimatol. Palaeoecol.
 , 
2008
, vol. 
261
 (pg. 
203
-
217
)
Scholander
P.F.
Hammel
H.T.
Hemmingsen
E.A.
Bradstreet
E.D.
Hydrostatic pressure and osmotic potential in leaves of mangroves and some other plants
Proc. Natl. Acad. Sci. USA
 , 
1964
, vol. 
52
 (pg. 
119
-
125
)
Sugimoto
A.
Yanagisawa
N.
Naito
D.
Fujita
N.
Maximov
T.C.
Importance of permafrost as a source of water for plants in east Siberian taiga
Ecol. Res.
 , 
2002
, vol. 
17
 (pg. 
493
-
503
)
Tchebakova
N.M.
Rehfeldt
G.E.
Parfenova
E.I.
Impacts of climate change on the distribution of Larix spp. and Pinus sylvestris and their climatypes in Siberia
Mitigat. Adapt. Strat. Glob. Change
 , 
2005
, vol. 
11
 (pg. 
861
-
882
)
Treter
U.
Recent extension and regeneration of the larch forest in the mountain forest steppe of north-west Mongolia
Marburger Geogr. Schr.
 , 
2000
, vol. 
135
 (pg. 
156
-
170
)
Velisevich
S.N.
Kozlov
D.S.
Effects of temperature and precipitation on radial growth of Siberian larch in ecotopes with optimal, insufficient, and excessive soil moistening
Russ. J. Ecol.
 , 
2006
, vol. 
37
 (pg. 
241
-
246
)
Vostokova
E.A.
Gunin
P.D.
Ecosystems of Mongolia, Atlas
 , 
2005
Moscow
Russian Academy of Sciences
48
Wallis de Vries
M.F.
Manizabar
N.
Dügerlham
S.
The vegetation of the forest-steppe region of Hustain Nuruu, Mongolia
Vegetatio
 , 
1996
, vol. 
122
 (pg. 
111
-
127
)
Walter
H.
Breckle
S.-W.
Ökologie der Erde. Band 3. Spezielle Ökologie der gemäßigten und arktischen Zonen Euro-Nordasiens: Zonobiom VI–IX
 , 
1994
2nd Edn.
Stuttgart
Fischer
726
Wang
J.R.
Spring and summer hydraulic conductivity of 14 woody species of the sub-boreal forest in British Columbia
Can. J. For. Res.
 , 
2005
, vol. 
35
 (pg. 
2727
-
2733
)