Effects of herbivores and litter on Lithocarpus hancei seed germination ...

7 downloads 257006 Views 1015KB Size Report
May 17, 2016 - Effects of herbivores and litter on Lithocarpus hancei seed germination and seedling survival in the understorey of a high diversity forest in SW ...
Plant Ecol DOI 10.1007/s11258-016-0610-0

Effects of herbivores and litter on Lithocarpus hancei seed germination and seedling survival in the understorey of a high diversity forest in SW China Jin-Jin Hu . Cheng-Chang Luo . Roy Turkington . Zhe-Kun Zhou

Received: 31 December 2015 / Accepted: 29 April 2016 Ó Springer Science+Business Media Dordrecht 2016

Abstract Tanoak Lithocarpus hancei (Fagaceae) is one of the dominant species in the high diversity subtropical evergreen broad-leaved forests in SW China. However, seedlings of L. hancei and other oaks are quite rare in the understorey. To investigate the effects of seed (acorn) predation and seedling herbivory by mammals, and litter, on acorn germination and seedling survival of L. hancei in these forests, we set up a 2 9 2 factorial experiment (litter present or removed; ±herbivore exclosures (fences); plus natural

Electronic supplementary material The online version of this article (doi:10.1007/s11258-016-0610-0) contains supplementary material, which is available to authorized users.

control; 5 replications) in the Ailaoshan National Nature Reserve, central Yunnan from 2010 to 2015. Acorns and transplanted seedlings of L. hancei were placed in the four treatments plots and the influence of these treatments on acorn germination and seedling survival was monitored. Fences protected L. hancei acorns and seedlings against herbivory by rodents and other mammals; litter had a positive effect on acorn survival but no effect on seedling establishment. Moreover, those seedlings that escaped herbivory were mostly killed by fungal attack. Our results indicate that while litter and pathogens have some influence, herbivores are probably the major cause of the low frequency of L. hancei seedlings in the understorey.

J.-J. Hu  Z.-K. Zhou (&) Key Laboratory of Tropical Forest Ecology, Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences, Mengla, Yunnan, China e-mail: [email protected]

Keywords Herbivory  Litter  Seed germination  Seedling establishment  Lithocarpus hancei  Tanoak

C.-C. Luo Ailaoshan Station for Subtropical Forest Ecosystem Research, Chinese Ecosystem Research Networks, Jingdong, Yunnan, China

Introduction

R. Turkington (&) Department of Botany, and Biodiversity Research Centre, University of British Columbia, Vancouver, BC, Canada e-mail: [email protected] Z.-K. Zhou Key Laboratory for Plant Diversity and Biogeography of East Asia, Kunming Institute of Botany, Chinese Academy of Sciences, Kunming, Yunnan, China

Lithocarpus hancei (Benth.) Rehder (Fagaceae), a kind of tanoak, is one of the dominant species in the subtropical evergreen broad-leaved forests in South Central Yunnan Province, China. These forests contain *6 % of the world’s total higher plants species ([20,000 species). As one of the dominant species in this ecosystem, L. hancei provides the major structural component for the forests and provides habitat for

123

Plant Ecol

many other plants and animals (Qiu et al. 1998) including epiphytic lichens (Li et al. 2015). Seedlings of L. hancei and other oaks, however, are not very abundant. For the maintenance of these diverse forests and the biodiversity they support, it is important to understand some of the factors that influence the survival of L. hancei, particularly the establishment and survival of seedlings. Similar to L. hancei, most of the Fagaceae species, which are one of the dominant taxa in the broad-leaved evergreen mixed forests in China, have poor natural regeneration in these forests (Crow 1988; Thadani and Ashton 1995; Gardiner and Hodges 1998). The main factors inhibiting seedling establishment are seed (acorn) predation and seedling herbivory, drought, frost damage to seedlings, destruction by pathogens, competition from understorey vegetation, low understorey light levels, allelopathic interaction or a physical barrier caused by litter, etc. (McGee 1975; Lorimer et al. 1994; Baskin and Baskin 2001; Moles and Westoby 2004; Qi et al. 2014). Among these potentially limiting factors, herbivores, mainly rodents and insects, and litter play an important role in the survival of acorns and seedlings of Fagaceae species. Herbivory by mammals and insects is one of the most common causes of seed and seedling mortality (Kitajima and Augspurger 1989; Augspurger and Kitajima 1992; Moles and Westoby 2004). A large proportion of tree seeds are consumed by mammals (Abbott 1961; Klinger and Rejma´nek 2013). The influence of mammals on slowing natural reseeding is profound (Gashwiler 1967; Kitajima and Augspurger 1989; Klinger and Rejma´nek 2013). Many studies have shown that rates of seed predation and dispersal can be strongly affected by individual mammal species. Willis (1914) and Shaw (1954) reported that a mouse (Peromyscus spp.) would consume (eat and store) 200–300 Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) seeds daily. Mice and voles eat and store naturally disseminated white pine seed; about half of the seeds are eaten in situ, the other half are stored as a food resource in winter (Abbott 1961). Acorns of Fagaceae species are favoured by mammals, such as rodents and wild boars (Xiao and Zhang 2012; Sunyer et al. 2015). Rodents will seek very young seedlings and eat cotyledons in the seed coat (Gashwiler 1967). Insects, especially weevils, also cause severe damage to acorns and seedlings of members of

123

the Fagaceae (Dalgleish et al. 2012; Mun˜oz et al. 2014). Litter may have a positive or negative influence on seed germination and seedling establishment (Molofsky and Augspurger 1992; Baskin and Baskin 2001; Rotundo and Aguiar 2005; Loydi et al. 2013). Litter usually affects seeds and seedlings by creating different microsites and having complex interactions with other biotic and abiotic factors (Molofsky and Augspurger 1992). Litter can reduce soil temperature fluctuations and increase soil moisture, thereby providing better conditions for seed germination (Sork 1983; Li and Ma 2003; Eckstein and Donath 2005; Dechoum et al. 2015). Litter may also protect seeds from being found and eaten by mammals and promote seed survival (Cintra 1997; Li et al. 2006). In contrast, litter can reduce seed germination and seedling recruitment by allelopathic interaction (Datta and Chatterjee 1980; Rai and Tripathi 1984; Das et al. 2012; Qi et al. 2014), by reducing light levels for seeds (Eckstein and Donath 2005), and by acting as a physical barrier that prevents seedling rooting and shoot emergence (Sydes and Grime 1981; Hamrick and Lee 1987; Eriksson 1995). Moreover, litter may provide a good substrate for fungal damage to seedlings by creating a shady and wet microenvironment favoured by fungal pathogens (Garcı´a-Guzma´n and Benı´tez-Malvido 2003). The primary objective of this study was to test if litter and herbivory by small mammals could account for the low frequency of seedlings of L. hancei in these diverse forests. We did this by manipulating litter and by reducing herbivory and monitoring the influence of these treatments on acorn germination and seedling establishment and survival.

Methods The study area The study was conducted near the Forest Ecosystem Research Station, Chinese Academy of Sciences at Xujiaba within the Ailaoshan National Nature Reserve (23°350 –24°440 N, 100°540 –101°30E; 2200 m asl). Ailaoshan has a monsoon climate with distinct wet seasons (from May to October) and dry seasons (from November to April the next year); mean annual rainfall is 1931 mm; mean annual temperature is

Plant Ecol

11.3 °C ranging from 5.4 °C in January to 16.4 °C in July (Qiu et al. 1998). The soils are acidic, pH 4.4–4.9, with a high organic matter and nitrogen content, and litter is typically 3–7 cm deep (Qiu et al. 1998; Liu et al. 2002). The Ailaoshan forests average about 68 woody plant species (with diameter at breast height C1 cm) per 6 ha (Gong et al. 2011a) with occasional hotspots up to 94 species per 0.4 ha (Young and Herwitz 1995). The forest is classified as subtropical montane evergreen broad-leaved forest (Editorial Committee for Vegetation of China 1980) or warm-temperate broadleaved evergreen/mixed forest (Ni 2001). These forests contain *6 % of the world’s total higher plants species ([20,000 species). The dominant species are Lithocarpus xylocarpus (Kurz) Markgr., L. hancei, Castanopsis wattii (King ex Hook.f.) A. Camus and Schima noronhae Reinw. ex Blume but other common tree species include Manglietia insignis (Wall.) Blume, Vaccinium duclouxii (H. Le´v.) Hand.-Mazz., Machilus viridis Hand.-Mazz., etc. (Qiu et al. 1998). The understorey is sparse with common tree species including Fargesia wuliangshanensis T. P. Yi, Machilus gamblei King ex Hook.f. and Symplocos ramosissima Wallich ex G. Don (Zhu and Yan 2009; Gong et al. 2011b; Qi et al. 2015). There are many animals in the area. The most common large mammals are the tufted deer (Elaphodus cephalophus Milne-Edwards), the black crested gibbon (Nomascus concolor Harlan) and stump-tailed macaque (Macaca arctoides Geoffroy) (Zhao et al. 1988). There are also many small mammals such as the South China field mouse (Apodemus draco Barrett-Hamilton), the Asian red-cheeked squirrel (Dremomys rufigenis Blanford), shrew gymnure (Neotetracus sinensis Trouessart) (Zhao et al. 1988), and birds such as the white-tailed leaf warbler (Phylloscopus davisoni Oates) and crested finchbill (Spizixos canifrons Blyth) (Wei et al. 1988; Wang et al. 2000). Experimental design The experiment was a fully factorial design with two levels of litter (present and removed), two levels of herbivory (present and reduced/absent), plus a natural control, all replicated 5 times for a total of 25 plots. Twenty-five 1 m 9 1 m plots were selected in parts of the forest in the understorey of L. hancei with a

copious litter layer at least 4 cm deep. None of the plots had any seedlings of any species. The plots were randomly divided among five treatments: (1) litter present along with a fence; (2) litter removed, and fenced; (3) litter present, and no fence; (4) litter removed, and no fence; (5) natural control (no treatment; natural litter and having no fence). Fences were 60 cm high and made of galvanized chicken wire with 0.5 cm square mesh, supported by galvanized pipes at each corner, and firmly stapled to the ground to prevent animals intruding under the fence. In addition, the fenced plots were covered with 1 cm mesh fishing net to reduce intrusion by birds; there was some obvious evidence of herbivores entering the fenced plots underground. In plots with a no-litter treatment, the litter (including acorns) was lightly brushed off the plots and all subsequent litter and acorns that fell to the ground or on top of nets were removed on a weekly basis. In plots that had litter, fallen acorns were removed and the litter spread evenly across the plots. The ‘natural control’ plots were marked at their corners by PVC pipes but otherwise were left untouched. In all 25 plots, a central area of 0.5 m 9 0.5 m was marked with small PVC pipes to demarcate the experimental area and was therefore surrounded by a 25 cm wide buffer zone to avoid edge-effects, i.e. an area surrounding the experimental area that was treated in exactly the same way as the experimental area but not used for any measurements. Lithocarpus hancei acorns were collected in this area under the canopy trees and these were used directly for germination tests and for the production of seedlings for transplanting. All acorns were tested by water floatation for soundness and those that were defective or infested by weevil (Curculio spp.) larvae were removed. In the central area of each of the 20 treatment plots, 10 acorns of L. hancei were planted in 2 rows of 5 and buried to a depth of 2 mm. Acorns were first placed in the plots on 20 Oct 2010 immediately after collection and repeated on four more occasions: 1 Jun 2011, 15 Nov 2011, 20 Nov 2012 and 11 Nov 2013. Because of mast seeding in the Fagaceae, there were almost no L. hancei acorns in 2012 and 2014, but there were abundant acorns in 2010, 2011 and especially in 2013. Therefore, no acorns were added to the plots in 2014 and the acorns sown in 2012 were collected in Nov 2011 and stored. In addition, 400 acorns were sown into pots in an

123

Plant Ecol

unheated greenhouse for over 6 months to produce seedlings for transplanting into the field plots the following year. This was done on four occasions: 20 Oct 2010, 15 Nov 2011, 20 Nov 2012 and 11 Nov 2013. The acorns used in 2012 were collected in Nov 2011. Between 7 and 10 seedlings of L. hancei were transplanted in 2 rows in each of the 20 treatment plots. The heights of all seedlings were recorded on the day of planting. Acorns rotten and absent were replaced by new acorns in the following October or November and seedling transplants dead or absent were also replaced by new seedlings. Monitoring began 20 Oct 2010, and continued weekly until 24 Jun 2015. At each of the weekly surveys we noted: Germination of acorns acorns were considered germinated when a radicle became visible; Survival of seedlings from germinated acorns recorded as present until death (the height of seedlings was recorded every 6 months); Cause of death of acorns mostly eaten or removed by rodents, identified by teeth marks on the acorn, rotten or infested by weevil larvae in natural control, or withered as seedlings; Survival of seedling transplants measured height every 6 months until death; Cause of death of transplants mostly killed by small mammals that cut the seedlings at ground level, or by weevils that made obvious holes and cuts on leaves. Some seedlings turned brown and died probably due to fungal pathogens which we have not yet been able to identify. In the natural controls, every acorn that naturally fell into the plots was marked and all of the same measurements were made as described above. Data analysis The data collected from each treatment over the 5 years were combined for statistical analyses. Missing data for survival time of acorns or transplants in the four treatments was \5 % and were replaced by the median value of the plots. In the natural controls, all acorns falling into the plots were counted and those not present in a subsequent survey were considered as ‘fate unknown’. A survival function (a discrete stepped survivorship curve) was estimated for L.

123

hancei acorns and transplants in each treatment using the Kaplan–Meyer method. Survival differences between treatments were tested using the Mantel– Cox test. All of the acorns and transplants in each treatment were present in the plots for about 2 years or longer during the experiment, so survival curves were generated for a 2-year period after placement in the plots. Because the error variance of survival time was not equal across treatments (Levene’s Test of equality of error variance) we compared differences in survival time using a Nonparametric Kruskal–Wallis one-way ANOVA (k samples). The growth of the seedling transplants was measured by Dheight (Dheight = the tallest height during the experiment time - the initial height of the seedling transplant). Differences of the Dheight of the seedlings transplanted into the four treatments were also tested by a Nonparametric Kruskal–Wallis one-way ANOVA (k samples) because of the unequal error variance of the Dheight across treatments. Survival functions and statistical analyses were done using SPSS Statistics version 19.0 (http://www. spss.com.cn). Figures were plotted using R version 3.0.2 (http://www.R-project.org), Adobe Illustrator CS5 and Adobe Photoshop CS5 (http://www.adobe. com).

Results The number of acorns and seedling transplants placed in the plots each year is shown in Table 1. Acorn survival was significantly different in the four treatments and the natural control (v2 = 499.6, df = 4, p \ 0.001; Fig. 1a). Acorns had higher survival rates and survived longer (p \ 0.001; Table 1) in fenced plots than in unfenced plots and the natural control. Acorns in plots with litter also had a higher survival rate than those in plots without litter. However, the survival rate of seedlings developed from acorns had no difference between plots with or without litter (v2 = 0.654, df = 1, p = 0.419). There were significant (p \ 0.001; Table 1) differences in survival time for pairwise comparisons of the four treatments and the natural control, except for the two treatments in fenced plots (p = 0.126); there was no significant difference in survival time between the two fenced treatments and both had longer survival time than the unfenced plots.

Plant Ecol Table 1 Number of Lithocarpus hancei acorns and transplanted seedlings added to the plots each year from 2010 to 2015 and survival time (months) under four experimental treatments and a ‘‘natural control’’ Treatment

Acorns

Transplanted seedlings

Number of acorns or transplanted seedlings

Survival time (months)

Oct 2010

Jun 2011

Nov 2011

Nov 2012

Nov 2013

Nov 2014

Total

Mean

Std. error

1 (fence?, litter?)

50

50

33

39

22

0

194

13.17a

0.66

2 (fence?, litter-) 3 (fence-, litter?)

50 50

50 50

25 50

38 50

35 50

0 0

198 250

11.77a 3.02b

0.71 0.29

4 (fence-, litter-)

50

50

47

50

50

0

247

1.87c

0.18

5 (control, fence-, litter?)

45

0

15

1

85

0

146

5.31d

0.30

1 (fence?, litter?)

36



37

5

19

0

97

9.34a

0.90

2 (fence?, litter-)

33



31

4

21

0

89

12.43a

1.36

3 (fence-, litter?)

38



33

10

20

0

101

2.74b

0.37

4 (fence-, litter-)

37



43

9

22

0

111

2.34b

0.27

‘‘?’’ means present; ‘‘–’’ means absent. Acorns sown in Jun 2011 were collected in Oct 2010 and buried in soil; those sown in 2012 were collected in Nov 2011. Survival time values with different letters are significantly different (p \ 0.05)

1.0

Treatment 1. Fence+, Litter+ 2. Fence+, Litter– 3. Fence–, Litter+ 4. Fence–, Litter– 5. Natural Control censored

Seed Survival

0.8 0.6

0.4 5

0.2

1 2

0.0

4

0

5

10

15

3

20

(b)

1.0

Treatment 1. Fence+, Litter+ 2. Fence+, Litter– 3. Fence–, Litter+ 4. Fence–, Litter– censored

0.8

Seedling Survival

(a)

0.6

0.4

0.2 2 1

4

3

0.0 25

Months after sowing

0

5

10

15

20

25

Months after transplanting

Fig. 1 Survival rates of a Lithocarpus hancei acorns and b L. hancei seedling transplants in experimental plots over a 2-year period after being sown or transplanted. Numbers beside the lines indicate the treatment

Transplanted seedlings also had a higher probability of survival (v2 = 141.0, df = 3, p \ 0.001; Fig. 1b) and lived longer (p \ 0.001; Table 1) in plots that were fenced. However, there was no difference in probability of seedling survival (v2 = 2.8, df = 1, p = 0.09) or survival time (p = 0.48; Table 1) with or without litter in the fenced plots, or the unfenced plots (probability of survival rate, v2 = 0.47, df = 1, p = 0.49; survival time, p = 0.54; Table 1).

Most of acorns survived for 2 years in fenced plots (Fig. 2a). Over 70 % of acorns survived from 2 weeks to 1.5 years in fenced plots and [10 acorns survived for more than 3 years (Table S1). About 10 % of the acorns germinated and grew to seedlings and about half of these seedlings survived for 2–4.5 years (Table S1). However, almost all of the acorns died after 1 year in unfenced plots (Fig. 2c) and most acorns survived for only 6 months (Table S1). For transplanted seedlings, about 90 % of them stayed

123

Plant Ecol

(a)

(b) Fenced Plots

80 60 40 20 0

80 60 40 20 0

(c)

Unfenced Plots

(d)

80 60 40 20 0

Unfenced Plots

100

Proportion of survival transplants (%)

100

80 60 40 20

15 Ju

n

20 n Ju

n

20

14

13 20

12 Ju

Ju

n

20

20 n Ju

O

11

0

ct 2 Ju 010 n N 201 ov 1 20 11 N ov 20 12 N ov 20 13 N ov 2 J u 01 n 4 20 15

Proportion of survival acorns (%)

Fenced Plots

100

Proportion of survival transplants (%)

Proportion of survival acorns (%)

100

(e) Natural Control

Number of survival acorns

80

Acorns added to the plots in 2010 or transplants from acorns planted in 2010 Acorns added to the plots in Jun 2011

60 40

Acorns added to the plots in Nov 2011 or transplants from acorns planted in Nov 2011

20

Acorns added to the plots in 2012 or transplants from acorns planted in 2012 Acorns added to the plots in 2013 or transplants from acorns planted in 2013

14 20

13 N

ov

20

12

ov N

ov

20

20 N

ov N

O

ct

20

10

11

0

Fig. 2 Survival dynamics of Lithocarpus hancei acorns (a in fenced plots; c in unfenced plots; e in natural control plots) and L. hancei seedling transplants (b in fenced plots; d in unfenced plots) from 2010 to 2015

alive for 2 years in fenced plots (Fig. 2b) and about 75 % survived from 2 weeks to 1 year and 10 seedlings survived for more than 3 years (Table S1). Similar to acorns, almost all of the transplanted seedlings died after 1 year in unfenced plots (Fig. 2d). About 50 % of the acorns and transplanted seedlings in unfenced plots survived for less than a month (Fig. 1). Almost all of the acorns in the natural control also survived no longer than 1 year (Fig. 2e).

123

Germination rates of acorns in fenced plots were significantly higher than those in unfenced plots and the natural control for all 5 times the acorns were sown to the plots during Oct 2010 to Nov 2013 and for total germination rates over the 5 years (Fig. 3a). The germination rate of acorns sown in Jun 2011 (collected in Oct 2010 and buried in soil for about 7 months) was significantly higher than the other 4 times (p \ 0.05). No acorns germinated in the natural control or

Plant Ecol

(b) 2.5 1 (Fence +, Litter +) 2 (Fence +, Litter −) 3 (Fence −, Litter +) 4 (Fence −, Litter −) 5 (Natural Control)

1

0.4 1 1 12

3 2

4

4

1 2

3 4

20

ta

13

12

11

l

3

3 4

N

ov

To

4

20

0.0

ov

5

N

3

20

12

2

0.5 1

Time

unfenced plots except those sown in Jun 2011. The Dheight of transplanted seedlings in fenced plots was also significantly greater, i.e. they grew taller, than those in unfenced plots (p \ 0.001) (Fig. 3b). However, there was no difference in Dheight of transplanted seedlings with or without litter in the fenced plots (p = 0.93) and in the unfenced plots (p = 0.57). Many factors contribute to the death of acorns (Fig. 4). About 9 and 5 % of the acorns were still alive at the end of the experiment in treatment 1 and 2 (fenced plots), respectively (Fig. 4a, b). Some of the living acorns became seedlings and others remained dormant. About 3 % of the acorns were still alive at the end of the experiment in treatment 3 (unfenced and with litter plots) (Fig. 4c) and all of the living ones were dormant. No acorns survived in treatment 4 (unfenced and no-litter plots) (Fig. 4d) or the natural control (Fig. 4e). About 65 % of the acorns were eaten or removed by rodents or other herbivores in fenced plots (Fig. 4a, b) because some rodents burrowed under the fence into the plots. Moreover, many acorns were rotten (about 20 %) and some of the seedlings from germinated acorns were withered (about 6 %). However, most of the acorns were eaten or removed in unfenced plots (Figs. 4c, d, 5a, b). In the natural control, 87 % of the acorns were eaten or removed, some were rotten and some were destroyed by weevil larvae (Fig. 4e). Acorns sown to the treatment plots had been selected by water floatation, so there were no acorns destroyed by weevil larvae. About 10 and 13 % of the transplanted seedlings were alive at the end of the experiment in treatment 1

1

20

2345

20

11

ov

4

1.0

13 To ta l

345

N

ov

20 n

20

11

10 20 Ju

ct O

345

3

ov

5

345

0.0

1

N

0.1

2 3

10

12

2 1

ov

4

2

0.2

1

1.5

N

0.3

1 (Fence +, Litter +) 2 (Fence +, Litter −) 3 (Fence −, Litter +) 4 (Fence −, Litter −)

2.0

20

0.5

2

ct

2

O

0.6

Seedling ΔHight (cm)

0.7

N

(a)

Germination Rate

Fig. 3 Seed germination rates in five treatments (a) and Dheight (cm) of transplanted seedlings in four treatments (b) from 2010 to 2015. The X axis is the date the acorns were sown in the plots (a) and the date that the acorns were sown into pots to produce seedlings transplanted to the plots (b). Error bars are 1 std. error. Numbers on top of bars indicate the treatment

Time

and 2 (fenced plots), respectively (Fig. 4f, g). The majority of transplanted seedlings withered in fenced plots (Fig. 4f, g). About 1 and 4 % of the transplanted seedlings were alive in treatment 3 and 4 (unfenced plots), respectively (Fig. 4h, i). The fate of the seedlings in unfenced plots was mainly being eaten or bitten by herbivores (Fig. 5b, c) and about 30 % of the seedlings were withered (Fig. 4h, i).

Discussion Effects of the treatments on acorns and transplants Fences provided protection for L. hancei acorns and seedlings to reduce herbivory by rodents and other mammals. Almost all of the acorns and seedlings were eaten, damaged or removed in unfenced plots but survival rate was significantly higher in fenced plots. Other research in Ailaoshan shows that L. hancei acorns are mainly eaten or hoarded by rodents, such as the Asian red-cheeked squirrel, the South China field mouse, the Chinese white-bellied rat (Niviventer coxingi Swinhoe) and chestnut rat (Niviventer fulvescens Gray) (Xiao and Zhang 2012) (Fig. 5a). These rodents depend heavily on Fagaceae acorns as an important food source. Although many studies have shown that litter reduces seedling emergence (Reader 1993; White 2014), in our study litter had a significant positive effect on acorn survival but had no effect on seedling establishment, neither those developed from germinated acorns in situ nor from seedling transplants. Since most rodents mainly

123

Plant Ecol

(a)

(c)

(b) Alive Bitten

Withered

Withered

(d) Rotten Alive

Alive

(e) Rotten Withered Rotten

Weevil larvae

Rotten Rotten

E/R

1. Fence+, Litter+

(f)

E/R E/R 3. Fence–, Litter+

2. Fence+, Litter–

(g) Alive Bitten

(h)

E/R 4. Fence–, Litter–

E/R 5. Natural Control

(i)

Bitten Eaten Alive

Alive

Alive Withered Bitten

Withered

Bitten

Withered

Withered

1. Fence+, Litter+

2. Fence+, Litter–

Eaten 3. Fence–, Litter+

Eaten 4. Fence–, Litter–

Fig. 4 Pie plots of the fate of seeds (acorns) (a–e) and transplanted seedlings (f–i) in five treatments. ‘‘E/R’’ means the acorns were eaten or removed by rodents or other mammals;

‘‘Bitten’’ means the seedlings were bitten and broken by herbivores including mammals or weevils

Fig. 5 Acorns and seedlings were eaten or bitten by herbivores. a An Asian red-cheeked squirrel eating acorns (photograph by Prof. Zhi-Shu Xiao, Institute of Zoology, Chinese Academy of

Sciences, Beijing); b, c Lithocarpus hancei seedlings bitten and broken by herbivores. The arrow in b shows an empty acorn hull

forage for food by olfaction (Price and Jenkins 1986), a thick litter layer makes it more difficult for them to find the acorns, thereby enhancing their survival (Cintra 1997; Li et al. 2006). In this study, litter had no significant effect on the survival of L. hancei seedlings most likely because the litter layer (about 4 cm deep) around the study area is not too thick for the shoot to

grow through and transplanted seedlings were also not influenced by the thin litter layer.

123

Causes of mortality Our study demonstrates that herbivores and pathogens were probably the most important sources of seed and

Plant Ecol

seedling mortality in L. hancei, a result consistent with Moles and Westoby (2004). Other sources of mortality may include low light levels, and at some times of the year desiccation. Acorns of L. hancei were removed or eaten by rodents once they had been shed which is similar to other seeds such as Quercus, Pinus, Vouacapoua, etc. (Vander Wall 1990; Jansen and Forget 2001; Xiao et al. 2005; Chang et al. 2012). In some species predators may remove or eat more than 90 % of available seeds (Dalling et al. 2011). The greatest losses of L. hancei acorns to rodents occurred during late fall immediately after their placement in the plots, and early spring when animal food resources are low, a result consistent with that of Gashwiler (1967). In this study, L. hancei acorns remained dormant for 5–12 months (mostly 6–9 months). A majority of the Fagaceae species have dormancy but there have not been many researches about it (Xiao et al. 2009). According to our observation, in genus Lithocarpus some species with thinner seed coat are nondormant, such as L. longinux (Hu) Y. C. Hsu & H. W. Jen and L. areca (Hickel & A. Camus) A. Camus, etc. Among the Fagaceae species in the Ailaoshan forest, Quercus variabilis Blume is one of the limited species that have no dormancy. Although many of the Fagaceae acorns have dormancy possibly due to the thick seed coat (physically dormant) (Burrows 1997), such a trait would be advantageous to a species in the absence of predation because this would increase the probability of acorn survival through the cold winter months and the dry season (from November to April the next year). However, it would be expected to be selectively disadvantageous in these forests where acorns are a preferred food for the rodents (Xiao and Zhang 2012). It is possible that although predation by rodents may cause high mortality, this mortality is overwhelmed in those sporadic years of mast seeding by L. hancei. In addition, during mast years acorn survival time increases because of food (acorn) saturation for the rodents. Thus, mast seeding is likely a plant trait selected by acorn predators (Kon et al. 2005). There were three mast years (2010, 2011 and 2013) and two nonmast years (in 2012 and 2014) during 2010–2014 in Ailaoshan. In all three mast years, especially 2013, there were abundant Fagaceae acorns including L. hancei. High seed density in mast years enhances recruitment because of predator satiation (Augspurger and Kitajima 1992; Vander Wall 2002), while in low seed density years, almost all the seeds are consumed by

animals (Jansen and Forget 2001; Vander Wall 2002). Acorns that fell on to the natural control plots in 2013 survived longer than in other years, a result consistent with previous studies indicating that high seed abundance helps some acorns escape predation (Vander Wall 2002; Xiao et al. 2005; Chang et al. 2012). Interestingly, in 2013 acorns in the natural control plots had higher survival rates and lived longer than those in other unfenced plots because high acorn densities could only be achieved in natural control plots; in all other treatment plots, all falling acorns were removed on a weekly basis and only the experimentally planted acorns allowed to remain. Pathogens and parasites are another primary source of mortality of seeds and seedlings (Augspurger 1984; Dalling et al. 2011). Pathogens and parasites are ubiquitous (Burdon and Leather 1990; Kohler and Wiley 1992) and yet, they are not considered a ‘‘trophic level’’, and it is virtually impossible to see disease organism by naked eyes, so they are difficult to incorporate into field tests (Polis and Strong 1996; Stiling and Rossi 1997). However, they are an important cause of seed and seedling mortality (Augspurger 1984; Dalling et al. 2011). The rotten L. hancei acorns in the fenced plots (about 20 % of all the acorns) were no doubt destroyed by fungi; the withered seedlings (about 6 % of all the acorns) and the majority of the transplants were also probably destroyed by fungi such as was reported by Lawrence and Rediske (1962). The high soil moisture in Ailaoshan along with relatively warm temperatures favours the development of pathogens (Schafer and Kotanen 2003). Conservation implications Lithocarpus hancei seedlings are infrequent in the understorey of these high diversity subtropical evergreen broad-leaved forests. This could be seen as a cause for concern considering that L. hancei is a dominant species in these forests. This concern could be amplified because although numerous acorns are produced annually most are eaten by rodents or rot. The acorns that do germinate to become seedlings, especially in mast years, are mostly eaten by herbivores or killed by fungal attack, what Harper (1977) referred to as ‘‘out of the frying pan and into the fire’’. However, these concerns may be misplaced because for persistence and long-term stability of the current

123

Plant Ecol

forest each individual simply has to replace itself. If the average L. hancei lives for 120 years (Young et al. 1992), then on average each individual has to leave only one successful offspring every 120 years. This requires two events. First, an adult has to die or be damaged to such a degree that there is an opening in the canopy. Death of mature adult trees is inevitable and damage is mostly caused by unusually harsh weather conditions. For example, many large branches of L. hancei were broken off by a severe heavy snowfall in winter of 2014 opening forest gaps and providing a suitable microenvironment for seedling regeneration. Second, the conditions in which these forests exist are relatively benign and this contributes to the high diversity of trees, insects, birds and other animals. The benign conditions ensure a fairly continuous high abundance of herbivores so the probability of tree seedling success remains low. Thus, successful reestablishment requires one episodic event, or ecological crunch, that radically reduces the number of herbivores for one or a few years. During these herbivore-reduced conditions, individuals already present as seedlings or saplings may be able to grow past the vulnerable stages to ‘escape height’ and successfully recruit to the community. Our ongoing research in the forest is now investigating the types of conditions that may temporarily reduce herbivory, permit recruitment, and thus maintain these high diversity forests. Acknowledgments Financial support was provided by a Joint Fund from the National Natural Science Foundation of China and Yunnan Provincial Government (Grant no. U1502231). We are grateful to the Ailaoshan Station for Subtropical Forest Ecosystem Research, Chinese Academy of Sciences, for permission and support to conduct this research in the Ailaoshan National Nature Reserve. We also thank Prof. ZhiShu Xiao, the Institute of Zoology, Chinese Academy of Sciences, Beijing, for providing the photograph of the Asian red-cheeked squirrel.

References Abbott HG (1961) White pine seed consumption by small mammals. J For 59:197–201 Augspurger CK (1984) Seedling survival of tropical tree species: interactions of dispersal distance, light-gaps, and pathogens. Ecology 65:1705–1712. doi:10.2307/1937766 Augspurger CK, Kitajima K (1992) Experimental studies of seedling recruitment from contrasting seed distributions. Ecology 73:1270–1284. doi:10.2307/1940675

123

Baskin CC, Baskin JM (2001) Seeds: ecology, biogeography, and evolution of dormancy and germination. Academic Press, San Diego Burdon JJ, Leather SR (1990) Pests, pathogens and plant communities. Blackwell Scientific Publications, London Burrows CJ (1997) Reproductive ecology of New Zealand forests: 1. Natural seed storage phenomena. N Z Nat Sci 23:31–52 Chang G, Jin TZ, Pei JF, Chen XN, Zhang B, Shi ZJ (2012) Seed dispersal of three sympatric oak species by forest rodents in the Qinling Mountains, Central China. Plant Ecol 213:1633–1642. doi:10.1007/s11258-012-0118-1 Cintra R (1997) Leaf litter effects on seed and seedling predation of the palm Astrocaryum murumuru and the legume tree Dipteryx micrantha in Amazonian forest. J Trop Ecol 13:709–725 Crow TR (1988) Reproductive mode and mechanisms for selfreplacement of northern red oak (Quercus rubra)—a review. For Sci 34:19–40 Dalgleish HJ, Shukle JT, Swihart RK (2012) Weevil seed damage reduces germination and seedling growth of hybrid American chestnut. Can J For Res 42:1107–1114. doi:10. 1139/x2012-067 Dalling JW, Davis AS, Schutte BJ, Arnold AE (2011) Seed survival in soil: interacting effects of predation, dormancy and the soil microbial community. J Ecol 99:89–95. doi:10. 1111/j.1365-2745.2010.01739.x Das CR, Mondal NK, Aditya P, Datta JK, Banerjee A, Das K (2012) Allelopathic potentialities of leachates of leaf litter of some selected tree species on gram seeds under laboratory conditions. Asian J Exp Biol Sci 3:59–65 Datta SC, Chatterjee AK (1980) Allelopathic potential of Polygonum orientale L. in relation to germination and seedling growth of weeds. Flora 169:456–465 Dechoum MS, Zenni RD, Castellani TT, Zalba SM, Rejma´nek M (2015) Invasions across secondary forest successional stages: effects of local plant community, soil, litter, and herbivory on Hovenia dulcis seed germination and seedling establishment. Plant Ecol 216:823–833. doi:10.1007/ s11258-015-0470-z Eckstein RL, Donath TW (2005) Interactions between litter and water availability affect seedling emergence in four familial pairs of floodplain species. J Ecol 93:807–816. doi:10.1111/j.1365-2745.2005.01015.x Editorial Committee for Vegetation of China (1980) Vegetation of China. Science Press, Beijing Eriksson O (1995) Seedling recruitment in deciduous forest herbs: the effects of litter, soil chemistry and seed bank. Flora 190:65–70 Garcı´a-Guzma´n G, Benı´tez-Malvido J (2003) Effect of litter on the incidence of leaf-fungal pathogens and herbivory in seedlings of the tropical tree Nectandra ambigens. J Trop Ecol 19:171–177. doi:10.1017/S0266467403003195 Gardiner ES, Hodges JD (1998) Growth and biomass distribution of cherrybark oak (Quercus pagoda Raf.) seedlings as influenced by light availability. For Ecol Manag 108:127–134. doi:10.1016/s0378-1127(98)00220-5 Gashwiler JS (1967) Conifer seed survival in a western Oregon clearcut. Ecology 48:431–438 Gong HD, Yang GP, Lu ZY, Liu YH (2011a) Diversity and spatial distribution patterns of trees in an evergreen broad-

Plant Ecol leaved forest in the Ailao Mountains, Yunnan. Biodivers Sci 19:143–150 (in Chinese with English abstract) Gong HD, Yang GP, Lu ZY, Liu YH, Cao M (2011b) Composition and spatio-temporal distribution of tree seedlings in an evergreen broad-leaved forest in the Ailao Mountains, Yunnan. Biodivers Sci 19:151–157. doi:10.3724/SP.J. 1003.2011.07010 (in Chinese with English abstract) Hamrick JL, Lee JM (1987) Effect of soil surface topography and litter cover on the germination, survival, and growth of musk thistle (Carduus nutans). Am J Bot 74:451–457. doi:10.2307/2443821 Harper JL (1977) Population biology of plants. Academic Press, London Jansen PA, Forget PM (2001) Scatterhoarding rodents and tree regeneration. In: Bongers F, Charles-Dominique P, Forget PM, The´ry M (eds) Nouragues: dynamics and plant-animal interactions in a neotropical rainforest. Kluwer Academic Publisher, Dordrecht, pp 275–288 Kitajima K, Augspurger CK (1989) Seed and seedling ecology of a monocarpic tropical tree, Tachigalia versicolor. Ecology 70:1102–1114. doi:10.2307/1941379 Klinger R, Rejma´nek M (2013) Experimental seed predator removal reveals shifting importance of predation and dispersal limitation in early life history stages of tropical forest trees. Folia Geobot 48:415–435. doi:10.1007/ s12224-013-9178-9 Kohler SL, Wiley MJ (1992) Parasite-induced collapse of populations of a dominant grazer in Michigan streams. Oikos 65:443–449 Kon H, Noda T, Terazawa K, Koyama H, Yasaka M (2005) Evolutionary advantages of mast seeding in Fagus crenata. J Ecol 93:1148–1155. doi:10.1111/j.1365-2745.2005.01040.x Lawrence WH, Rediske JH (1962) Fate of sown Douglas-fir seed. For Sci 8:210–218 Li QK, Ma KP (2003) Factors affecting establishment of Quercus liaotungensis Koidz. under mature mixed oak forest overstory and in shrubland. For Ecol Manag 176:133–146. doi:10.1016/s0378-1127(02)00274-8 Li HJ, Zhang HM, Zhang ZB (2006) Acorn removal of Liaodong oak (Quercus liaotungensis) by rodents. Acta Theriol Sin 26:8–12 Li S, Liu WY, Li DW, Song L, Shi XM, Lu HZ (2015) Species richness and vertical stratification of epiphytic lichens in subtropical primary and secondary forests in southwest China. Fungal Ecol 17:30–40. doi:10.1016/j.funeco.2015. 02.005 Liu WY, Fox JED, Xu ZF (2002) Biomass and nutrient accumulation in montane evergreen broad-leaved forest (Lithocarpus xylocarpus type) in Ailao Mountains, SW China. For Ecol Manag 158:223–235 Lorimer CG, Chapman JW, Lambert WD (1994) Tall understorey vegetation as a factor in the poor development of oak seedlings beneath mature stands. J Ecol 82:227–237. doi:10.2307/2261291 Loydi A, Eckstein RL, Otte A, Donath TW (2013) Effects of litter on seedling establishment in natural and semi-natural grasslands: a meta-analysis. J Ecol 101:454–464. doi:10. 1111/1365-2745.12033 McGee CE (1975) Change in forest canopy affects phenology and development of northern red and scarlet oak seedlings. For Sci 21:175–179

Moles AT, Westoby M (2004) What do seedlings die from and what are the implications for evolution of seed size? Oikos 106:193–199 Molofsky J, Augspurger CK (1992) The effect of leaf litter on early seedling establishment in a tropical forest. Ecology 73:68–77. doi:10.2307/1938721 Mun˜oz A, Bonal R, Espelta JM (2014) Acorn—weevil interactions in a mixed-oak forest: outcomes for larval growth and plant recruitment. For Ecol Manag 322:98–105. doi:10.1016/j.foreco.2014.02.039 Ni J (2001) Plant functional types and biomes of China at a regional scale. Acta Bot Sin 43:419–425 (in Chinese with English abstract) Polis GA, Strong DR (1996) Food web complexity and community dynamics. Am Naturalist 147:813–846 Price MV, Jenkins SH (1986) Rodents as seed consumers and dispersers. In: Murray DR (ed) Seed dispersal. Academic Press, Sydney, pp 191–235 Qi SS, Dai ZC, Miao SL et al (2014) Light limitation and litter of an invasive clonal plant, Wedelia trilobata, inhibit its seedling recruitment. Ann Bot 114:425–433. doi:10.1093/ aob/mcu075 Qi JH, Zhang YJ, Zhang YP, Liu YH, Lu ZY, Wu CS, Wen HD (2015) The influence of changes in water availability on seedling mortality of a subtropical evergreen broadleaf forest on Ailao Mountain. Acta Ecol Sin 35:2521–2528 (in Chinese with English abstract) Qiu XZ, Xie SC, Liu WY et al (1998) Studies on the forest ecosystem in Ailao Mountains, Yunnan. Yunnan Science and Technology Press, Kunming (in Chinese with English abstract) Rai JPN, Tripathi RS (1984) Allelopathic effects of Eupatorium riparium on population regulation of two species of Galinsoga and soil microbes. Plant Soil 80:105–117. doi:10.1007/BF02232944 Reader RJ (1993) Control of seedling emergence by ground cover and seed predation in relation to seed size for some old-field species. J Ecol 81:169–175. doi:10.2307/2261232 Rotundo JL, Aguiar MR (2005) Litter effects on plant regeneration in arid lands: a complex balance between seed retention, seed longevity and soil-seed contact. J Ecol 93:829–838. doi:10.1111/j.1365-2745.2005.01022 Schafer M, Kotanen PM (2003) The influence of soil moisture on losses of buried seeds to fungi. Acta Oecol 24:255–263. doi:10.1016/j.actao.2003.09.001 Shaw EW (1954) Direct seeding in the Pacific Northwest. J For 52:827–828 Sork VL (1983) Distribution of pignut hickory (Carya glabra) along a forest to edge transect, and factors affecting seedling recruitment. B Torrey Bot Club 110:494–506. doi:10. 2307/2996284 Stiling P, Rossi AM (1997) Experimental manipulations of topdown and bottom-up factors in a tri-trophic system. Ecology 78:1602–1606 Sunyer P, Boixadera E, Mun˜oz A, Bonal R, Espelta JM (2015) The interplay among acorn abundance and rodent behavior drives the spatial pattern of seedling recruitment in mature Mediterranean oak forests. PLoS One 10:e0129844. doi:10.1371/journal.pone.0129844 Sydes C, Grime JP (1981) Effects of tree leaf litter on herbaceous vegetation in deciduous woodland: II. An

123

Plant Ecol experimental investigation. J Ecol 69:249–262. doi:10. 2307/2259829 Thadani R, Ashton PMS (1995) Regeneration of banj oak (Quercus leucotrichophora A. Camus) in the central Himalaya. For Ecol Manag 78:217–224. doi:10.1016/ 0378-1127(95)03561-4 Vander Wall SB (1990) Food hoarding in animals. University of Chicago Press, Chicago and London Vander Wall SB (2002) Masting in animal-dispersed pines facilitates seed dispersal. Ecology 83:3508–3516 Wang ZJ, Carpenter C, Young SS (2000) Bird distribution and conservation in the Ailao Mountains, Yunnan, China. Biol Conserv 92:45–57. doi:10.1016/S0006-3207(99)00058-0 Wei TH, Wang ZJ, Cui QY (1988) The birds in northern and middle parts of the Ailaoshan Mountains. In: Xu YC, Jiang HQ (eds) Comprehensive survey reports of the Ailaoshan mountains natural reserve. Yunnan Ethnic Press, Kunming, pp 206–230 (in Chinese with English abstract) Willis CP (1914) The control of rodents in field seeding. Proc Soc Am For 9:365–379 White TJ (2014) The effects of Phragmites Australis litter on seed emergence in the Erie-Huron Corridor, Michigan. Master Thesis, Wayne State University Xiao ZS, Zhang ZB (2012) Behavioural responses to acorn germination by tree squirrels in an old forest where white oaks have long been extirpated. Anim Behav 83:945–951. doi:10.1016/j.anbehav.2012.01.013

123

Xiao ZS, Zhang ZB, Wang YS (2005) The effects of seed abundance on seed predation and dispersal by rodents in Castanopsis fargesii (Fagaceae). Plant Ecol 177:249–257. doi:10.1007/s11258-005-2321-9 Xiao ZS, Gao X, Steele MA, Zhang ZB (2009) Frequency-dependent selection by tree squirrels: adaptive escape of nondormant white oaks. Behav Ecol. doi:10.1093/beheco/ arp169 Young SS, Herwitz SR (1995) Floristic diversity and co-occurrences in a subtropical broad-leaved forest and two contrasting regrowth stands in central-west Yunnan Province, China. Vegetatio 119:1–13 Young SS, Carpenter C, Wang ZJ (1992) A study of the structure and composition of an old growth and secondary broad-leaved forest in the Ailao Mountains of Yunnan, China. Mt Res Dev 12:269–284. doi:10.2307/3673670 Zhao TG, Wu DL, Deng XF (1988) Mammals in the Ailaoshan mountain natural reserve. In: Xu YC, Jiang HQ (eds) Comprehensive survey reports of the Ailaoshan mountains natural reserve. Yunnan Ethnic Press, Kunming, pp 194–205 (in Chinese with English abstract) Zhu H, Yan LC (2009) List of seed plants in the Ailao Mts of Yunnan Province. Yunnan Science and Technology Press, Kunming, China