This is a repository copy of Drought generates large, long-term changes in tree and liana
regeneration in a monodominant Amazon forest.
White Rose Research Online URL for this paper:
https://eprints.whiterose.ac.uk/161899/
Version: Accepted Version
Article:
Marimon, BS, Oliveira-Santos, C, Marimon-Junior, BH et al. (8 more authors) (2020)
Drought generates large, long-term changes in tree and liana regeneration in a
monodominant Amazon forest. Plant Ecology. ISSN 1385-0237
https://doi.org/10.1007/s11258-020-01047-8
© Springer Nature B.V. 2020. This is an author produced version of a journal article
published in Plant Ecology. Uploaded in accordance with the publisher's self-archiving
policy.
eprints@whiterose.ac.uk
https://eprints.whiterose.ac.uk/
Reuse
Items deposited in White Rose Research Online are protected by copyright, with all rights reserved unless
indicated otherwise. They may be downloaded and/or printed for private study, or other acts as permitted by
national copyright laws. The publisher or other rights holders may allow further reproduction and re-use of
the full text version. This is indicated by the licence information on the White Rose Research Online record
for the item.
Takedown
If you consider content in White Rose Research Online to be in breach of UK law, please notify us by
emailing eprints@whiterose.ac.uk including the URL of the record and the reason for the withdrawal request.
1
Drought generates large, long-term changes in tree and liana regeneration in a 1
monodominant Amazon forest 2
3
Beatriz S. Marimon
a,b,
*, Claudinei Oliveira-Santos
c
, Ben Hur Marimon-Junior
a,b
, Fernando Elias
d
, 4
Edmar A. de Oliveira
a
, Paulo S. Morandi
a
, Nayane C. C. dos Santos Prestes
b
, Lucas H. Mariano
b
, 5
Oriales R. Pereira
a
, Ted R. Feldpausch
e
and Oliver L. Phillips
f
6
7
a
Universidade do Estado de Mato Grosso (UNEMAT), Departamento de Ciências Biológicas, 8
Laboratório de Ecologia Vegetal, Caixa Postal 08, CEP 78.690-000, Nova Xavantina, MT, Brazil; 9
b
Universidade do Estado de Mato Grosso (UNEMAT), Programa de Pós-graduação em Ecologia e 10
Conservação, Caixa Postal 08, CEP 78.690-000, Nova Xavantina, MT, Brazil; 11
c
Universidade Federal de Goiás (UFG), Programa de Pós-Graduação em Ciências Ambientais, 12
CEP 74.690-900, Goiânia, GO, Brazil; 13
d
Universidade Federal do Pará (UFPA)/Embrapa Amazônia Oriental, Programa de Pós-14
graduação em Ecologia, CEP 66.075-110, Belém, PA, Brazil; 15
e
College of Life and Environmental Sciences, University of Exeter, Exeter, EX4 4RJ, UK; 16
f
School of Geography, University of Leeds, LS2 9JT, Leeds, UK. 17
*Corresponding author. E-mail: biamarimon@unemat.br (Faxphone: +55 6634381224) 18
19
Acknowledgments 20
This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível 21
Superior – Brazil (CAPES) – Finance Code 001. We also thank the Conselho Nacional de 22
Desenvolvimento Científico e Tecnológico/Projetos Ecológicos de Longa Duração - CNPq/PELD 23
(Nr. 401279/2014-6 and 441244/2016-5), and Fundação de Amparo à Pesquisa do Estado de Mato 24
Grosso, Project RedeFlor 0589267/2016, for financial support. B.S. Marimon and B.H. Marimon-25
Junior acknowledge CNPq for their productivity grants (305029/2015-0 and 303680/2016-4), and 26
P.S. Morandi acknowledge CAPES for his post-doc grant (88887.185186/2018-00). O.L. Phillips 27
was supported by an ERC Advanced Grant (Tropical Forests in the Changing Earth System) and a 28
Royal Society Wolfson Research Merit Award. 29
5,298 words 30
2
Abstract 31
The long-term dynamics of regeneration in tropical forests dominated by single tree species remains 32
largely undocumented, yet is key to understanding the mechanisms by which one species can gain 33
dominance and resist environmental change. We report here on the long-term regeneration 34
dynamics in a monodominant stand of Brosimum rubescens Taub. (Moraceae) at the southern 35
border of the Amazon forest. Here the climate has warmed and dried since the mid-1990’s. Twenty-36
one years of tree and liana regeneration were evaluated in four censuses in 30 plots by assessing 37
species abundance, dominance, and diversity in all regeneration classes up to 5 cm diameter. The 38
density of B. rubescens seedlings declined markedly, from 85% in 1997 to 29% in 2018 after the 39
most intense El Niño-driven drought. While the fraction contributed by other tree species changed 40
little, the relative density of liana seedlings increased from just 1% to 54% and three-quarters of 41
liana species underwent a ten-fold or greater increase in abundance. The regeneration community 42
experienced a high rate of species turnover, with changes in the overall richness and species 43
diversity determined principally by lianas, not trees. Long-term maintenance of monodominance in 44
this tropical forest is threatened by a sharp decline in the regeneration of the monodominant species 45
and the increase in liana density, suggesting that monodominance will prove to be a transitory 46
condition. The close association of these rapid changes with drying indicates that monodominant B. 47
rubescens forests are impacted by drought-driven changes in regeneration, and therefore are 48
particularly sensitive to climatic change. 49
50
Keywords: drought • regeneration dynamics • saplings • seedlings • lianas. 51
52
Introduction 53
Tropical forests are renowned as being global centres of tree species richness and diversity 54
(Connell et al. 1984; Gentry 1988). In general, tropical forests are also hyper-diverse at the local, 55
community scale, with one hectare having as much as 30 times more tree species than an equivalent 56
3
area in temperate forests (e.g., Gentry 1988; Condit et al. 1996; Torti et al. 2001). Such high alpha 57
diversity typically pertains across the most extensive tropical forests of all in Amazonia (e.g., ter 58
Steege et al. 2003), yet for some tropical forests the rule of high alpha-diversity does not hold. 59
Across the tropics, several studies have reported ‘monodominant’ mature forests even on well-60
drained soils, in which one species may comprise from 50% to as much as 100% of the individuals 61
and the total biomass (Connell and Lowman 1989; Marimon et al. 2001a; Peh et al. 2011b). 62
Several researchers have attempted to identify factors and mechanisms that enable a single 63
species to reach and maintain monodominance in the tropics (e.g., Connell and Lowman 1989; Hart 64
1990; Torti et al. 2001; Marimon and Felfili 2006; Marimon et al. 2008, 2012, 2014; Peh et al. 65
2014; Nascimento et al. 2017; Elias et al. 2018). Peh et al. (2011b) discussed a total of eight 66
hypotheses that have been proposed to explain the origin and maintenance of monodominance in 67
tropical forests. They concluded that a variety of mechanisms likely interact to induce a species to 68
attain monodominance locally, even when that species also grows in similar environmental 69
conditions in adjacent mixed forests which have much greater tree diversity. 70
Meanwhile, it has long been appreciated that disturbance regimes can have a strong impact 71
on tropical forest diversity; Connell’s ‘intermediate disturbance’ hypothesis (Connell 1978), for 72
example, suggests that there is an optimal level of disturbance frequency and intensity above and 73
below which species diversity declines. Indeed disturbance regimes appear to be intimately 74
associated with the phenomenon of monodominance (Tovar et al. 2019). Thus, while Ibanez and 75
Birnbaum (2014) observed that monodominance can occasionally be a non-persistent state that 76
occurs after severe disturbances (‘early successional monodominance’), one of the factors most 77
often hypothesised as responsible for favouring a species to reach monodominance is a long-term 78
lack of large-scale disturbances (Connell and Lowman 1989; Hart 1990; Marimon et al. 2012). In 79
this situation, species whose seedlings are able to grow under deep shade, and which also have high 80
parental survivorship and potential to dominate canopies are expected to gain a long-term 81
advantage. In all, multiple related biological traits and environmental characteristics are likely to be 82
4
responsible for helping a species develop a monodominant state, including low gap formation 83
frequency, strong interspecific competition, high parental survivorship and high survivorship under 84
strong shading, litter characteristics, large seed size, and masting events (i.e., massive, infrequent 85
seed production in regional synchrony with trees of the same species) (Connell and Lowman 1989; 86
Hart 1990; Torti et al. 2001; Marimon and Felfili 2006; Peh et al. 2011b; Marimon et al. 2012; Hart 87
2012; Read et al. 2017; Henkel and Mayor 2019). 88
Evidently then most of the factors and mechanisms invoked in explanations of 89
monodominance may be best evaluated through examining species regeneration (Connell and 90
Lowman 1989; Read et al. 2017). For example, if diversity in tropical forests is normally 91
maintained by compensatory mechanisms that benefit rare species (Connell et al. 1984), then 92
monodominance may occur when regeneration mechanisms uncharacteristically instead favour 93
common species and tree diversity remains low. More generally, evaluation of long-term 94
regeneration dynamics should be able to determine whether changes in diversity occur in 95
conjunction with changes in disturbance mechanisms, such as droughts, logging and anthropogenic 96
climate change – all of which impact tree mortality, recruitment, carbon sequestration and species 97
composition in Amazonia (Phillips et al. 2009; Costa et al. 2010; Brienen et al. 2015; Meir et al. 98
2015; Feldpausch et al. 2016; Esquivel-Muelbert et al. 2019). Yet studies of the stand dynamics of 99
monodominant forest regeneration are difficult, extremely few, and so far have relied on single 100
census inventories or short-term monitoring (c.f., Hart 1995; Marimon et al. 2012; Valverde-101
Berrantes and Rocha 2014). 102
As well as understanding species regeneration processes, exploring the ecological 103
interactions between trees and lianas is also essential for understanding how tropical forests 104
function (Caballé and Martin 2001; Comita et al. 2007). Lianas not only compete effectively with 105
trees for water, light, space and nutrients (e.g., van der Heijden and Phillips 2009), but by 106
contributing to gap formation they can accelerate processes of species substitution and forest 107
dynamics (e.g., Phillips et al. 2005; van der Heijden and Phillips 2009; van der Heijden et al. 2013; 108