SciELO - Scientific Electronic Library Online

vol.33 issue1Arbuscular mycorrhizal fungal communities in soils under three phytophysiognomies of the Brazilian Atlantic ForestReproductive morphology and phenological aspects of one morphological variant of Hypnea pseudomusciformis (Gigartinales, Rhodophyta) author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Acta Botanica Brasilica

Print version ISSN 0102-3306On-line version ISSN 1677-941X

Acta Bot. Bras. vol.33 no.1 Belo Horizonte Jan./Mar. 2019  Epub Nov 29, 2018 


Fire records in tree rings of Moquiniastrum polymorphum: potential for reconstructing fire history in the Brazilian Atlantic Forest

Arno Fritz das Neves Brandes1  *

Andrea Sánchez-Tapia2

Jerônimo Boelsums Barreto Sansevero3

Rafael Perpetuo Albuquerque2

Cláudia Franca Barros2

1 Setor Botânica, Departamento de Biologia Geral, Instituto de Biologia, Universidade Federal Fluminense, 24020-141, Niterói, RJ, Brazil

2 Diretoria de Pesquisa Científica, Escola Nacional de Botânica Tropical, Instituto de Pesquisas Jardim Botânico do Rio de Janeiro, 22460-030, Rio de Janeiro, RJ, Brazil

3 Departamento de Ciências Ambientais, Instituto de Florestas, Universidade Federal Rural do Rio de Janeiro, 23890-000, Seropédica, RJ, Brazil


Fire disturbance affects the composition, structure and dynamics of vegetation. Historical records of fire events exist in some places, but they are generally limited in temporal and spatial extent. Tree-ring research is a useful tool for fire history reconstruction and can contribute important long-term ecological data. We tested the hypotheses that Moquiniastrum polymorphum (Less.) G. Sancho, a widespread species in Brazil that occurs in burnt areas of Atlantic Forest, produces annual growth rings and that its wood can record fire incidence by datable fire scaring. Our results corroborate these hypotheses and indicate that the species has potential for fire history reconstruction.

Keywords: Camará; Cambará; dendrochronology; fire history; fire scars; Gochnatia polymorpha; growth rings; Moquiniastrum polymorphum; pyrochronology


Fire disturbance can be the result of natural factors or human activity and affects the composition, structure and dynamics of vegetation (Thonicke et al. 2001). Dendrochronology is useful for the reconstruction of fire history (pyrochronology) and has been used in this way in both temperate and tropical forests (Worbes 1995; Grau et al. 2003; Lopez et al. 2012; Mundo et al. 2013; Rink & Thompson 2014). In tropical regions many species are known to produce distinct growth rings with potential for dendroecological research (Worbes 2002; Brienen et al. 2009; Rozendaal & Zuidema 2011). In the Atlantic Forest, growth ring research has helped to understand individual growth and species development, as well as the associated factors (Callado et al. 2001; Brandes et al. 2011; 2015; Andreacci et al. 2014; Shimamoto et al. 2014; Costa et al. 2015). Dendrochronology offers a temporal and spatial record of fire disturbance that is useful for understanding historical ecologies in this complex biome.

Historically, fire has been a common practice for clearing forest in the Atlantic Forest, a practice known as slash-and-burn (Dean 1996). Fire events remain frequent in the Atlantic Forest, with more than 136,000 outbreaks being recorded from 2007 to 2016, 37 % of which have been in the last three years (INPE 2017). Therefore, impacts of fire at the community scale, such as decreases in tree species richness and abundance (Melo & Durigan 2010; Sansevero et al. 2017), may be underestimated since fire data are restricted to a few places and over a short time span. Tree-ring analysis could expand the temporal and spatial extent of fire records. In trees and shrubs, the occurrence of fire produces scaring, which is preserved in woody tissue. In long-lived woody species with distinct annual growth rings, these scars have proven useful for dating fire occurrence and periodicity (McBride 1983; Fritts & Swetnam 1989; Worbes 1995; Stahle 1999).

Moquiniastrum polymorphum (Asteraceae) (new combination for Gochnatia polymorpha), is a shrub or small tree that is widely distributed in Brazil, having been reported from the Northeast, Central-West, Southeast and South regions of the country. It occurs in the Atlantic Forest and Cerrado biomes, more precisely in cerrado (lato sensu), gallery forest, seasonal forest and ombrophilous forest vegetation types (Moquiniastrum in Flora do Brasil 2020 em construção 2017). Popularly known as Camará and Cambará, the national conservation status for the species is “Least Concern” (CNCFlora 2017), although regionally it is considered “Vulnerable" according to the red list of threatened plants for the state of Rio Grande do Sul (State Decree number 42,099 from December 31, 2002). The species dominates some early regeneration stages in areas previously covered by ombrophilous forest that have been disturbed by deforestation, slash-and-burn practices and fire (Uhlmann et al. 2014; Sansevero et al. 2017). It is also freeze resistant (Brando & Durigan 2005) and survives fire (Kolb 1993; Sánchez-Tapia 2011) due to its thick bark. Its wide distribution, local relative abundance and fire resistance make M. polymorphum a good candidate for pyrochronological study in the Atlantic Forest.

In the present study, we demonstrate that M. polymorphum produces annual growth rings and that when exposed to the fire, fire scars appear and the year of these events can be determined. These results have important implications for detecting past fire events and for understanding the ecological resilience and shifts in species composition of communities subjected to fire in the Brazilian Atlantic Forest, a global biodiversity hotspot.

Materials and methods

Research was conducted at Poço das Antas Biological Reserve (PABR), (22°30'-22°33'S 42°15'-42°19'W) in the state of Rio de Janeiro, Brazil. This reserve occupies approximately 5,160 hectares that include forest and non-forest phytophysiognomies (Lima et al. 2006). Non-forest formations with anthropic interference have a significant presence in the study area (47.9 %). Anthropic actions, including wood exploitation, fire, artificial drainage and river course modification, have contributed to the attrition of forest species in these areas (Lima et al. 2006). Fire has been reported to have been frequent in recent decades (Lima et al. 2006). Since 1990, PABR has recorded and mapped all fire events by official technical reports. Topography of the area is characterized by plains and undulations with hills and hillocks with rounded profiles varying from 19 to 200 m in elevation and separated by flat-bottom floodplain wetlands. The climate is of the Aw type according to the Köppen-Geiger system (tropical with a dry season in winter). Average annual rainfall is 1,995 mm and the average annual temperature is 25.5 °C (Lima et al. 2006; Moraes et al. 2006).

Moquiniastrum polymorphum (Less.) G. Sancho is a representative species of the floristic composition of secondary forests and areas in early regeneration in PABR (Lima et al. 2006). In disturbed areas, the species occurs in high densities and appears to play an important ecological role in post-fire succession (Neves & Peixoto 2008; Sánchez-Tapia 2011; Prieto et al. 2017), due to its ability to survive and resprout quickly following fire.

We sampled 53 individuals in sites where we had historical records of fire, according to PABR technical reports, with burning marks at the base and with or without externally visible wounds. Stem samples were collected from plants in areas with recorded incidence of fire in the winter: 1990, 14 samples; 2002, 17 samples; 2005, 2 samples; 2002 and 2005, 5 samples; and 2010, 15 samples (Tab. 1). Samples were collected in July 2007 and April 2012 from the main stem using a hacksaw and chisel (section, 31 samples), and an increment borer (22 samples). We collected samples from where external wounds were visible (nine samples) and, when external wounds were not visible, from where the bark appeared more burnt (44 samples). The wood samples were deposited in the wood collection of the Herbário de Niterói (NITw).

Table 1  Sample number, year of fire incidence, year of collection, method of collection, and the presence of fire scars in the year of fire incidence. *** = Individual with externally visible wound. 

Sample number Collection year Collection method Fire year Fire scar
NITw909 2007 section 2002 present***
NITw910 2007 section 2002 present
NITw911 2007 section 2002 present
NITw912 2007 section 2002 present
NITw913 2007 section 2002 present
NITw914 2007 section 2005 present
NITw915 2007 section 2005 present
NITw916 2007 section 2005 / 2002 present***
NITw917 2007 section 2005 / 2002 present
NITw918 2007 section 2005 / 2002 present
NITw919 2007 section 2005 / 2002 present***
NITw920 2007 section 2005 / 2002 present***
NITw921 2012 increment borer 1990
NITw922 2012 increment borer 1990
NITw923 2012 increment borer 1990
NITw924 2012 increment borer 1990
NITw925 2012 increment borer 1990
NITw926 2012 increment borer 1990
NITw927 2012 increment borer 1990
NITw928 2012 increment borer 1990
NITw929 2012 increment borer 1990
NITw930 2012 section 1990
NITw931 2012 section 1990
NITw932 2012 section 1990
NITw933 2012 section 1990
NITw934 2012 section 1990
NITw935 2012 increment borer 2002
NITw936 2012 increment borer 2002
NITw937 2012 increment borer 2002
NITw938 2012 increment borer 2002
NITw939 2012 increment borer 2002
NITw940 2012 section 2002 present***
NITw941 2012 section 2002 present***
NITw942 2012 section 2002
NITw943 2012 section 2002
NITw944 2012 section 2002
NITw945 2012 section 2002
NITw946 2012 section 2002
NITw947 2012 increment borer 2010
NITw948 2012 increment borer 2010
NITw949 2012 increment borer 2010
NITw950 2012 increment borer 2010
NITw951 2012 increment borer 2010
NITw952 2012 increment borer 2010
NITw953 2012 increment borer 2010
NITw954 2012 increment borer 2010
NITw955 2012 section 2010 present***
NITw956 2012 section 2010 present***
NITw957 2012 section 2010 present***
NITw958 2012 section 2010
NITw959 2012 section 2010
NITw960 2012 section 2010
NITw961 2012 section 2010

Collected samples were polished with the help of a razor and/or an orbital sander using sandpaper with progressively finer grit to facilitate the observation of, and clear distinction between, growth rings. Observations were performed with a stereoscopic microscope (Leica). Dating of growth rings followed Schulman’s (1956) convention for dendrochronological research in the Southern Hemisphere. We evaluated the presence of fire scars for the recorded fire dates. The annual periodicity of growth ring formation was evaluated by the correspondence of fire scars with the known dates of fire (Worbes 1995; Lopez et al. 2012).

For anatomical description of the growth rings in bright-field microscopy, samples were softened in boiling water and glycerin, 12 to 30 µm-thick sections were made in transversal plain using a Spencer 860 sliding microtome. After cleaning, they were stained with safranin and astra blue (Bukatsch 1972), dehydrated, and mounted on Entellan resin in permanent slides. Slides were observed using a Primo Star Zeiss microscope, and images were captured with ZEN software for Windows linked to the microscope through a Media Axiocam ERC 5s video camera.

Results and discussion

Moquiniastrum polymorphum produces annual growth rings, which can exhibit visible scars from fire events such as pith flecks, traumatic canals and external wounds that expose wood. We observed such scars in 17 individuals that corresponded with the dates of fires reported by PABR (Tab. 1). In individuals where fire damaged a wide area of the vascular cambium, an external wound would be present where new wood was not produced and which would contrast with adjacent areas where the cambium remained productive (Fig. 1A). In samples where fire had damaged a more limited area of the vascular cambium, pith flecks and traumatic canals were produced, while the vascular cambium reestablished continuity to produce secondary xylem and secondary phloem (Fig. 1B).

Figure 1  Transverse sections (stereomicroscopy) of M. polymorphum. A. Sample with externally visible wound. Fire incidence in 2002. B. Sample without externally visible wounds. Fire incidence in 2002. C. Sample collected with increment borer without fire scars. Bars indicate growth ring boundaries. Black arrow indicates fire scar. Clear arrow indicates differentiated early wood. Scale bars = 1 cm. 

Moquiniastrum polymorphum can be regarded as fire resistant because 36 of the individuals exposed to fire did not show scars (Tab. 1), which may be related to its thick bark (Sansevero 2013). This finding is in agreement with in situ observations made by several researchers (Neves & Peixoto 2008; Sánchez-Tapia 2011) but represents the first anatomical observational confirmation of fire resistance for the species. Bark thickness has been shown to be an important feature for stem survival through fire (Lawes et al. 2011).

None of the samples collected with the increment borer showed fire scars (Fig. 1C, Tab. 1), but it is unclear whether these individuals were undamaged by fire or whether the limited sample-size obtained by the increment borer failed to obtain tissue from fire-wounded areas of the stem. In some samples, differentiated early wood was detected that was similar to the early wood observed adjacent to fire scars (Fig. 1A-B), which was characterized by abundant axial parenchyma (Fig. 2A-B), however, these were not considered fire scars.

Figure 2  Transverse sections (bright-field microscopy) of M. polymorphum. A - Fire scar and adjacent differentiated early wood with abundant axial parenchyma. Scale bar = 200 µm. B - Differentiated early wood with abundant axial parenchyma. Scale bar = 100 µm. Clear arrow indicates differentiated early wood. Black arrow indicates growth ring boundary. 

The annual growth rings of this species are delimited by tangentially arranged early wood vessels and radially flattened late wood fibers (Fig. 3A-B). The anatomy of growth rings match with features described by Sonsin et al. (2014). Tomazello-Filho et al. (2004) described M. polymorphum as having scarcely distinct annual growth rings, but in our samples we detected distinct rings regardless of the collection method.

Figure 3  Transverse sections of M. polymorphum with growth ring boundaries delimited by tangentially arranged early wood vessels and radially flattened late wood fibers. A - Bright-field microscopy. Scale bar = 100 µm. B - Stereomicroscopy. Scale bar = 1 mm. Black arrows indicate growth ring boundaries. 

This result indicates that M. polymorphum has potential for reconstructing fire history. Grau et al. (2003) suggest that pyrochonology may be useful in the reconstruction of historic fire regimes when three important features are present, specifically, the fire must have been abundant in the system; indicator species must be damaged by fire so as to produce visible scars, but yet survive fire damage; and fire scars must be datable. Previous studies in PABR have reported the persistence of M. polymorphum after fire (Neves & Peixoto 2008; Sánchez-Tapia 2011), indicating that many individuals of this species did not die as a result. We sampled live individuals with datable fire scars and post-fire growth. Technical reports from PABR report an incidence of fire (size and duration of events) sufficient to produce a historical record. Populations of M. polymorphum in PABR appear seem to meet the requirements necessary for establishing a dendrochronological record of fire regimes in this ecosystem.

The number of scars in the stems of M. polymorphum may not represent the total number of fire episodes, given that fire scars were not always recorded after fire passage. The intensity of fire, fire type (surface fire, crown fire and ground fire) and other environmental conditions would affect the fire record (Taylor & Skinner 2003; Speer 2010). Lack of scars in some trees does not indicate that fire event does not wounded other trees on nearby sites and does not reduce the potential of the species for reconstructing fire history. To overcome this condition, researches for reconstruction of fire histories perform collection of many trees and individuals with externally visible fire scars (Grau et al. 2003; Taylor & Skinner 2003; Horne & Fulé 2006; Everett 2008). In the present study all individuals with visible external wound recorded fire events.

We conclude that M. polymorphum produces annual growth rings and seems to produce datable fires scars that can be used in the reconstruction of fire history. Considering the widespread distribution of M. polymorphum (two different biomes in Brazil) our results have important implications for future research that investigates fire events on a broad scale. For future research using this approach we recommend that individuals with externally visible fire scars be selected and that sections, as opposed to increment borer samples, be used for sampling wood.


We are grateful to Reserva Biológica de Poço das Antas for permission and Adilson Martins Pintor for his help conducting fieldwork. The Programa de Pesquisa em Biodiversidade (PPBio), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Projeto Nacional de Ações Integradas Público-Privadas para Biodiversidade (Probio II), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ) and Global Environment Facility (GEF) all provided financial support for this project.


Andreacci F, Botosso P, Galvão F. 2014. Sinais Climáticos em anéis de crescimento de Cedrela fissilis em diferentes tipologias de florestas ombrófilas do sul do Brasil. Floresta 44: 323-332. [ Links ]

Brandes AFN, Lisi CS, Barros CF. 2011. Dendrochronology of lianas of the Leguminosae family from the Atlantic Forest, Brazil. Trees 25: 133-144. [ Links ]

Brandes AFN, Lisi CS, Silva LDSAB, Rajput KS, Barros CF. 2015. Seasonal cambial activity and wood formation in trees and lianas of Leguminosae growing in the Atlantic Forest: a comparative study. Botany 93: 211-220. [ Links ]

Brando PM, Durigan G. 2005. Changes in cerrado vegetation after disturbance by frost (São Paulo State, Brazil). Plant Ecology 175: 205-215. [ Links ]

Brienen RJW, Lebrija-Trejos E, Breugel M, et al. 2009. The potential of tree rings for the study of forest succession in Southern Mexico. Biotropica 41: 186-195. [ Links ]

Bukatsch F. 1972. Bemerkungen zur doppelfa¨rbung astrablau-safranin. Mikrokosmos 61: 33-36. [ Links ]

Callado C, Neto SS, Scarano F, Costa C. 2001. Periodicity of growth rings in some flood-prone trees of the Atlantic Rain Forest in Rio de Janeiro, Brazil. Trees 15: 492-497. [ Links ]

CNCFlora - Centro Nacional de Conservação da Flora. 2017. Gochnatia polymorpha subsp. polymorpha in Lista Vermelha da Flora Brasileira versão 2012. polymorpha subsp. polymorpha . 12 Apr. 2017. [ Links ]

Costa MS, Ferreira KEB, Botosso PC, Callado CH. 2015. Growth analysis of five Leguminosae native tree species from a seasonal semidecidual lowland forest in Brazil. Dendrochronologia 36: 23-32. [ Links ]

Dean W. 1996. A ferro e fogo: a história e a devastação da Mata Atlântica brasileira. São Paulo, Companhia das Letras. [ Links ]

Everett RG. 2008. Dendrochronology-based fire history of mixed-conifer forests in the San Jacinto Mountains, California. Forest Ecology and Management 256: 1805-1814. [ Links ]

Flora do Brasil 2020 em construção . 2017. Moquiniastrum. Jardim Botânico do Rio de Janeiro. . 12 Apr. 2017. [ Links ]

Fritts HC, Swetnam TW. 1989. Dendroecology: a tool for evaluating variations in past and present forest environments. Advances in Ecological Research 19: 111-188. [ Links ]

Grau H, Easdale T, Paolini L. 2003. Subtropical dendroecology-dating disturbances and forest dynamics in northwestern Argentina montane ecosystems. Forest Ecology and Management 177: 131-143. [ Links ]

INPE - Instituto Nacional de Pesquisas Espaciais. 2017. Programa queimadas. . 12 Apr. 2017 [ Links ]

Horne ML, Fulé PZ. 2006. Comparing methods of reconstructing fire history using fire scars in a southwestern United States ponderosa pine forest. Canadian Journal of Forest Research 36: 855-867. [ Links ]

Kolb SR. 1993. Islands of secondary vegetation in degraded pastures of Brazil: their role in reestablishing Atlantic Coastal Forest. PhD Thesis, University of Georgia, USA. [ Links ]

Lawes MJ, Adie H, Russell-Smith J, Murphy B, Midgley JJ. 2011. How do small savanna trees avoid stem mortality by fire? The roles of stem diameter, height and bark thickness. Ecosphere 2: 1-13. [ Links ]

Lima HC, Pessoa SVA, Guedes-Bruni RR, et al. 2006. Caracterização fisionômico-florística e mapeamento da vegetação da Reserva Biológica de Poço das Antas, Silva Jardim, Rio de Janeiro, Brasil. Rodriguésia 53: 369-389. [ Links ]

Lopez L, Villalba R, Peña-Claros M. 2012. Determining the annual periodicity of growth rings in seven tree species of a tropical moist forest in Santa Cruz, Bolivia. Forest Systems 21: 508-514. [ Links ]

McBride JR. 1983. Analysis of tree rings and fire scars to establish fire history. Tree-Ring Bulletin 43: 51-67. [ Links ]

Melo ACG, Durigan G. 2010. Impacto do fogo e dinâmica da regeneração da comunidade vegetal em borda de Floresta Estacional Semidecidual (Gália, SP, Brasil). Revista Brasileira de Botânica 33: 37-50. [ Links ]

Moraes LFD, Assumpção JM, Luchiari C, Pereira TS. 2006. Plantio de espécies arbóreas nativas para a restauração ecológica na Reserva Biológica de Poço das Antas, Rio de Janeiro, Brasil. Rodriguésia 57: 477-489. [ Links ]

Mundo IA, Kitzberger T, Juñent FAR, Villalba R, Barrera MD. 2013. Fire history in the Araucaria araucana forests of Argentina: Human and climate influences. International Journal of Wildland Fire 22: 194-206. [ Links ]

Neves GMS, Peixoto AL. 2008. Florística e estrutura da comunidade arbustivo-arbórea de dois remanescentes em regeneração de Floresta Atlântica secundária na Reserva Biológica de Poço das Antas, Silva Jardim, Rio de Janeiro. Pesquisa Botânica 59: 71-112. [ Links ]

Prieto PV, Seger GDS, Sánchez-Tapia A, Sansevero JBB, Braga JMA, Rodrigues PJF. 2017. Secondary succession and fire disturbance promote dominance of a late-diverging tree lineage in a lowland Neotropical forest. Plant Ecology and Diversity 10: 311-322. [ Links ]

Rink WJ, Thompson J. 2014. Encyclopedia of scientific dating methods. Dordrecht, Springer. [ Links ]

Rozendaal DMA, Zuidema PA. 2011. Dendroecology in the tropics: a review. Trees 25: 3-16. [ Links ]

Sánchez-Tapia A. 2011. Regeneração natural e restauração ecológica em capoeiras submontanas de Mata Atlântica submetidas a queimadas. MSc Thesis, Universidade Federal do Rio de Janeiro, Rio de Janeiro. [ Links ]

Sansevero JBB. 2013. Classificação de grupos funcionais e caracterização de trajetórias sucessionais na Floresta Atlântica. PhD Thesis, Instituto de Pesquisas Jardim Botânico do Rio de Janeiro / Escola Nacional de Botânica Tropical, Rio de Janeiro. [ Links ]

Sansevero JBB, Prieto PV, Sanchez-Tapia A, Braga JMA, Rodrigues PJFP. 2017 Past land-use and ecological resilience in a lowland Brazilian Atlantic Forest: implications for passive restoration. New Forests 48: 573-586. [ Links ]

Schulman E. 1956. Dendroclimatic changes in semiarid America. Tucson, University of Arizona Press. [ Links ]

Shimamoto CY, Botosso PC, Marques MCM. 2014. How much carbon is sequestered during the restoration of tropical forests? Estimates from tree species in the Brazilian Atlantic forest. Forest Ecology and Management 329: 1-9. [ Links ]

Sonsin JO, Gasson P, Machado SR, Caum C, Marcati CR. 2014. Atlas da diversidade de madeiras do cerrado paulista. Botucatu, Fundação de Estudos e Pesquisa Agrícolas e Florestais. [ Links ]

Speer JH. 2010. Fundamentals of tree-ring research. Tucson, The University of Arizona Press. [ Links ]

Stahle DW. 1999. Useful strategies for the development of tropical tree-ring chronologies. Iawa Journal 20: 249-253. [ Links ]

Taylor AH, Skinner CN. 2003. Spatial patterns and controls on historical fire regimes and forest structure in the Klamath mountains. Ecological Applications 13: 704-719. [ Links ]

Thonicke K, Venevsky S, Sitch S, Cramer W. 2001. The role of fire disturbance for global vegetation dynamics: Coupling fire into a dynamic global vegetation model. Global Ecology and Biogeography 10: 661-677. [ Links ]

Tomazello-Filho M, Lisi CS, Hansen N, Cury G. 2004. Anatomical features of increment zones in different tree species in the State of São Paulo, Brazil. Scientia Forestalis 66: 46-55. [ Links ]

Uhlmann A, Bonnet A, Curcio GR, Silva AP, Gonçalves FLA, Resende AS. 2014. A cobertura vegetal das florestas e pastagens. In: Prado RB, Fidalgo ECC, Bonnet A. (eds.) Monitoramento da revegetação do COMPERJ-Etapa inicial. EMBRAPA, Brasília. P. 223-244 [ Links ]

Worbes M. 1995. How to measure growth dynamics in tropical trees: a review. Iawa Journal 16: 337-351. [ Links ]

Worbes M. 2002. One hundred years of tree-ring research in the tropics-a brief history and an outlook to future challenges. Dendrochronologia 20: 217-231. [ Links ]

Received: August 16, 2018; Accepted: September 24, 2018

* Corresponding author:

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License