New palaeontological assemblage, sedimentological and chronological data from the Pleistocene Ma U'Oi cave (northern Vietnam)

Aim Current species distributions are shaped by present and past biotic and abiotic factors. Here, we assessed whether abiotic factors (habitat availability) in combination with past connectivity and a biotic factor (body mass) can explain the unique distribution pattern of Southeast Asian mammals, which are separated by the enigmatic biogeographic transition zone, the Isthmus of Kra (IoK), for which no strong geophysical barrier exists. Location Southeast Asia. Taxon Mammals. Methods We projected habitat suitability for 125 mammal species using climate data for the present period and for two historic periods: mid‐Holocene (6 ka) and last glacial maximum (LGM 21 ka). Next, we employed a phylogenetic linear model to assess how present species distributions were affected by the suitability of areas in these different periods, habitat connectivity during LGM and species body mass. Results Our results show that cooler climate during LGM provided suitable habitat south of IoK for species presently distributed north of IoK (in mainland Indochina). However, the potentially suitable habitat for these Indochinese species did not stretch very far southwards onto the exposed Sunda Shelf. Instead, we found that the emerged landmasses connecting Borneo and Sumatra provided suitable habitat for forest dependent Sundaic species. We show that for species whose current distribution ranges are mainly located in Indochina, the area of the distribution range that is located south of IoK is explained by the suitability of habitat in the past and present in combination with the species body mass. Main conclusions We demonstrate that a strong geophysical barrier may not be necessary for maintaining a biogeographic transition zone for mammals, but that instead a combination of abiotic and biotic factors may suffice.

Section: Discussionsupporting

confidence: 62%

Distributions of mammals in Southeast Asia: The role of the legacy of climate and species body mass

Radchuk

Kramer‐Schadt

Fickel

et al. 2019

“…Our past potential distribution maps predicted the presence of both L. herberti and L. neilli in northern Thailand and northern Vietnam during the LGM and are concordant with the available fossil data if we consider their LGM distribution as representative of their distribution during other Pleistocene glacial periods and that the fossil teeth from Chaimanee () and Bacon et al . () belong to a Leopoldamys species other than L. sabanus . Leopoldamys sabanus teeth have also been identified in eastern Java and are representative of the Punung fauna of early LIG age (between 118 ± 3 ka and 128 ± 15 ka) (Storm et al ., ; Westaway et al ., ), thus corroborating our predicted LIG distribution for L. sabanus .…”

Section: Discussionmentioning

confidence: 98%

“…a). This area encompasses two relevant biogeographical regions where the genus Leopoldamys is currently distributed and where fossils of this genus have been discovered (Medway, ; Zheng, ; Chaimanee, ; Storm et al ., ; Bacon et al ., ).…”

Section: Methodsmentioning

confidence: 99%

Influence of past and future climate changes on the distribution of three Southeast Asian murine rodents

Latinne

Meynard

Herbreteau

et al. 2015

Aim We tested the influence of Pleistocene climatic fluctuations and the potential effect of future climate change on Southeast Asian small mammal distributions using two forest-dwelling (Leopoldamys herberti and Leopoldamys sabanus) and one karst (Leopoldamys neilli) endemic rodent species as models.Location Southeast Asia.Methods We used presence-absence data of genetically identified individuals, bioclimatic variables and species distribution modelling techniques to predict potential distributions of the three studied species under current, past [Last Interglacial (LIG) and Last Glacial Maximum (LGM)] and future conditions. We applied a variety of modelling techniques and then used consensus techniques to draw up robust maps of potential distribution ranges at all stages.Results According to our models, these three Leopoldamys species did not experience significant range contraction during the LGM. Our models revealed substantial range contraction during the LIG for L. herberti in northern Indochina, while its distribution expanded in southern Indochina. Evidence of a southward range expansion during that period was also obtained for L. neilli, whereas L. sabanus remained widely distributed in insular Southeast Asia but experienced a range contraction on the Thai-Malay Peninsula. The two future climate change scenarios used predicted that large climatically suitable areas would still be available in the future for the three species. Main conclusions Our model predictions contradict the well-established hypothesis that Southeast Asian forest-dwelling species were confined to small refugia during the LGM. Moreover, our results suggest that some Southeast Asian taxa may have been distributed in their refugial state during the LIG rather than the LGM. This could be because of vegetation changes that may have occurred at that time as a result of the increased seasonality observed during the LIG. These Pleistocene refugia may have been localized in northern Indochina but our study also revealed that southern Indochina could provide major potential refugia.

“…Bacon et al. (2008b) suggested a humid forested area with some open areas for the Late Pleistocene site of Duoi U’Oi ( c. 66 ka), whereas the site of Ma U’Oi ( c. 49 ka) in northern Vietnam showed drier and more open conditions (Bacon et al. , 2006).…”

Section: Discussionmentioning

confidence: 99%

Palaeoecology of Southeast Asian megafauna‐bearing sites from the Pleistocene and a review of environmental changes in the region

Louys

Meijaard

2010

100

Aim To reconstruct the palaeoenvironments of megafauna‐bearing sites from Pleistocene Southeast Asia, and to describe general environmental changes in the region. Location Indochina and Sundaland, including Myanmar, Laos, Vietnam, Cambodia, Thailand, the Malay Peninsula, Borneo, Sumatra and Java. Methods This study reconstructs the habitat types of 25 Pleistocene sites in Southeast Asia through a synecological (community‐based) method. This method specifically targets medium‐ and large‐bodied mammals, and ecovariables that could be directly assessed from species lists were chosen. The methods allow the reconstruction of fossil sites as closed (continuous tree cover), mixed (heterogeneous tree cover) and open (very limited to no tree cover) through discriminant functions analysis of community guild structure. Results Four Pleistocene sites can confidently be assigned to one of the three habitat types. Tam Hang (south), a Middle Pleistocene site from Laos, is classified as mixed. Ban Fa Suai, a Middle Pleistocene site from Thailand, is also classified as mixed. Trinil, a Middle Pleistocene site from Java, is classified as open. Lastly, Hang Hum II, a Late Pleistocene site from Vietnam, is classified as open. Insufficient species are present in the fossil faunas of remaining sites to allow statistically confident habitat assignment. Nevertheless, conditional habitat assignments can be achieved, and these are largely congruent with other palaeoenvironmental evidence. Main conclusions Medium‐ and large‐bodied mammals are the most frequently recovered mammals from Pleistocene sites in Southeast Asia. Previous palaeoenvironmental reconstructions of these sites have been hampered by this body size bias, as well as by limited site and faunal descriptions. However, our analysis demonstrates that reconstructions can still be achieved for megafauna‐bearing sites in the region. The reconstructions suggest that through much of the Pleistocene, Southeast Asia had significant areas of mixed habitats, and that the widespread distribution of rain forests, such as found today, was a relatively rare phenomenon.