Karst ecosystems in southern China are species-rich and have high levels of endemism, yet little is known regarding the evolutionary processes responsible for the origin and diversification of karst biodiversity. The genus Primulina (Gesneriaceae) comprises ca. 170 species endemic to southern China with high levels of ecological (edaphic) specialization, providing an exceptional model to study the plant diversification in karsts. We used molecular data from nine chloroplast and 11 nuclear regions and macroevolutionary analyses to assess the origin and cause of species diversification due to palaeoenvironmental changes and edaphic specialization in Primulina. We found that speciation was positively associated with changes in past temperatures and East Asian monsoons through the evolutionary history of Primulina. Climatic change around the mid-Miocene triggered an early burst followed by a slowdown of diversification rate towards the present with the climate cooling. We detected different speciation rates among edaphic types, and transitions among soil types were infrequently and did not impact the overall speciation rate. Our findings suggest that both global temperature changes and East Asian monsoons have played crucial roles in floristic diversification within the karst ecosystems in southern China, such that speciation was higher when climate was warmer and wetter. This is the first study to directly demonstrate that past monsoon activity is positively correlated with speciation rate in East Asia. This case study could motivate further investigations to assess the impacts of past environmental changes on the origin and diversification of biodiversity in global karst ecosystems, most of which are under threat.

SignificanceIdentifying and explaining regional differences in tropical forest dynamics, structure, diversity, and composition are critical for anticipating region-specific responses to global environmental change. Floristic classifications are of fundamental importance for these efforts. Here we provide a global tropical forest classification that is explicitly based on community evolutionary similarity, resulting in identification of five major tropical forest regions and their relationships: (i) Indo-Pacific, (ii) Subtropical, (iii) African, (iv) American, and (v) Dry forests. African and American forests are grouped, reflecting their former western Gondwanan connection, while Indo-Pacific forests range from eastern Africa and Madagascar to Australia and the Pacific. The connection between northern-hemisphere Asian and American forests is confirmed, while Dry forests are identified as a single tropical biome.

No abstract

A large image dataset plays a crucial role in building automatic vision recognition system. However, collecting and labeling data are tedious, laborious and time-consuming tasks. In some cases, it is chicken and egg problem: it is only possible to get application data after the system deployment. In our study, we are interested in building automatic plant identification systems from images. As plants distribution on the world is not uniform and may change in response to the availability of resources, the availability of species in different areas is different. That is why some species are very abundant in one region and non-existing in others regions. Even the distribution of plant species is diverse, plant species in the planet share common features. They all have organ types such as leaf, flower, etc. Taking into this observation, in this paper, we propose a new approach for building an image-based plant identification without an available image database based on the combination of deep learning, transfer learning, and crowd-sourcing. The proposed approach consists of four main steps: plant organ detection, plant image collection, data validation and plant identification. Plant organ detection aims to learn organ type characteristic from available image datasets of plants while the purpose of the data collection step is to crawl dataset from crowd-sourced sources. Then, plant organ detection will be used in data validation in order to remove the unwanted/invalid images while keeping the valid ones. Finally, plant identification method will be developed and evaluated from the new image dataset. We illustrate and demonstrate the use of the proposed approach for building a Vietnamese medicinal plant retrieval system. Index Terms-Organ detection, plant identification, deep learning, convolutional neural network.

This paper examined how forest has contributed to rural households’ livelihood in Da river basin, the northwest mountainous region of Vietnam. The results revealed that forest predominantly contributes to the total income of rural residents in the region. Specifically, forestry land area, access to non-timber forest products, and payment for forest environmental services significantly affected household’s income in the region. However, rural people in the region have still faced several difficulties that constrain household’s livelihood. Of these difficulties, lack of financial capital, epidemic diseases in animal husbandry, limited access to market information and natural disaster are popular barriers to livelihood of people in the region. This paper also recommended several policies to improve rural livelihood in Da river basin. These includes: (i) integrating issues regarding payment for forest environmental services and REDD+ into socioeconomic development plan; (ii) improving awareness of local people on sustainable natural capital use through ecosystem conservation policy; (iii) providing preferential credit and training on agricultural production techniques; and (iv) encouraging market-oriented agriculture.