Search

The effective and efficient monitoring of revegetation outcomes is a key component of ecosystem restoration. Monitoring often involves labor‐intensive manual methods, which are difficult to deploy when sites are inaccessible or involve large areas of revegetation. This study aimed to identify plant species and quantify α‐diversity index on a sub‐meter scale at Manlailiang Mine Site in Northwestern China using unmanned aerial vehicles (UAVs) as a means to semiautomate large‐scale vegetation monitoring. UAVs equipped with multispectral sensors were combined with three industry‐standard supervised classification algorithms (support vector machine [SVM], maximum likelihood, and artificial neural network) to classify plant species. Spectral vegetation indices (normalized difference vegetation index [NDVI], difference vegetation index [DVI], visible‐band difference vegetation index, soil‐adjusted vegetation index, modified soil‐adjusted vegetation index, and excess green–excess red) were used to assess vegetation diversity obtained from on‐ground survey plot data (the Margalef, Pielou, Simpson, and Shannon–Wiener indices). Our results showed that SVM outperformed other algorithms in species identification accuracy (overall accuracy of 84%). Significant relationships were observed between vegetation indices and diversity indices, with the DVI performing significantly better than many more commonly used indices such as the NDVI. The findings highlight the potential of combining UAV multispectral data, spectral vegetation indices and ground surveys for effective and efficient fine‐scale monitoring of vegetation diversity in the ecological restoration of mining areas. This has significant practical benefits for improving adaptive management of restoration through improved monitoring tools. [ABSTRACT FROM AUTHOR]

Tropical lowland rainforests are considered to be ecologically important refugia of national and global significance, yet wide‐scale deforestation and degradation of tropical lowland rainforests lead to losses in biodiversity. Seed storage behavior information is required for the effective restoration of these forests. Here, we studied the seed storage behavior of 42 selected tropical lowland rainforest tree species from Sri Lanka with the aim to test the reliability of current experimental methods and predictive models to predict the desiccation sensitivity of seeds. Seed storage behavior was experimentally determined through the 100 seed method and the results were compared with three predictive models; thousand seed weight–moisture content (TSW–MC), seed coat ratio–seed mass (SCR–SM), and phylogenetic affiliation models. Based on the 100 seed method, 28 species were identified as desiccation sensitive and 14 species were identified as desiccation tolerant. Compared to the 100 seed method, the predictability of the SCR–SM model was 70%, whereas the TSW–MC and phylogenetic affiliation models showed 52% and 58% predictability, respectively. Due to its higher predictability, the SCR–SM model has a high potential to be an effective predictive tool of seed storage behavior. This study revealed that tropical lowland rainforests are dominated by trees producing desiccation‐sensitive seeds. Further, species dispersing seeds in the peak rainy season have desiccation‐sensitive seeds, while those dispersing seeds in the normal wet season have desiccation‐tolerant seeds. Another key finding was that 100% of the subcanopy species produce nonorthodox, that is, desiccation‐sensitive seeds, which gradually decrease from subcanopy to understory. [ABSTRACT FROM AUTHOR]

Open science is vital to the interdisciplinary field of ecology due to its integrative nature and use of longitudinal datasets that build upon earlier data collections. To highlight the importance of open science in the rapidly growing discipline of restoration ecology, we conducted a "computational reproducibility" assessment of a publication on a mining restoration program spanning several decades and over 250 km2 in a global biodiversity hotspot. Open data and code provided alongside the original publication were assessed for consistency with the results and conclusions of the original publication, as were potential limitations in findings due to the methodology. The impacts of inconsistencies and limitations were qualitatively assessed against the key findings from the publication and data were reanalyzed where impacts were potentially significant. Of the six inconsistencies and limitations identified, two had a significant impact on 5 of the 11 key findings of the original publication and one new key finding was made. The impact of this is of high ecological significance as the findings related to key restoration parameters: species richness (similarity of species richness between forest and 25‐year‐old restoration), functional diversity (correlation of species richness and functional diversity), and the restoration trajectory (long‐term trends and restored areas' resilience to disturbance). These outcomes highlight the importance of open data and the value of detailed third‐party data reviews, particularly in restoration ecology which relies on research findings to inform decision‐making and policy and drive adaptive management. [ABSTRACT FROM AUTHOR]