Aims: Our objective is to provide a better understanding on the background and community structure of fish diversity in inland waters of Shanghai. 

Methods: We use specimens from 13 rivers and one reservoir in Shanghai to investigate the species resources and community structure of fish in the inland water bodies of Shanghai. These specimens were examined qualitatively and quantitatively from July to August and then from October to November in 2023. 

Results: A total of 20,160 specimens were collected of which16 orders, 24 families, 57 genera, and 71 species are identified. Included are Cypriniformes, Gobiiformes, and Siluriformes which account for 57.74%, 9.86%, and 5.63% of the total species, respectively. There are 57 species of freshwater fish, 11 species of estuarine fish, 3 species of migratory fish, and 4 exotic or artificial species. The high taxonomic composition of series Percomorpha and the rich ecological types of estuarine fish represent the characteristics of fish composition in estuarine tidal river networks in eastern China. The 14 water bodies average 35.9 species of fish each. The Punan Canal and Pudong Canal have the largest number with 47 species each, and Dishui Lake has the smallest number with 24 species. There are 15 dominant species and 14 common species with the average being 5 dominant species and 10 common species in each water body. Carassius auratus, Acheilognathus macropterus, Culter dabryi, and Chanodichthys erythropterus are the dominant species in the 14 water bodies. The richness index (D), diversity index (H'), evenness index (J'), and dominance index (C) show fish diversity is higher in the 13 rivers such as Yuanxie River and lowest in Dishui Lake. The βc diversity index of each water body ranged from 4.0–15.5 with an average of 9.9. A large species diversity is shown by the βr diversity index ranging from 12.44 to 28.76 with an average of 19.8. The ABC curve shows W is negative in most water bodies, and the abundance dominance curve is mostly above the biomass curve, showing the characteristics of community structure dominated by small and medium-sized individual fish. The fish communities in 14 water bodies could be divided into 4 groups by cluster analysis. 

Conclusion: The similarities within the groups are approximately 74%, with the exception of the first group, and the differences between the groups are 31.42%–49.65%. The highest contribution of differences is Carassius auratus, Pseudorasbora parva, and Chanodichthys mongolicus. Our results suggest the differences of fish community structure of inland water bodies in Shanghai may be caused by natural conditions, regional location, release activities of the water bodies, and the land use forms where the water bodies are located. The water of the national central city with dense buildings can also maintain a high diversity of fish species under the premise of protecting the water environment and fish habitat.

Aims: Environmental DNA (eDNA) technology provides a non-invasive approach for biodiversity monitoring. Recent studies have demonstrated that airborne eDNA collected from the air can be utilized to monitor wildlife in forest ecosystems. Compared with other eDNA survey methods, airborne eDNA offers greater flexibility in sampling site selection, especially in survey areas lacking environmental media such as water bodies. Therefore, airborne eDNA holds significant potential for biodiversity monitoring in forest ecosystems. This study aims to assess the effectiveness of airborne eDNA in monitoring terrestrial vertebrate diversity in tropical rainforests. 

Methods: The research was conducted in a 20-hectare forest dynamics plot in Xishuangbanna, China. Airborne eDNA technology was employed to survey terrestrial vertebrate diversity, and results were compared with those obtained from infrared camera monitoring data. A total of 20 airborne eDNA samplers were deployed, and three 24-hour sampling sessions were conducted over six days in November 2023. Collected samples were amplified using 12SV05 primers targeting the 12S rRNA gene fragment and sequenced on the Illumina NovaSeq 6000 platform. Sequences data were taxonomically annotated, and species detection efficiency was compared between airborne eDNA and infrared cameras. 

Results: Across the three Airborne eDNA sampling sessions, 66 operational taxonomic units (OTUs) were detected, representing birds, mammals, reptiles, and amphibians. In comparison, 20 infrared cameras placed at the same sites recorded 15 mammal species and 15 bird species over 5,682 camera days. Comparative analyses revealed that airborne eDNA was more effective in detecting species diversity. Additionally, evaluation of the alpha diversity accumulation curve indicated that diversity accumulation curve indicated that the diversity plateaued when 10 airborne eDNA samples were collected. Suggesting that within the current experimental conditions, 10 samples collected over three days are sufficient for maximizing species detection. 

Conclusion: This study demonstrates that airborne eDNA serves as an effective tool for monitoring terrestrial vertebrate diversity in tropical rainforests, enabling rapid and comprehensive biodiversity assessments. Compared to infrared cameras, airborne eDNA shows distinct advantages in rapid species detection. Although the technology is still under development and its stability and accuracy under specific environmental conditions require further improvement, ongoing advancements are likely to establish airborne eDNA as a critical tool for cross-trophic and multi-species biodiversity monitoring. It holds strong potential to support large-scale, standardized biodiversity monitoring networks in China.

Aims: Niche differentiation is an important mechanism for carnivores to coexist when they have similar functional or morphological characteristics. Of the most common methods of niche differentiation (e.g., spatial, nutritional, and temporal niches), temporal niche differentiation (i.e., activity-rest cycles within circadian rhythms) is often considered the easiest for animals to adjust. The manner in which these sympatric small- and medium-carnivores maintain their temporal niche has attracted increasing attention from researchers, but is not well-studied. 

Methods: We conducted a systemic camera-trapping survey in Chishui Alsophila National Nature Reserve Areas, Guizhou Province, China (GCANNR) to investigate patterns of coexistence between sympatric small- and medium-sized carnivores (i.e., Prionailurus bengalensis, Paguma larvata, Viverricula indica, Arctonyx collaris, Melogale moschata, Mustela kathiah, M. nivalis and M. sibirica). We detected 2,299-independent photograph of carnivores and used kernel density estimation and overlap coefficient index (Δ) to examine how circadian activity rhythms for each species differed in the presence or absence of human disturbance (i.e., outside or within GCANNR, respectively). We also quantified the temporal differentiation of these carnivores. 

Results: The circadian activity rhythms of small- and medium-sized carnivores revealed that the dominant species P. larvata, V. indica, and M. moschata and the common species P. bengalensis exhibited nocturnal activity patterns. Meanwhile, the common species A. collaris and M. kathiah, and the rare species M. nivalis and M. sibirica, showed cathemeral activity patterns. There was high overlap in the circadian activity rhythms of nocturnal dominant species (Δ > 0.90, 95%CI, 0.79–0.97) between within and outside of GCANNR. Despite this, dominant carnivores outside GCANNR had a peak in early-morning activity 1–2 h earlier than those inside GCANNR. Common species had a low overlap coefficient for their circadian activity rhythms between within and outside of GCANNR, as well as those of rare species (Δ < 0.80). Further analysis was conducted on the overlapping activity rhythms among the sympatric dominant and common species, which use the same trapping sites. P. bengalensis and A. collaris showed the lowest overlap coefficient (Δ = 0.49, 95% CI, 0.28–0.69). Meanwhile, there was no significant temporal differentiation in the activity rhythms among the three dominant species (P. larvata, V. indica and M. moschata; Δ > 0.90), and the overlap coefficient index (Δ) was 0.6–0.8 between the other species pairs. 

Conclusion: Circadian activity rhythms of the sympatric small- and medium-sized carnivores in GCANNR were affected by the human activities. These carnivores were able to coexist largely due to temporal differentiation. Most of the common species avoided the dominant species in temporal activity. In contrast, the coexistence among dominant species in the same distribution was not promoted by temporal niche differentiation, so the role of nutrient niche differentiation may have been more important. Thus, our results indicated there were patterns of temporal differentiation for several sympatric small-and-medium sized carnivores, but nutritional or spatial niche differentiation also aided in integrating these carnivores, helping them coexist. These results indicate that it is important to pay attention to the ecological niche differentiation pattern of species while managing wildlife protection areas; doing so would help formulate more effective conservation policies.

Aims: To understand the complex genetic structure and biogeographic history of the brown algae Fucus distichus in the North Pacific, the Northeast Atlantic, and the Northwest Atlantic, including the identification of a marine glacial refugium in the east of Newfoundland, Canada. 

Methods: We sampled five F. distichus populations from the Grand Banks of Newfoundland, and conducted PCR-based amplification and sequencing of two mitochondrial markers: 23S rRNA-tRNA-Val intergenic spacer (IGS) and COX1. By integrating these samples with molecular datasets published in 2011, we calculated the number of haplotypes, haplotype diversity (h) and nucleotide diversity (π) for each marker. We also constructed haplotype networks and evaluated phylogenetic affinities among haplotypes using maximum-likelihood estimation and neighbor-joining trees for each marker. 

Results: IGS data showed that F. distichus populations from the Grand Banks each harbored 3-4 haplotypes of which most are private, whereas other populations from the North Pacific, the Northeast and the Northwest Atlantic mostly had 1–2 haplotypes. In particular, the Grand Banks populations exhibited much higher haplotype (mean h = 0.6533) and nucleotide diversity (mean π = 0.0067) than other populations (mean h = 0.1487; mean π = 0.0022), with the highest genetic indices. Haplotype networks inferred from IGS and COX1 both showed that the ancestral haplotype was widely distributed in the Northeast and the Northwest Atlantic, including the Grand Banks. Phylogenetic trees further revealed a clear genetic divergence between the private haplotypes in Grand Banks and others elsewhere. These phylogeographic results indicate that F. distichus populations on both sides of the North Atlantic experienced multiple large-scale extinction events due to sea-level fluctuations driven by glacial-interglacial cycles during the late Pleistocene. Afterwards, the surviving ancestor of F. distichus in the Arctic recolonized the Northeast Atlantic prior to the last glacial maximum, following with a trans-Atlantic migration from Europe to North America possibly during the Holocene. Our phylogeographic results also suggest that the Flemish Cap located to the east of the Grand Banks of Newfoundland, Canada was potentially a marine glacial refugium during the late Pleistocene ice ages. 

Conclusion: Phylogeographic diversity patterns and processes can be influenced by various kinds of environmental factors. Adding geographically unique specimens such as isolated or ice-age survived populations during the paleoclimate change can largely expand our understanding of how species responded to historical environmental change, particularly the dynamic survival relics and dispersal routes associated with population diversification and speciation. These phylogeographic insights are also valuable for guiding natural resource conservation and management, and understanding of climate-driven ecological adaptation.

Hubei Province, located in central China, features a complex and diverse topography that provides heterogeneous habitats for rich flora. Lycophytes and ferns are important elements in biodiversity. However, they were not included in the Flora of Hubei (1976–1998). To comprehensively document the diversity and distribution of lycophytes and ferns in Hubei Province, we compiled a checklist integrating data from the Catalogue of Life China: 2024 Annual Checklist, the National Specimen Information Infrastructure (NSII), published books, and journal articles. Families and genera were classified based on the Pteridophyte Phylogeny Group I (PPG I) system. Our results identified 31 families, 99 genera, 570 species, 3 subspecies and 27 varieties of lycophytes and ferns in Hubei Province. Among these, lycophytes comprised 2 families, 8 genera, 38 species and 1 subspecies, while ferns comprised 29 families, 92 genera, 532 species, 2 subspecies and 27 varieties. In addition, this study confirmed that 42 species previously recorded in Hubei Province were not actually distributed there, and we questioned the authenticity of 37 species reported in the province. To improve the accuracy of plant documentation in Hubei Province, we recommend strengthening field investigations, and ensuring rigor and precision in specimen collection records. Our results provide a valuable dataset for future study on lycophytes and ferns in Hubei Province.

Aims: The Yarlung Tsangpo River is a crucial river ecosystem on the Tibetan Plateau, and protozoa, as an essential component of eukaryotic microorganisms, play a key role in aquatic ecosystems. This study aimed to investigate the diversity patterns and geographical distribution of protozoan communities in the middle reaches of the Yarlung Tsangpo River during the dry season, focusing on their community structure, diversity, and environmental response mechanisms in both the mainstem and tributaries. 

Method: In the dry season of 2021, we conducted a comprehensive study of protozoan communities in the middle reaches of the Yarlung Tsangpo River. We analyzed community structure, diversity, and environmental drivers using high-throughput sequencing, generating ASVs (amplicon sequence variants). Statistical analyses included distance-decay, environmental-decay, neutral and null models, and co-occurrence network analysis to assess spatial and environmental influences on protozoan assemblages. 

Results: A total of 6,066 ASVs were detected, spanning 23 phyla, 55 classes, 86 orders, 114 families, 281 genera, and 323 species. Significant differences in community structure were observed between the mainstem and tributaries, with species turnover identified as the primary driver of β-diversity variations. Distance-decay and environmental-decay analyses revealed that protozoan communities in tributaries were more sensitive to geographical and environmental changes. Neutral and null model analyses indicated that community assembly in both habitats was primarily driven by homogenous selection, though stochastic processes played a greater role in tributaries. Co-occurrence networks showed distinct environmental influences on protozoan communities in the mainstem versus tributaries. 

Conclusion: These findings highlight the spatial heterogeneity and environmental response mechanisms of protozoan communities in river ecosystems. The study underscores the importance of tributaries as sensitive indicators of environmental and geographical changes, while emphasizing the dominant role of deterministic processes in community assembly. These insights contribute to a deeper understanding of microbial ecology in high-altitude river systems.

Wildlife diversity monitoring is an important tool for assessing progress in conservation programs in protected areas. For example, Gaowangjie National Nature Reserve is located in Guzhang County, Hunan Province, in the hinterland of Wuling Mountain. This reserve covers an area of 17,169.8 ha, and it is home to 12 species of wildlife under national protection, such as small Indian civet (Viverricula indica), leopard (Prionailurus bengalensis), Elliot’s pheasant (Syrmaticus ellioti) and Koklass pheasant (Pucrasia macrolopha). Camera-trapping technology is a practice often used in wildlife monitoring, providing effective support for species cataloging in nature reserves. The aim of this study, therefore, was to use infrared camera technology to assess the success of conservation efforts in the Gaowangjie National Nature Reserve. Infrared camera monitoring records from 2021–2023 in the Gaowangjie National Nature Reserve were collated and species identification was performed. The dataset included 42,947 camera working days and captured a total of 45,838 images. Among them, there were 19,972 animal records and 8,039 independent detections, covering 12 species of animals in 3 orders and 8 families. There were 5,694 bird records and 2,212 independent detections, involving 43 species of birds in 9 orders and 23 families. By showing detailed coordinates of the camera sites, the working days of the cameras, the independent detections of birds and animals, and the list of bird and animal species monitored, these data can aide in revealing the biodiversity of a protected area. This insight provides an important foundation for developing evidence-based conservation plans and management strategies in protected areas.

Aims: Meconopsis, the second largest genus in the Papaveraceae, is widely distributed across the Eastern Himalayas and Hengduan Mountains and is highly valued for its ornamental qualities. Recently, numerous new taxa, based on subtle morphological differences, have been described from Xizang, Yunnan, and Sichuan, creating considerable challenges in the taxonomy of this genus. This study aims to comprehensively revise the Meconopsis species in China, clarifying their species classifications and geographical distributions, and provide an updated checklist for the genus. 

Methods: This research is based on an extensive review of taxonomic literature, databases, specimen identification, and field surveys, along with molecular systematics data, to comprehensively update the catalogue of Meconopsis species in China. 

Results: This study clarifies for the first time the horizontal and vertical distribution ranges of the genus Meconopsis in China, as well as the county-level distribution information for each species within the genus. The study confirms that Meconopsis in China includes 61 species, 4 subspecies, and 3 varieties, with 34 endemic species. We present an updated catalogue and identification key for the genus, as well as brief discussions on selected taxa. We synonymize several species: M. pseudointegrifolia subsp. daliensis Grey-Wilson, M. prattii (Prain) Prain, M. atrovinosa T. Yoshida & H. Sun, M. biluoensis Y. Fan & L. Wei, M. lhasaensis Grey-Wilson, M. horridula subsp. drukyulensis Grey-Wilson, and M. merakensis T. Yoshida, R. Yangzom & D. G. Long. M. zhongdianensis Grey-Wilson is downgraded to a variety, and M. purpurea T. Yoshida & H. Sun and M. pulchella var. melananthera T. Yoshida are newly combined and also downgraded to varieties. Additionally, this paper reports two newly recorded Meconopsis species in China, M. ludlowii Grey-Wilson and M. sinuata Prain, as well as one newly recorded Meconopsis species in Xizang, M. sherriffii G. Taylor. 

Conclusion: Based on the latest taxonomic research findings, this study provides updated catalogue information and distribution data for the genus Meconopsis in China, offering a reference for future taxonomic studies and species conservation of Meconopsis in China. Future efforts should focus on targeted field surveys of the Meconopsis, combining morphological traits with modern molecular systematics to further refine its classification system and distribution range.

Aims：Scientific protection and effective management of wildlife highly rely on baseline data regarding biological resources, and the precision of survey methods directly affects the objectivity and accuracy of resource assessments. Environmental DNA technology (eDNA) has been widely used in fish species monitoring, but its application in amphibian resource surveys remains relatively limited and is mostly focused on the detection of specific species. This study uses the Hunan Mangshan National Nature Reserve as an example to evaluate the detection efficiency and accuracy of eDNA technology in amphibian surveys. 

Methods：eDNA test data were collected from 19 water sampling sites within the reserve between July to August 2023. These data were then compared with data collected during the same period using traditional transect survey methods. By calculating alpha and beta diversity indices, a comprehensive evaluation of the two methods was conducted. 

Results：Both methods detected 34 amphibian species, with 24 species identified by both approaches. Significant differences (P < 0.05) were observed between the two methods in terms of the Shannon-Wiener diversity index, Simpson diversity index, and Pielou’s evenness index. Non-metric multidimensional scaling (NMDS) analysis showed significant differences in species composition between the two methods (P < 0.05), potentially influenced by species behavior, sampling location and timing, and environmental disturbances. 

Conclusion：This study demonstrates that while eDNA technology offers clear advantages in amphibian resource survey, it cannot yet completely replace traditional transect methods. It is suggested that eDNA techniques be integrated with traditional approaches to obtain more comprehensive and accurate baseline data on amphibian resources, thereby supporting more informed conservation and management efforts.

Aims: Analyzing biodiversity patterns across multiple dimensions such as species, evolution, and function is essential for deeply understanding biological diversification and informing effective conservation strategies. As the province with the richest biodiversity in China and an important distribution center of Lauraceae plants globally, Yunnan is a key region for studying the multidimensional diversity patterns of Lauraceae plants and their environmental driving mechanisms. To gain a comprehensive understanding of the multidimensional patterns of biodiversity and to enhance the conservation system in Yunnan, we focused on Lauraceae, which is highly endemic and threatened in Yunnan, to conduct studies on diversity patterns and conservation. 

Methods: By collecting distribution data, we examined the multidimensional diversity patterns of Lauraceae in Yunnan, identified its endemism and diversity hotspots, analyzed the dominant environmental factors shaping these patterns, conducted conservation gap analysis to inform conservation planning. 

Results: (1) The results indicate that there are 21 genera and 225 species of Lauraceae in Yunnan, and the diversity patterns of the different dimensions were mismatched. (2) Water and soil were the dominant factors affecting the patterns. (3) Southeast Yunnan, South Yunnan, Southwest Yunnan, Northwest Yunnan, Northeast Yunnan, as well as the Ailaoshan-Wuliangshan area in the central part are the diversity hotspots and centers of endemism. (4) Current protected areas in Yunnan do not adequately cover Lauraceae conservation hotspots, leaving 63.4% of the critical areas unprotected. Notably, 52.6% of these conservation hotspots were located in narrow areas adjacent to neighboring provinces and countries. 

Conclusion: A multidimensional comprehensive analysis method should be employed to fully understand the distribution patterns of regional biodiversity, and develop scientific and effective conservation planning. The conservation planning of Yunnan needs to focus particularly on edge conservation hotspot grids that are adjacent to neighboring provinces or countries.

Background: The decision to develop a multilateral mechanism, including a global fund, for sharing the benefits derived from the use of digital sequence information on genetic resources (DSI) was adopted during Phrase II of the 15^th meeting of the Conference of Parties to the Convention on Biological Diversity (COP15) at the end of 2022. It was further resolved that the modalities of the mechanism would be finalized at COP16. On November 2^nd, 2024, COP16 adopted a subsequent decision on these modalities, including the establishment of the Cali DSI Fund. This mechanism is considered a potential solution to globally concerning issue of digital bio-piracy. According to its rules and requirements, Parties, non-Parties and other stakeholders are expected to take action to facilitate the implementation of the mechanism at both global and national levels. Its operation is anticipated to have significant implications for DSI governance at these levels. 

Results: The multilateral mechanism applies to DSI that is publicly available in accordance with national legislation and that is not covered by mutually agreed terms (MATs) for access to genetic resources and by other international instruments on access and benefit-sharing. It grants Parties, non-Parties, public databases operators, and users corresponding rights and obligations in aspects such as monetary benefit-sharing, Cali Fund governance, non-monetary benefit-sharing, data governance, and operation management. The launch of Cali Fund signifies that the multilateral mechanism has entered the global implementing stage. Although the multilateral mechanism is not legally binding, its execution is expected to facilitate the transition of open access to DSI derived from genetic resources within national jurisdictions from scientific practice to customary international law. It is also anticipated to promote the advancement of international and domestic legislation related to DSI. As a major producer and user of DSI in scientific research, China has begun to establish a foundational DSI management system. However, DSI generated from domestic genetic resources is currently stored primarily in large international public databases. China’s commercialization of DSI lags behind that of major developed countries, and gaps remain in its DSI management framework, particularly in areas such as property right. 

Recommendations: DSI is considered a fundamental, crucial and strategic production factor for the development of the bioeconomy and for maintaining global competitiveness. There is an urgent need to improve the DSI governance system and to develop a cross-sectoral coordination mechanism based on unified leadership and clear division of responsibilities by integrating DSI into national big data governance. To enhance the legal framework for DSI management, national competent authorities should conduct legislative studies on issues such as property rights attribution, data generation, collection and storage, classification, research and development, data infrastructure construction, cross-border transmission, and data security. Furthermore, a national biodiversity digitalization project should be launched to create a unified plan generation, storage, application and benefit-sharing of DSI. This project should aim to digitalize and informatize key ecosystems, species, genes, and traditional knowledge, while establishing a set of standards for data collection, storage and benefit-sharing. Finally, the potential spillover effects of the multilateral mechanism on other ongoing or future international negotiations should be thoroughly examined.

Aims: This study aims to evaluate the effectiveness of the Baited Remote Underwater Video (BRUV) technique in monitoring nearshore reef fish communities in the Nanpeng Archipelago, China, and to compare its performance with the traditional Underwater Visual Census (UVC) method. The focus is on assessing both techniques in terms of species diversity, composition, and their ability to capture different ecological groups of reef fish. 

Methods: Simultaneous BRUV and UVC surveys were carried out at four coral reef sites. Each BRUV unit, baited with 2 kg of chopped clupeoid fish (Clupea sp), was deployed until the camera was out of battery. UVC surveys were done by divers recording fish species and abundances along transects. Functional diversity was assessed using traits such as body size, trophic level, mobility, gregariousness, water column, and diet. 

Results: BRUV and UVC recorded a similar number of species (59 for BRUV vs. 61 for UVC), but BRUV exhibited a higher functional richness index (0.987 vs. 0.783). Functional trait analysis showed that BRUV was more effective at capturing large-bodied, highly mobile, schooling predators such as Caranx sexfasciatus, whereas UVC detected a higher number of cryptic reef-associated species. Analysis of species accumulation curves indicated an optimal BRUV deployment duration of 50-60 minutes, beyond which additional sampling time yielded diminishing returns in species detection. 

Conclusions: Both BRUV and UVC have distinct strengths and limitations in monitoring reef fish communities. Our findings demonstrate that BRUV and UVC exhibit distinct yet complementary monitoring efficiencies: UVC is more effective for monitoring fish communities in coral-rich reefs, while BRUV performs better in monitoring rocky reef fish diversity. In practical applications, a combination of both methods would provide a more comprehensive understanding of reef fish diversity and ecological function.

Aims: To reveal the cascading effects of plant diversity-grasshoppers-carabids across grassland types and their influences on primary productivity. 

Methods: In Ningxia’s four temperate grassland types, 42 research sites were established to assess plant diversity (quadrat-based species composition and cover degree) and primary productivity. Concurrently, the abundance and richness of grasshoppers were sampled via sweep-netting, the abundance and richness of carabids were sampled via pitfall traps, with subsequent multivariate analysis of all datasets. 

Results: The plant species richness in meadow steppe, typical steppe, desert steppe, and steppe desert was 41, 35, 22, and 16 species, respectively; grasshopper (Orthoptera) species richness was 9, 12, 16, and 6 species, while carabid beetle (Coleoptera: Carabidae) species richness was 8, 12, 12, and 3 species. In the meadow steppe, no significant pairwise correlations were found among plant diversity, grasshoppers, carabids, and primary productivity. In the typical steppe, primary productivity exhibited significant positive correlations with plant diversity, carabid abundance, and carabid species richness. In the desert steppe, primary productivity showed significant positive correlation with plant diversity, but carabid species richness showed significant negative correlation with plant diversity. However, in the steppe desert, not only were grasshopper species richness and abundance significantly positively correlated with plant diversity, but carabid abundance was also significantly positively correlated with grasshopper abundance, while primary productivity demonstrated significant negative correlations with plant diversity, grasshopper abundance, and species richness. 

Conclusion: Therefore, among the four grassland types, the steppe desert exhibited the strongest cascading effects of plant diversity-grasshoppers-carabids, but increased plant diversity promotes locust proliferation and leads to a decline in primary productivity. This study highlights that elevated plant diversity may trigger locust outbreaks and reduce ecosystem productivity. Therefore, when using plant diversity to restore steppe deserts, it is necessary to consider the grassland type and its impact on the interactions between pests and natural enemies and thereby on the primary productivity of the grassland.