Adjacent Scales Research Articles

A Comparative study of the mechanical properties of the bio-inspired overlapped scales fabricated by 3D printing

The extreme contrast in stiffness between the hard scales and the surrounding soft issue of animals produces an unusual and attractive mechanism that inspires the designs of biomimetic armor. Despite the growing interest, there are few guidelines for the choice of material, optimal thickness, size, shape and arrangement of the protective scales. Here, we designed six biomimetic armors through simplifying and normalizing the biological armor structures while retaining the main features. We fabricated these structures by 3D printing technology and performed a comparative study of their mechanical properties. We found that the interactions between the adjacent scales significantly enhanced the mechanical performance. We investigated the effect of two important geometrical parameters, the overlapped depth ratio k and slanted angle θ, on the mechanical properties of the composite structures. Moreover, we analyzed the k values for the scales from different body parts of the herbivorous and carnivorous teleost. The optimized designs with different k-values were identified in different parts of the teleost. This study yields new insights into the mechanisms of the overlapped structures and suggests new designs for biomimetic protective systems.

Adaptive learning Unet‐based adversarial network with CNN and transformer for segmentation of hard exudates in diabetes retinopathy

AbstractAccurate segmentation of hard exudates in early non‐proliferative diabetic retinopathy can assist physicians in taking appropriate treatment in a more targeted manner, in order to avoid more serious damage to vision caused by the deterioration of the disease in the later stages. Here, an Adaptive Learning Unet‐based adversarial network with Convolutional neural network and Transformer (CT‐ALUnet) is proposed for automatic segmentation of hard exudates, combining the excellent local modelling ability of Unet with the global attention mechanism of transformer. Firstly, multi‐scale features are extracted through a CNN dual‐branch encoder. Then, the information fusion of features at adjacent scale is realized and the fused features are selected adaptively to maintain the overall consistency of features by attention‐guided multi‐scale fusion blocks (AGMFB). After that, the high‐level encoded features are input to transformer blocks to extract global contexts. Finally, these features are fused layer‐by‐layer to achieve accurate segmentation of hard exudates. In addition, adversarial training is incorporated into the above segmentation model, which improves Dice scores and MIoU scores by 7.5% and 3%, respectively. Experiments demonstrate that CT‐ALUnet shows more reliable segmentation and stronger generalization ability than other SOTA methods, which lays a good foundation for computer‐assisted diagnosis and assessment of efficacy.

ADD-UNet: An Adjacent Dual-Decoder UNet for SAR-to-Optical Translation

Synthetic aperture radar (SAR) imagery has the advantages of all-day and all-weather observation. However, due to the imaging mechanism of microwaves, it is difficult for nonexperts to interpret SAR images. Transferring SAR imagery into optical imagery can better improve the interpretation of SAR data and support the further fusion research of multi-source remote sensing. Methods based on generative adversarial networks (GAN) have been proven to be effective in SAR-to-optical translation tasks. To further improve the translation results of SAR data, we propose a method of an adjacent dual-decoder UNet (ADD-UNet) based on conditional GAN (cGAN) for SAR-to-optical translation. The proposed network architecture adds an adjacent scale of the decoder to the UNet, and the multi-scale feature aggregation of the two decoders improves structures, details, and edge sharpness of generated images while introducing fewer parameters compared with UNet++. In addition, we combine multi-scale structure similarity (MS-SSIM) loss and L1 loss as loss functions with cGAN loss together to help preserve structures and details. The experimental results demonstrate the superiority of our method compared with several state-of-the-art methods.

Abstract A015: Dynamics and lipid interactions of the RAS-RBD-CRD protein complex

Abstract The high-resolution structure of the RAS G-domain complexed to the RBD and CRD domains of RAF provides critical insights into the assembly and conformation of the RAS-RAF signaling complex[1]. The influence and impact of the membrane on both the RAS and RAF proteins is a factor that we are just beginning to understand and appreciate. Molecular simulation is an ideal methodology to further study the complicated relationship between the membrane and associated proteins. Our recent work, using MuMMI[2] (Multiscale Machine-learned Modeling Infrastructure), investigated the different lipid compositions around KRAS4b and the interplay between the protein behavior and these membrane environments[3]. MuMMI uses machine learning to couple adjacent simulation scales and can be effectively scaled across some of the world’s largest high performance computing systems. The MuMMI multi-resolution framework has been expanded to include the RAF RBD-CRD domains. Here we present the overall simulation results from this latest MuMMI campaign. Tens of thousands of coarse-grained molecular dynamics simulations were completed, sampled from a variety of protein/lipid composition configurations. The orientations of the RAS-RBD-CRD complex on the membrane occupies distinct configurational states. Furthermore, the spatial patterns of lipid arrangements around these different protein states are also unique to each state. Interactions of the CRD with the membrane indicate the enrichment of very specific lipids in precise locations act to stabilize the complex. The extent, and size of the lipid ‘fingerprint’ imposed on the membrane by the RAS-RBD-CRD protein complex is significantly larger than observed for just the RAS protein on its own. We do not observe any statistically significant defined protein-protein orientations within the simulation ensemble. These observations indicate that spatial co-localize the RAS-RBD-CRD proteins in the same vicinity may be assisted by specific membrane environments. [1] Tran, Timothy H., et al. "KRAS interaction with RAF1 RAS-binding domain and cysteine-rich domain provides insights into RAS-mediated RAF activation." Nature communications 12.1 (2021): 1-16. [2] Di Natale, Francesco, et al. "A massively parallel infrastructure for adaptive multiscale simulations: modeling RAS initiation pathway for cancer." Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis. 2019. [3] Ingólfsson, Helgi I., et al. "Machine learning–driven multiscale modeling reveals lipid-dependent dynamics of RAS signaling proteins." Proceedings of the National Academy of Sciences 119.1 (2022): e2113297119. Citation Format: Felice C. Lightstone, Timothy S. Carpenter. Dynamics and lipid interactions of the RAS-RBD-CRD protein complex [abstract]. In: Proceedings of the AACR Special Conference: Targeting RAS; 2023 Mar 5-8; Philadelphia, PA. Philadelphia (PA): AACR; Mol Cancer Res 2023;21(5_Suppl):Abstract nr A015.

Machine Learning-Driven Multiscale Modeling: Bridging the Scales with a Next-Generation Simulation Infrastructure.

Interdependence across time and length scales is common in biology, where atomic interactions can impact larger-scale phenomenon. Such dependence is especially true for a well-known cancer signaling pathway, where the membrane-bound RAS protein binds an effector protein called RAF. To capture the driving forces that bring RAS and RAF (represented as two domains, RBD and CRD) together on the plasma membrane, simulations with the ability to calculate atomic detail while having long time and large length- scales are needed. The Multiscale Machine-Learned Modeling Infrastructure (MuMMI) is able to resolve RAS/RAF protein-membrane interactions that identify specific lipid-protein fingerprints that enhance protein orientations viable for effector binding. MuMMI is a fully automated, ensemble-based multiscale approach connecting three resolution scales: (1) the coarsest scale is a continuum model able to simulate milliseconds of time for a 1 μm2 membrane, (2) the middle scale is a coarse-grained (CG) Martini bead model to explore protein-lipid interactions, and (3) the finest scale is an all-atom (AA) model capturing specific interactions between lipids and proteins. MuMMI dynamically couples adjacent scales in a pairwise manner using machine learning (ML). The dynamic coupling allows for better sampling of the refined scale from the adjacent coarse scale (forward) and on-the-fly feedback to improve the fidelity of the coarser scale from the adjacent refined scale (backward). MuMMI operates efficiently at any scale, from a few compute nodes to the largest supercomputers in the world, and is generalizable to simulate different systems. As computing resources continue to increase and multiscale methods continue to advance, fully automated multiscale simulations (like MuMMI) will be commonly used to address complex science questions.

Progressive Dynamic Registration Method for Tile Maps Based on Optimum Multi-Features

Different tile maps may use different coordinate systems, and it is difficult to superimpose maps from different sources. In this regard, we propose a progressive dynamic registration (PDR) method based on optimum features extracted from images of tile map screenshots of a certain scale. Using this method, we can automatically register maps in roughly the same area from different sources without knowing the map project information. The better features among feature points and feature surfaces are selected for image registration based on the richness of the features in the map, and a new matching filter that combines the characteristics of the tile map and the features is proposed. In the progressive registration process during map zooming, the result of the adjacent scales are used as reference values for coarse registration. After experimental verification in different areas, the root mean square error of PDR is below 2.5 in different maps and is better than that of the registration method using feature points only. Moreover, the registration accuracies of remote sensing maps in all areas and vector maps in nonurban areas are better than that of the method based on coordinate system transformation. The calculation results of PDR can register not only the tile maps but also other nontiled vector or remote sensing data.

Miper-MVS: Multi-scale iterative probability estimation with refinement for efficient multi-view stereo

Multi-view stereo reconstruction aims to construct 3D scenes from multiple 2D images. In recent years, learning-based multi-view stereo methods have achieved significant results in depth estimation for multi-view stereo reconstruction. However, the current popular multi-stage processing method cannot solve the low-efficiency problem satisfactorily owing to the use of 3D convolution and still involves significant amounts of calculation. Therefore, to further balance the efficiency and generalization performance, this study proposed a multi-scale iterative probability estimation with refinement, which is a highly efficient method for multi-view stereo reconstruction. It comprises three main modules: 1) a high-precision probability estimator, dilated-LSTM that encodes the pixel probability distribution of depth in the hidden state, 2) an efficient interactive multi-scale update module that fully integrates multi-scale information and improves parallelism by interacting information between adjacent scales, and 3) a Pi-error Refinement module that converts the depth error between views into a grayscale error map and refines the edges of objects in the depth map. Simultaneously, we introduced a large amount of high-frequency information to ensure the accuracy of the refined edges. Among the most efficient methods (e.g., runtime and memory), the proposed method achieved the best generalization on the Tanks & Temples benchmarks. Additionally, the performance of the Miper-MVS was highly competitive in DTU benchmark. Our code is available at https://github.com/zhz120/Miper-MVS.

Entanglement negativity in a nonequilibrium steady state

We study entanglement properties in a nonequilibrium steady state of a free-fermion chain, that emerges after connecting two half-chains prepared at different temperatures. The entanglement negativity and the R\'enyi mutual information between two adjacent intervals scale logarithmically in the system size, with prefactors that we calculate analytically as a function of the bath temperatures. In particular, we show that the negativity and the R\'enyi mutual information with index $\ensuremath{\alpha}=1/2$ are described by different prefactors, and thus the two quantities provide inequivalent information about the state. Furthermore, we show that the logarithmic growth of the negativity during time evolution is also governed by the steady-state prefactor.

Study on Pore Structure Evolution Characteristics of Weakly Cemented Sandstone under Freeze–Thaw Based on NMR

Taking saturated, weakly cemented sandstone as the research object, nuclear magnetic resonance (NMR) tests were performed before and after six freeze–thaw cycles without water replenishment in order to study and reveal the evolution characteristics of the pore structure of weakly cemented sandstone under a freeze–thaw cycle. The evolution of pore structure under repeated freeze–thaw cycles was studied using T2 fractal theory and spectral peak analysis. The results show that the evolution of the pore structure of weakly cemented sandstone can be divided into three stages during the freeze–thaw cycle. In stage 1, the rock skeleton can still significantly restrict frost heave, and the effect of rock pore expansion occurs only on the primary pore scale, primarily in the transformation between adjacent scales. In stage 2, as the restraint effect of the skeleton on frost heave decreases, small-scale secondary pores are gradually produced, pore expansion occurs step by step, and its connectivity is gradually enhanced. In stage 3, as rock pore connectivity improves, the effect of pore internal pressure growth in the freezing process caused by water migration is weakened, making it impossible to break through the skeleton constraint. Thus, it becomes difficult for freezing and thawing to have an obvious expansion effect on the rock pore structure. The strength of the freeze–thaw cycle degradation effect is determined by the effect of the rock skeleton strength under the freeze–thaw cycles and the connectivity of small-scale pores in the rock. The lower the strength of the rock skeleton, the worse the connectivity of pores, and the more obvious the freeze–thaw degradation effect, and vice versa.

MDF-Net: A Multi-Scale Dynamic Fusion Network for Breast Tumor Segmentation of Ultrasound Images.

Breast tumor segmentation of ultrasound images provides valuable information of tumors for early detection and diagnosis. Accurate segmentation is challenging due to low image contrast between areas of interest; speckle noises, and large inter-subject variations in tumor shape and size. This paper proposes a novel Multi-scale Dynamic Fusion Network (MDF-Net) for breast ultrasound tumor segmentation. It employs a two-stage end-to-end architecture with a trunk sub-network for multiscale feature selection and a structurally optimized refinement sub-network for mitigating impairments such as noise and inter-subject variation via better feature exploration and fusion. The trunk network is extended from UNet++ with a simplified skip pathway structure to connect the features between adjacent scales. Moreover, deep supervision at all scales, instead of at the finest scale in UNet++, is proposed to extract more discriminative features and mitigate errors from speckle noise via a hybrid loss function. Unlike previous works, the first stage is linked to a loss function of the second stage so that both the preliminary segmentations and refinement subnetworks can be refined together at training. The refinement sub-network utilizes a structurally optimized MDF mechanism to integrate preliminary segmentation information (capturing general tumor shape and size) at coarse scales and explores inter-subject variation information at finer scales. Experimental results from two public datasets show that the proposed method achieves better Dice and other scores over state-of-the-art methods. Qualitative analysis also indicates that our proposed network is more robust to tumor size/shapes, speckle noise and heavy posterior shadows along tumor boundaries. An optional post-processing step is also proposed to facilitate users in mitigating segmentation artifacts. The efficiency of the proposed network is also illustrated on the "Electron Microscopy neural structures segmentation dataset". It outperforms a state-of-the-art algorithm based on UNet-2022 with simpler settings. This indicates the advantages of our MDF-Nets in other challenging image segmentation tasks with small to medium data sizes.

Vocal imprecision as a universal constraint on the structure of musical scales

Theories of the origin of musical scales from the ancient Greeks to the present day have assumed that the intervals comprising scales are defined by specific mathematical ratios. Such theories are predicated on pre-tunable instruments, and yet the voice is almost certainly the original musical instrument. Therefore, the analysis of vocal scales offers a more naturalistic approach to understanding the origin of musical scales. In the present study, we conducted a large-scale computational analysis of vocal pitch-class properties and their implications for scale structure. We analyzed 418 field recordings of solo, unaccompanied songs from across 10 principal musical-style regions of the world. The results revealed a mean vocal pitch-class imprecision of approximately 1.5 semitones, and comparable results were obtained across all regions. These results suggest that vocal imprecision is universal and is mainly derived from the physiological limitations of the voice. Such vocal imprecision fundamentally constrains the formation of musical scale structure: it provides a lower limit on the spacing between adjacent scale tones and thus an upper limit on the number of scale tones that an octave can contain. We discuss these results in terms of an Interval Spacing model of the evolution of musical scales.

MB-Net: multiscale boundary interaction learning for image manipulation localization

Image editing techniques can modify the content of images indiscriminately, which causes a grave threat to the security of society. Hence, the localization of manipulated images is inevitable. A serious challenge for image manipulation detection is the lack of strategies for perceiving global features and refining edges. In this paper, we present a multiscale boundary interaction learning network for image manipulation localization to solve both problems. This network contains an adjacent-scale mutual module to enrich the global perception domain by interactively learning adjacent scale features. It avoids the tremendous noise interference caused by the direct fusion of all scale features. To effectively suppress semantic content segmentation, the boundary pixel disparity module computes interpixel differences at specific angles to enhance boundary artifact recognition between tampered and real regions. The fusion attention module is proposed to combine scale and edge messages, integrating spatial and channel correlations in a compatible way. Extensive experimental results indicate that our proposed method is significantly superior to current state-of-the-art methods on public standard datasets.

The state space and travelling-wave solutions in two-scale wall-bounded turbulence

The computation of invariant solutions and the visualisation of the associated state space have played a key role in the understanding of transition and the self-sustaining process in wall-bounded shear flows. In this study, an extension of this approach is sought for a turbulent flow which explicitly exhibits multi-scale behaviour. The minimal unit of multi-scale near-wall turbulence, which resolves two adjacent spanwise integral length scales of motion, is considered using a shear stress-driven flow model (Doohan, Willis & Hwang J. Fluid Mech., vol. 913, 2021, A8). The edge state, 26 travelling waves and two periodic orbits are computed, which represent either the large- or small-scale self-sustaining processes. Given that the spanwise length scales are not widely separated here, it could be envisaged that turbulent trajectories visit these solutions in the state space. Considering the intra- and inter-scale dynamics of the flow, numerous phase portraits are examined, but the turbulent state is not found to approach any of these solutions. A detailed analysis reveals that this is due to the lack of scale interaction processes captured by the invariant solutions, including the mean–fluctuation interaction, the energy cascade in the streamwise wavenumber space and the cascade-driven energy production discovered recently. There is a single solution that resembles turbulence much more than the others, which captures two-scale energetics and a scale interaction process involving energy feeding from small to large spanwise scales through the subharmonic sinuous streak instability mode. Based on these observations, it is conjectured that the state space view of turbulent trajectories wandering between solutions would need suitable refinement to model multi-scale turbulence, when each solution does not represent multi-scale processes of turbulence. In particular, invariant solutions that are inherently multi-scale would be required.

Lightweight convolutional neural network for aircraft small target real-time detection in Airport videos in complex scenes

Airport aircraft identification has essential application value in conflict early warning, anti-runway foreign body intrusion, remote command, etc. The scene video images have problems such as small aircraft targets and mutual occlusion due to the extended shooting distance. However, the detection model is generally complex in structure, and it is challenging to meet real-time detection in air traffic control. This paper proposes a real-time detection network of scene video aircraft-RPD (Realtime Planes Detection) to solve this problem. We construct the lightweight convolution backbone network RPDNet4 for feature extraction. We design a new core component CBL module(Conv (Convolution), BN (Batch Normalization), RELU (Rectified Linear Units)) to expand the range of receptive fields in the neural network. We design a lightweight channel adjustment module block by adding separable depth convolution to reduce the model’s structural parameters. The loss function of GIou loss improves the convergence speed of network training. the paper designs the four-scale prediction module and the adjacent scale feature fusion technology to fuse the adjacent features of different abstract levels. Furthermore, a feature pyramid structure with low-level to high-level is constructed to improve the accuracy of airport aircraft’s small target detection. The experimental results show that compared with YOLOv3, Faster-RCNN, and SSD models, the detection accuracy of the RPD model improved by 5.4%, 7.1%, and 23.6%; in terms of model parameters, the RPD model was reduced by 40.5%, 33.7%, and 80.2%; In terms of detection speed, YOLOv3 is 8.4 fps while RPD model reaches 13.6 fps which is 61.9% faster than YOLOv3.

MFNet: Panoptic segmentation network based on multiscale feature weighted fusion and frequency domain attention mechanism

Existing panoptic segmentation networks based on multiscale methods do not distinguish shared features according to different scales, which leads to a loss of feature information and suboptimal results. To solve this problem, the authors improve on UPSNet (a unified panoptic segmentation network) by introducing a multiscale fusion module cascaded between layers into the semantic segmentation head of UPSNet. On the one hand, this enriches the feature information for differently scaled features by cascading the feature information of adjacent scales, and on the other hand, the feature information at each scale can be combined adaptively to achieve a better segmentation effect. Furthermore, a frequency domain attention mechanism is introduced into the backbone network of UPSNet to improve the feature extraction ability of the backbone network by learning more frequency domain feature information. The experimental results of the authors’ improved network on the Common Objects in Context (COCO) and Cityscapes data sets show that compared with UPSNet, the performance is significantly improved: the panoptic quality is improved by 0.9% and 1.6%, respectively.

IRSTFormer: A Hierarchical Vision Transformer for Infrared Small Target Detection

Infrared small target detection occupies an important position in the infrared search and track system. The most common size of infrared images has developed to 640×512. The field-of-view (FOV) also increases significantly. As the result, there is more interference that hinders the detection of small targets in the image. However, the traditional model-driven methods do not have the capability of feature learning, resulting in poor adaptability to various scenes. Owing to the locality of convolution kernels, recent convolutional neural networks (CNN) cannot model the long-range dependency in the image to suppress false alarms. In this paper, we propose a hierarchical vision transformer-based method for infrared small target detection in larger size and FOV images of 640×512. Specifically, we design a hierarchical overlapped small patch transformer (HOSPT), instead of the CNN, to encode multi-scale features from the single-frame image. For the decoder, a top-down feature aggregation module (TFAM) is adopted to fuse features from adjacent scales. Furthermore, after analyzing existing loss functions, a simple yet effective combination is exploited to optimize the network convergence. Compared to other state-of-the-art methods, the normalized intersection-over-union (nIoU) on our IRST640 dataset and public SIRST dataset reaches 0.856 and 0.758. The detailed ablation experiments are conducted to validate the effectiveness and reasonability of each component in the method.

Adaptive weighted multiscale feature fusion for small drone object detection

Drone object detection in low-altitude airspace plays an essential role in many practical applications, such as security and airspace monitoring. Despite the remarkable progress made by many methods, drone object detection still remains challenging due to the complex background and huge differences in scales of drones. To address the above issues, an improved fully convolutional one-stage object detection (FCOS) model based on adaptive weighted feature fusion (AWFF) module is proposed for multiscale drone object detection in complex background. By learning the spatial relevance of feature maps at each scale and improving the scale invariance of features based on the channel attention mechanism, AWFF module could adaptively fuse the features of adjacent scale. In addition, a receptive field enhancement module is designed to reduce the information loss in the feature fusion process. Extensive experiments are conducted to evaluate the effectiveness of the proposed module and method on the constructed low-altitude drone dataset, which concludes that the mean average precision of the AWFF-FCOS is increased by 2.1% compared with the baseline method. And extensive ablation experiments further demonstrate that the proposed AWFF module and REF module could be integrated into the state-of-the-art method to improve the performance of drone object detection.

Nexus of ambient flow and squall line via turbulence in the March 2018 meso-scale convective system over Southeast China

On March 4, 2018, an extremely strong squall line occurs suddenly over Southeast China, incurring devastatingly strong surface winds. In this study, we first present a high-resolution model simulation, then apply the recently developed theory of canonical transfer, which bears a form reminiscent of the Poisson bracket in Hamiltonian mechanics, to investigate the underlying dynamical processes. Using a recently developed functional analysis apparatus, multiscale window transform, the fields are reconstructed onto three scale windows, namely, ambient flow window, squall line-scale window, and turbulence window, which for convenience are denoted as windows 0, 1, 2, respectively. It is found that, among all the canonical transfers of kinetic energy (KE), only that between windows 0 and 2, and that between windows 1 and 2, are significant, and the two correspond remarkably well in space, except for opposite signs. Specifically, at the leading edge, KE is first transferred from the ambient flow over the spectrum all the way to the turbulence scale processes, and then goes to window 1 to form the squall line. At the trailing edge, this process is reversed. While the downscale cascading at the leading edge and upscale transfer at the trailing edge are expected in the classical shock hypothesis of squall lines, the existence of the secondary upscale and downscale transfers, respectively, at the leading and trailing edge, are not seen before. Most importantly, these secondary canonical transfers make the turbulence-scale window function like a hub, bridging the ambient flow and the squall line. To our knowledge, such an energetic scenario, i.e., the nexus of two adjacent scales in a spectrum via a “remote mediator” is not seen before in the literature. This study testifies to the importance of turbulent flows in the maintenance of squall lines, and may yield a clue to a better simulation.

Skin structure, coloration, and habitat utilization in typical and melanistic morphs of the grass snake (Natrix natrix).

Melanism is a polymorphic phenotype caused by the number and density of melanocyte cells producing melanin pigment in the skin and widely observed in snakes. The frequency of this coloration in populations is associated with its opposing fitness consequences and can be closely related to species-specific characteristics such as sex, reproduction, and nutrition, as well as environmental factors such as climate and geography. Although melanism is frequently seen in snakes, the skin structure of melanistic individuals has not been studied in detail. Also, the impact of the black phenotype on habitat use has not yet been clarified in this species. Here, we show a comparison of typical and melanistic morphs of the grass snake Natrix natrix population of Anatolia for the first time in terms of skin structure, habitat, and sex. We found that melanistic individuals, in which partial melanism is more abundant than total melanism, comprise 13% of the population. Melanocyte area of the skin is 1.4 times greater in melanistic compared to the typical individuals. The epidermis is thicker in typical morphs by 7.7%. Hinge regions between adjacent scales do not bare melanocytes in both morphs. As for habitat utilization, we revealed that melanistic individuals of the Işıklı population tend to occur closer to water bodies than typical ones. Our data provide a new perspective on poorly known aspects of color polymorphism and habitat use of widely distributed, semi-aquatic Natrix natrix.

High flower richness and abundance decrease pollen transfer on individual plants in road verges but increase it in adjacent fields in intensively managed agroecosystems

Road verges can facilitate the movement of pollinators and pollen in intensively managed agroecosystems, reducing the negative impacts of fragmentation (i.e., isolation of wild pollinator communities) and enhancing the pollination service provided to wild plants. To assess the role of road verges in sustaining the pollination service to wild plants, we studied gardens consisting of two plant species arranged within sixteen road verges and their adjacent fields during two consecutive seasons in the Argentine Pampas. We assessed the influence of attributes of road verges on pollination service at different scales: road verges width, species richness of flowering plant, and flower abundance (patch scale), crops and successional pastures (adjacent land use scale), and landscape heterogeneity (landscape scale). We measured the pollination service provided to two species of potted plants (Hirschfeldia incana and Lobularia maritima) in three different ways: visitation rate, conspecific and heterospecific pollen transfer to stigmas, and distance of pollen transfer from a fixed source. The number of conspecific and heterospecific pollen grains deposited on stigmas decreased with an increased abundance of floral units in associated road verges, which is caused by a dilution effect on individual plant pollination. Also, the pollen transfer into adjacent fields decreased with an increased abundance of floral units in the nearby road verges, which is caused by a decrease of pollen spillover into adjacent crop fields. The visitation rate within adjacent fields increased with higher flowering plant species richness in associated road verges. Also, the number of conspecific pollen grains in the road verges adjacent to successional pastures increased with higher flowering plant species richness in road verges, which results in a concentration effect of pollinator visits and pollen transfer into road verges. We conclude that managing flower abundance and increasing flowering plant species richness in road verges are crucial practices for maintaining the pollination service of entomophilous plants in intensively managed agroecosystems.