Residual Pool Research Articles

YOLO with feature enhancement and its application in intelligent assembly

Object detection is the most important part in intelligent assembly tasks, accurate and fast detection for different targets can complete positioning and assembly tasks more automatically and efficiently. In this paper, a feature enhancement object detection model based on YOLO is proposed. Firstly, the expression ability of feature layer is enhanced through RFP (Recursive Feature Pyramid) structure. The ARSPP (Atrous Residual Spatial Pyramid Pooling) is proposed to have a further enhancement for the feature layers output by the backbone network, it improves the recognition performance for multi-scale targets of model by using different size of dilated convolution and residual connection. Finally, the contiguous pyramid features are fused and enhanced through the attention mechanism, the results are used for the input of next recursive or predictive output. The model proposed in this paper effectively improves the detection accuracy of YOLO, it has 3% MAP improvement in PASCAL VOC dataset. The validity and accuracy of the model are verified in the robot intelligent assembly recognition task.

A-088 Clinically Significant Errors Using the Diazyme Total Bile Acid Assay on the Roche c502 Analyzer: Investigating a Result Error Identifies a Novel Large-Scale Testing Error Caused by Reagent Carryover

Abstract Background Thoroughly investigating single patient result errors may identify systemic, large-scale testing errors. The laboratory observed an event where a revision to a Total Bile Acid (TBA, Diazyme Laboratories, Inc.) patient result following a 22S quality control (QC) failure displayed a larger than expected difference (52 to 5mcmol/L). TBA is primarily used for diagnosis and monitoring of intrahepatic cholestasis of pregnancy with results >10mcmol/L considered elevated. Retrospective review of result revisions following QC failures (3/15/2017-5/30/2023) revealed that 50%(20/40) yielded result differences of >10mcmol/L(>20SD based on assay imprecision). The aim of this study was to apply a systematic approach to i) estimate the rate of TBA errors, ii) ensure accurate result reporting during investigation period, iii) perform root cause analysis (RCA), and iv) determine corrective and preventative action(s). Methods Residual samples from TBA testing (6/24-7/5/2023) were retested(n=158), differences >+/-3.0mcmol/L or 15% were confirmed on an alternate analyzer, and reports were revised. Automated repeat testing of TBA samples >10mcmol/L was operationalized 7/6/2023. Initial and repeat TBA results were compared(n=448) and results differing by >+/-20% were remeasured on an alternate analyzer. RCA was conducted using a fishbone diagram. NaOH reagent probe washes were implemented for TBA and the error rate was re-assessed. Assays run prior to TBA samples with errors (n=15) were identified. To assess reagent carryover a residual serum pool (TBA ∼5mcmol/L) was aliquoted into 5 tubes in a sample rack. The first sample in the rack was programmed to run an assay suspected of causing carryover followed by 4 TBA measurements. Mean±standard deviation(SD) TBA concentrations were calculated. TBA was also measured in the liquid reagent for amylase, lipase, acetylcholinesterase (ACE) and fructosamine. Results Initial TBA retesting yielded no errors when initial TBA was ≤10mcmol/L(n=51). For samples with TBA >10mcmol/L(n=107), 9(8.4%) had differences exceeding criteria with 8/9 being revised to ≤10mcmol/L. Analysis of automated patient repeat data showed a 3.8%(17/448) error rate when initial TBA >10mcmol/L. After NaOH reagent probe washes were implemented, the error rate decreased to 0.7%(3/448). Assays run directly before an erroneously high TBA result included: lactate, fructosamine, soluble transferrin receptor(STFR), lipase, and ACE. A serum TBA pool(mean±SD=5.1±0.4 mcmol/L,n=15) measured 37.6±1.2mcmol/L(n=3) and 6.7±0.7mcmol/L(n=3) after lipase and fructosamine, respectively. No other assays demonstrated carryover. TBA in lipase reagent compartment B and C was 653 and 649mcmol/L, respectively, and 653mcmol/L in fructosamine reagent compartment B. Conclusions A single patient TBA result revision was investigated and led to identification of reagent carryover causing erroneously high TBA results on the Roche c502 analyzer. Automatic, real-time, repeat testing was implemented to prevent reporting incorrect patient results until RCA could identify the cause. Occurrence rate of erroneous high TBA results decreased from 3.8% to 0.7% after implementation of NaOH washes on the instruments. Carryover experiments confirmed that lipase and fructosamine assays cause carryover when run prior to a TBA sample due to measurable TBA in the reagent. The lab is pursuing moving the fructosamine and lipase reagent to the Roche Cobas c701 to prevent the issue.

Non-extractable residues of perfluorooctanoic acid (PFOA) in soil

Perand polyfluoroalkyl substances have gained increased attention due to their persistence, ubiquitous presence in the environment, and toxicity. We hypothesised that the formation of non-extractable residues [NER] occurs in soils and contributes to the overall persistence of these priority pollutants, and that NER formation is controlled by temperature. To test these hypotheses, we used 14C-labelled perfluorooctanoic acid [PFOA] as target compound, added it to two arable soils (Cambisol, Luvisol), and incubated them at 10°C and 20°C in the dark. To support potential co-metabolic decomposition, some samples were additionally fed with glucose to enhance microbial activity. The PFOA residues were then sequentially extracted using 0.01 M CaCl2, followed by accelerated solvent extraction (ASE) with methanol or methanol/acetic acid after 0, 1, 3, 9, 30, 62, and 90 days of incubation. In addition, we monitored the release of 14C into the gas phase as well as [14C]-PFOA-NER after dry combustion and liquid scintillation counting. After 90 days, we found that the [14C]-PFOA content declined in the extraction order of CaCl2 ((bio)available fraction) > ASE (residual fraction) > NER > gas fraction), with most rapid changes occurring in the first 9 days of incubation. NER formation was different in the two soils and reached 5-9% of the applied amount in the Cambisol and Luvisol, respectively. Noteworthy the proportion of 14C-PFOA in the (bio)available fraction remained relatively stable over time at 56-62% of the applied amount, indicating the reversible transfer into this fraction from a bi-exponentially declining residual (ASE) pool. These dissipation patterns were neither influenced by temperature nor by the addition of glucose. We conclude that NER exist for PFOA, but that the majority of PFOA remains in (bio)available form, thus maintaining toxicity and mobility in soil for prolonged periods of time.

MA-VoxelMorph: Multi-scale attention-based VoxelMorph for nonrigid registration of thoracoabdominal CT images

This paper aims to develop a nonrigid registration method of preoperative and intraoperative thoracoabdominal CT images in computer-assisted interventional surgeries for accurate tumor localization and tissue visualization enhancement. However, fine structure registration of complex thoracoabdominal organs and large deformation registration caused by respiratory motion is challenging. To deal with this problem, we propose a 3D multi-scale attention VoxelMorph (MA-VoxelMorph) registration network. To alleviate the large deformation problem, a multi-scale axial attention mechanism is utilized by using a residual dilated pyramid pooling for multi-scale feature extraction, and position-aware axial attention for long-distance dependencies between pixels capture. To further improve the large deformation and fine structure registration results, a multi-scale context channel attention mechanism is employed utilizing content information via adjacent encoding layers. Our method was evaluated on four public lung datasets (DIR-Lab dataset, Creatis dataset, Learn2Reg dataset, OASIS dataset) and a local dataset. Results proved that the proposed method achieved better registration performance than current state-of-the-art methods, especially in handling the registration of large deformations and fine structures. It also proved to be fast in 3D image registration, using about 1.5 s, and faster than most methods. Qualitative and quantitative assessments proved that the proposed MA-VoxelMorph has the potential to realize precise and fast tumor localization in clinical interventional surgeries.

Automatic crack segmentation model based on multi-branch aggregation transformer

Crack detection plays a crucial role in evaluating the safety and durability of civil infrastructure. However, detecting cracks of uneven intensity in complex backgrounds is challenging. To overcome this problem, we propose a dual decoder network (CSMT) based on a multi-branch aggregation Transformer, which uses residual atrous spatial pyramid pooling (RASPP) and Transformer dual decoding branches to extract local and global features of different structures. To enhance global feature extraction, we designed a multi-branch aggregation Transformer (MAT) that adaptively weights the features of two attention heads from spatial and channel dimensions to achieve intra block feature aggregation between dimensions. Meanwhile, to obtain multi-scale semantic information, we constructed a new decoding branch, RASPP, which embeds a squeeze-and-excitation (SE) module and residual structures into standard ASPP. Finally, we propose a feature adaptive fusion module (FAM) to enhance feature fusion between adjacent layers and codec layers. Many experiments on three benchmark datasets have shown that the proposed CSMT segmentation network provides excellent performance in a variety of complex scenarios.

Spatial heterogeneity in nutrient utilization during the end-Devonian ocean anoxic event: a case study of the Western Canada sedimentary basin

The Devonian-Carboniferous (D-C; 359 Ma) boundary is marked by widespread deposition of organic-matter-rich black shales associated with the Hangenberg mass extinction event. The Exshaw Formation spans the D-C boundary in the Western Canada Sedimentary Basin (WCSB) and includes the basal Exshaw Shale deposited under broadly anoxic waters. The sediments at the base of the Exshaw Shale were deposited synchronously during a transgressive event across the WCSB, spanning the geographic variability of the basin. The variable Corg content of the shale was affected by local nutrient upwelling and paleotectonic features impacting water depth and circulation. To characterize the link between paleogeography and nutrient cycling, geographic (N = 20 locations) and stratigraphic (N = 6 locations) trends of δ13Corg and δ15Nbulk were examined throughout the WCSB, representing a range of depositional settings. The δ15Nbulk values range between 0.0 and 6.3‰ and δ13Corg from −29.5 to −26.8‰. Phytoplankton production in focused upwelling zones acquired a relatively 15N-depleted signature through isotopic fractionation during nutrient assimilation, and the residual nutrient pool was 15N-enriched. The advection of surface waters away from the location of upwelling supported additional phytoplankton growth and the deposition of sediments with higher δ15N values. The stratigraphic sections include black laminated and burrowed mudrock sequences that record changes in paleoredox conditions, water depth, and tectonism over time. Up-core from the base of the Exshaw, the Corg content decreases and simultaneously δ15Nbulk increases, suggesting a decrease in eutrophic conditions. Variable δ13Corg and δ15Nbulk trends demonstrate that there is no “type” isotopic profile spanning the D-C boundary in the WCSB.

Protein-Protein Interaction Prediction via Structure-Based Deep Learning.

Protein-protein interactions (PPIs) play an essential role in life activities. Many artificial intelligence algorithms based on protein sequence information have been developed to predict PPIs. However, these models have difficulty dealing with various sequence lengths and suffer from low generalization and prediction accuracy. In this study, we proposed a novel end-to-end deep learning framework, RSPPI, combining residual neural network (ResNet) and spatial pyramid pooling (SPP), to predict PPIs based on the protein sequence physicochemistry properties and spatial structural information. In the RSPPI model, ResNet was employed to extract the structural and physicochemical information from the protein three-dimensional structure and primary sequence; the SPP layer was used to transform feature maps to a single vector and avoid the fixed-length requirement. The RSPPI model possessed excellent cross-species performance and outperformed several state-of-the-art methods based either on protein sequence or gene ontology in most evaluation metrics. The RSPPI model provides a novel strategy to develop an AI PPI prediction algorithm.

Calcium isotope fractionation during weathering of argillaceous carbonates in a humid climate

The continental weathering is a key process that controls calcium (Ca) transportation from the continental crust to the waters. To elucidate the behavior of Ca isotopes during carbonate weathering, the concentrations and δ44/40Ca (relative to NIST SRM 915a) of bulk saprolites, exchangeable, acid-leachable and residual phases of a weathering profile developed on the marine carbonates, Guangdong province, South China, were investigated. Upwards the profile, δ44/40Ca values of the bulk saprolites systematically decrease from 0.77 ± 0.12 ‰ to −0.44 ± 0.12 ‰, suggesting that significant Ca isotope fractionation occurred during chemical weathering. The exchangeable fractions have δ44/40Ca values higher than those of the bulk saprolites with Δ44/40Caexchangeable-saprolite varying from −0.01 ‰ to 0.73 ‰, suggesting that heavy isotopes are preferentially adsorbed onto the clays. The acid-leachable phases display a relatively narrow δ44/40Ca range from 0.52 ‰ to 0.74 ‰ with Ca fractions varying from 7.4 % to 100.3 %, potentially indicating that limited Ca isotopic fractionation occurs during the dissolution of primary carbonates. The residual Ca pool is strongly fractionated with δ44/40Ca ranging from 0.64 ± 0.08 ‰ to −0.98 ± 0.02 ‰, systematically lower than their bulk saprolites, perhaps indicating light Ca isotopes are preferentially incorporated into the clay lattices. Altogether, it seems that the Ca isotopic fractionation directions are opposite between clay structural incorporation and adsorption. Our study provides important insight of Ca behavior and Ca isotopic fractionation during chemical weathering, which is critical to shape Ca isotopic compositions of the upper continental crust and trace the global biogeochemical cycle of Ca.

Dual-branch network based on transformer for texture recognition

Texture recognition is the task of identifying and understanding the patterns and structures present in textures within digital images, constituting a highly challenging research topic. With the rapid advancement of deep learning, CNN-based methods have progressively supplanted traditional approaches, emerging as the predominant techniques for texture recognition. Nevertheless, these methods face limitations in capturing the structural information of textures from global features, attributed to the constraints of CNNs. Structural information stands out as a crucial factor in describing and recognizing real-world texture images. To address this challenge, we propose DBTrans, a dual-branch network based on Transformer architecture for texture recognition. DBTrans leverages the transformer's extraction capability for global features (low-frequency information) to capture the structural nuances of texture images. Additionally, recognizing the distinctive nature of texture recognition tasks, we propose a Transformer variant module tailored for this purpose, named Residual Pooling Transformer (RPT). Finally, we investigate the impact of different branches and modules on network performance and showcase the performance of DBTrans on five public datasets, achieving state-of-the-art accuracy.

National forest carbon harvesting and allocation dataset for the period 2003 to 2018

Abstract. Forest harvesting is one of the anthropogenic activities that most significantly affect the carbon budget of forests. However, the absence of explicit spatial information on harvested carbon poses a huge challenge in assessing forest-harvesting impacts, as well as the forest carbon budget. This study utilized provincial-level statistical data on wood harvest, the tree cover loss (TCL) dataset, and a satellite-based vegetation index to develop a Long-term harvEst and Allocation of Forest Biomass (LEAF) dataset. The aim was to provide the spatial location of forest harvesting with a spatial resolution of 30 m and to quantify the post-harvest carbon dynamics. The validations against the surveyed forest harvesting in 133 cities and counties indicated a good performance of the LEAF dataset in capturing the spatial variation of harvested carbon, with a coefficient of determination (R2) of 0.83 between the identified and surveyed harvested carbon. The linear regression slope was up to 0.99. Averaged from 2003 to 2018, forest harvesting removed 68.3 ± 9.3 Mt C yr−1, of which more than 80 % was from selective logging. Of the harvested carbon, 19.6 ± 4.0 %, 2.1 ± 1.1 %, 35.5 ± 12.6 % 6.2 ± 0.3 %, 17.5 ± 0.9 %, and 19.1 ± 9.8 % entered the fuelwood, paper and paperboard, wood-based panels, solid wooden furniture, structural constructions, and residue pools, respectively. Direct combustion of fuelwood was the primary source of carbon emissions after wood harvest. However, carbon can be stored in wood products for a long time, and by 2100, almost 40 % of the carbon harvested during the study period will still be retained. This dataset is expected to provide a foundation and reference for estimating the forestry and national carbon budgets. The 30 m × 30 m harvested-carbon dataset from forests in China can be downloaded at https://doi.org/10.6084/m9.figshare.23641164.v2 (Wang et al., 2023).

Deep learning-based prediction of thermal residual stress and melt pool characteristics in laser-irradiated carbon steel

Residual stress is a crucial factor that must be managed in laser manufacturing processes. Neglecting it can result in product failure or unsuitability for the intended use. In this article, we present the first surrogate deep learning model based on generative adversarial network that predicts thermal residual stress and melt pool characteristics on the cross-section of laser-irradiated carbon steel. The model was trained using simulation data obtained from the unified momentum equation approach-based numerical model. Our model was designed to predict various aspects of the laser melting process only from three scalar inputs of laser power, beam diameter, and irradiation time at two different points in time (immediately after the laser heating is finished and when the steel is completely cooled down) and successfully predicted the distributions of 12 variables, including residual stress and temperature. Prediction results were not only qualitatively very similar to their ground truth, but also quantitatively accurate with the R2 values ranging from 0.975 to 0.999 and the mean absolute errors between 0.0015 and 0.0126. We demonstrated that complex experiments can be quickly performed using the surrogate model because a single deep learning prediction took only 0.13 s, compared to the 2.5–8 h numerical simulation times.

PIS-Net: Efficient Medical Image Segmentation Network with Multivariate Downsampling for Point-of-Care.

Recently, with more portable diagnostic devices being moved to people anywhere, point-of-care (PoC) imaging has become more convenient and more popular than the traditional "bed imaging". Instant image segmentation, as an important technology of computer vision, is receiving more and more attention in PoC diagnosis. However, the image distortion caused by image preprocessing and the low resolution of medical images extracted by PoC devices are urgent problems that need to be solved. Moreover, more efficient feature representation is necessary in the design of instant image segmentation. In this paper, a new feature representation considering the relationships among local features with minimal parameters and a lower computational complexity is proposed. Since a feature window sliding along a diagonal can capture more pluralistic features, a Diagonal-Axial Multi-Layer Perceptron is designed to obtain the global correlation among local features for a more comprehensive feature representation. Additionally, a new multi-scale feature fusion is proposed to integrate nonlinear features with linear ones to obtain a more precise feature representation. Richer features are figured out. In order to improve the generalization of the models, a dynamic residual spatial pyramid pooling based on various receptive fields is constructed according to different sizes of images, which alleviates the influence of image distortion. The experimental results show that the proposed strategy has better performance on instant image segmentation. Notably, it yields an average improvement of 1.31% in Dice than existing strategies on the BUSI, ISIC2018 and MoNuSeg datasets.

A distinctive new species of Protozantaena Perkins, 1997 from lowland Namaqualand, South Africa (Coleoptera, Hydraenidae).

Protozantaena gigantea sp. nov. is described, based on specimens collected from residual pools in a drying seasonal river in Namaqualand, Northern Cape Province, South Africa. Morphologically, the new species appears to be related to P. labrata Perkins, 1997 from Namibia and P. birdi Bilton, 2022, from the Great Escarpment in the Eastern Cape Province. At up to 2.0 mm in body length, the new species, whilst small, is by far the largest African Protozantaena Perkins, 1997 known to date. The opportunity is also taken to report a new record for P. birdi in the Eastern Cape Drakensberg.

Tyrosine Kinase Inhibitor Discontinuation in Chronic Myeloid Leukemia: Strategies to Optimize Success and New Directions.

The discovery that patients suffering from chronic myeloid leukemia who obtain deep and long-lasting molecular responses upon treatment with tyrosine kinase inhibitors may maintain their disease silent for many years after therapy discontinuation launched the era of treatment-free remission as a key management goal in clinical practice. The purpose of this review on treatment-free remission is to discuss clinical advances, highlight knowledge gaps, and describe areas of research. Patients in treatment-free remission are a minority, and it is believed that some may still retain a reservoir of leukemic stem cells; thus, whether they can be considered as truly cured is uncertain. Strengthening BCR::ABL1 inhibition increases deep molecular responses but is not sufficient to improve treatment-free remission, and we lack biomarkers to identify and specifically target residual cells with aggressive potential. Another level of complexity resides in the intra- and inter-patient clonal heterogeneity of minimal residual disease and characteristics of the bone marrow environment. Finding determinants of deep molecular responses achievement and elucidating varying biological mechanisms enabling either post-tyrosine kinase inhibitor chronic myeloid leukemia control or relapse may help develop innovative and safe therapies. In the light of the increasing prevalence of CML, targeting the residual leukemic stem cell pool is thought to be the key.

Ancient mural segmentation based on multiscale feature fusion and dual attention enhancement

To address the fuzzy segmentation boundaries, missing details, small target losses and low efficiency of traditional segmentation methods in ancient mural image segmentation scenarios, this paper proposes a mural segmentation model based on multiscale feature fusion and a dual attention-augmented segmentation model (MFAM). The model uses the MobileViT network, which integrates a coordinate attention mechanism, as the feature extraction backbone network. It attains global and local expression capabilities through self-attention, class convolution, and coordinate attention and focuses on location information to expand the receptive field and achieve improved feature extraction efficiency. An A_R_ASPP feature enhancement module is proposed for the attention-optimized residual atrous spatial pyramid pooling module. The module uses residual connections to solve the small target loss problem in murals caused by the excessive sampling rate of atrous convolution and uses a feature attention mechanism to adaptively adjust the feature map weight according to the channel importance levels. A dual attention-enhanced feature fusion module is proposed for multiscale decoder feature fusion to improve the mural segmentation effect. This module uses a cross-level aggregation strategy and an attention mechanism to weight the importance of different feature levels to obtain multilevel semantic feature representations. The model improves the mean intersection over union (MIoU) by 3.06% and the MPA by 1.81% on a mural dataset compared with other models. The model is proven to be effective at improving the segmentation details, efficiency and small target segmentation results produced for mural images, and a new method is proposed for segmenting ancient mural images.

Interguild fungal competition in litter and soil inversely modulate microbial necromass accumulation during Loess Plateau forest succession

Microbial interactions determine ecosystem carbon (C) and nutrient cycling, yet it remains unclear how interguild fungal interactions modulate microbial residue contribution to soil C pools (SOC) during forest succession. Here, we present a region-wide investigation of the relative dominance of saprophytic versus symbiotic fungi in litter and soil compartments, exploring their linkages to soil microbial residue pools and potential drivers along a chronosequence of secondary Chinese pine (Pinus tabulaeformis) forests on the Loess Plateau. Despite minor changes in C and nitrogen (N) stocks in the litter or soil layers across successional stages, we found significantly lower soil phosphorus (P) stocks, higher ratios of soil C: N, soil N: P and soil C: P but lower ratios of litter C: N and litter C: P in old (>75 years) than young stands (<30 years). Pine stand development altered the saprotroph: symbiotroph ratios of fungal communities to favor the soil symbiotrophs versus the litter saprotrophs. The dominance of saprotrophs in litter is positively related to microbial necromass contribution to SOC, which is negatively related to the dominance of symbiotrophs in soils. Antagonistic interguild fungal competition in litter and soil layers, in conjunction with increased fungal but decreased bacterial necromass contribution to SOC, jointly contribute to unchanged total necromass contribution to SOC with stand development. The saprotroph: symbiotroph ratios in litter and soil layers are mainly driven by soil P stocks and stand parameters (e.g., stand age and slope), respectively, while substrate stoichiometries primarily regulate microbial necromass accumulation and fungal: bacterial necromass ratios. These results provide novel insights into how microbial interactions at local spatial scales modulate temporal changes in SOC pools, with management implications for mitigating regional land degradation.

Biomedical image segmentation algorithm based on dense atrous convolution.

Biomedical images have complex tissue structures, and there are great differences between images of the same part of different individuals. Although deep learning methods have made some progress in automatic segmentation of biomedical images, the segmentation accuracy is relatively low for biomedical images with significant changes in segmentation targets, and there are also problems of missegmentation and missed segmentation. To address these challenges, we proposed a biomedical image segmentation method based on dense atrous convolution. First, we added a dense atrous convolution module (DAC) between the encoding and decoding paths of the U-Net network. This module was based on the inception structure and atrous convolution design, which can effectively capture multi-scale features of images. Second, we introduced a dense residual pooling module to detect multi-scale features in images by connecting residual pooling blocks of different sizes. Finally, in the decoding part of the network, we adopted an attention mechanism to suppress background interference by enhancing the weight of the target area. These modules work together to improve the accuracy and robustness of biomedical image segmentation. The experimental results showed that compared to mainstream segmentation networks, our segmentation model exhibited stronger segmentation ability when processing biomedical images with multiple-shaped targets. At the same time, this model can significantly reduce the phenomenon of missed segmentation and missegmentation, improve segmentation accuracy, and make the segmentation results closer to the real situation.

Monocular Depth Estimation Based on Residual Pooling and Global-Local Feature Fusion

Monocular Depth Estimation Based on Residual Pooling and Global-Local Feature Fusion

CMPF-UNet: a ConvNeXt multi-scale pyramid fusion U-shaped network for multi-category segmentation of remote sensing images

Most U-shaped convolutional neural network (CNN) methods have the problems of insufficient feature extraction and fail to fully utilize global/multi-scale context information, which makes it difficult to distinguish similar objects and shadow occluded objects in remote sensing images. This article proposes a ConvNeXt multi-scale pyramid fusion U-shaped network (CMPF-UNet). In this work, we first propose a novel backbone network based on ConvNeXt to enhance image feature extraction, and use ConvNeXt bottleneck blocks to reconstruct the decoder. Furthermore, a scale aware pyramid fusion (SAPF) module and Residual Atrous Spatial Pyramid Pooling (RASPP) module are proposed to dynamically fuse the rich multi-scale context information in advanced features. Finally, multiple Global Pyramid Guidance (GPG) modules are embedded in the network, aiming to provide different levels of global context information for the decoder by reconstructing skip-connections. Experiments on the Vaihingen and Potsdam datasets indicate that the proposed CMPF-UNet segmentation achieves more accurate results.

Marsh sediments chronically exposed to nitrogen enrichment contain degraded organic matter that is less vulnerable to decomposition via nitrate reduction

Blue carbon habitats, including salt marshes, can sequester carbon at rates that are an order of magnitude greater than terrestrial forests. This ecosystem service may be under threat from nitrate (NO3−) enrichment, which can shift the microbial community and stimulate decomposition of organic matter. Despite efforts to mitigate nitrogen loading, salt marshes continue to experience chronic NO3− enrichment, however, the long-term consequence of this enrichment on carbon storage remains unclear. To investigate the effect of chronic NO3− exposure on salt marsh organic matter decomposition, we collected sediments from three sites across a range of prior NO3− exposure: a relatively pristine marsh, a marsh enriched to ~70 μmol L−1 NO3− in the flooding seawater for 13 years, and a marsh enriched between 100 and 1000 μmol L−1 for 40 years from wastewater treatment effluent. We collected sediments from 20 to 25 cm depth and determined that sediments from the most chronically enriched site had less bioavailable organic matter and a distinct assemblage of active microbial taxa compared to the other two sites. We also performed a controlled anaerobic decomposition experiment to test whether the legacy of NO3− exposure influenced the functional response to additional NO3−. We found significant changes to microbial community composition resulting from experimental NO3− addition. Experimental NO3− addition also increased microbial respiration in sediments collected from all sites. However, sediments from the most chronically enriched site exhibited the smallest increase, the lowest rates of total NO3− reduction by dissimilatory nitrate reduction to ammonium (DNRA), and the highest DNF:DNRA ratios. Our results suggest that chronic exposure to elevated NO3− may lead to residual pools of organic matter that are less biologically available for decomposition. Thus, it is important to consider the legacy of nutrient exposure when examining the carbon cycle of salt marsh sediments.