Search for: All records

Abstract The consequences of climate change on boreal ecosystems are evident in declining permafrost extent, amplifying positive climate feedback loops, and altering the timing and intensity of hydrologic events. Thawing permafrost in the discontinuous permafrost zone could affect carbon and nutrient cycling in stream ecosystems. We examined stream chemistry and climate trends over a 20+‐year period across catchments in the Caribou Poker Creeks Research Watershed underlain with varying extents of permafrost (4%–53%). The study aimed to evaluate patterns in dissolved inorganic carbon (DIC,pCO₂), dissolved organic carbon (DOC), nitrogen (Dissolved organic nitrogen, and NO₃⁻), geochemical solutes (Ca²⁺, Mg²⁺, SO₄²⁻), and discharge to determine how altered terrestrial flowpaths and climate change‐related trends in temperature and precipitation have transformed solute transport in high‐latitude watersheds during the ice‐free season. We analyzed long‐term trends in stream chemistry using Thiel‐Sen analysis and a mixed effects model to quantify the influence of abiotic factors on solute concentrations. Results indicate significant declines in DOC (−109.0 to −169.9 μg L⁻¹ yr⁻¹) andpCO₂(−24.1 ppmv yr⁻¹) in higher permafrost extent sub‐catchments. The highest permafrost catchment is experiencing the greatest amount of change, contrasting sharply with opposite to fewer trends in the catchments with lower permafrost extent. Model results indicate the importance of moisture conditions and discharge (p < 0.05), especially for changes in organic solutes. As climate change progresses, the role of these abiotic factors and permafrost thaw will remain important for solute transport dynamics in boreal headwater streams, with consequences for in‐stream communities and downstream solute yields. 

Background: Software Package Registries (SPRs) are an integral part of the software supply chain. These collaborative platforms unite contributors, users, and packages, and they streamline pack- age management. Much engineering work focuses on synthesizing packages from SPRs into a downstream project. Prior work has thoroughly characterized the SPRs associated with traditional soft- ware, such as NPM (JavaScript) and PyPI (Python). Pre-Trained Model (PTM) Registries are an emerging class of SPR of increasing importance, because they support the deep learning supply chain. Aims: A growing body of empirical research has examined PTM reg- istries from various angles, such as vulnerabilities, reuse processes, and evolution. However, no existing research synthesizes them to provide a systematic understanding of the current knowledge. Furthermore, much of the existing research includes unsupported qualitative claims and lacks sufficient quantitative analysis. Our research aims to fill these gaps by providing a thorough knowledge synthesis and use it to inform further quantitative analysis. Methods: To consolidate existing knowledge on PTM reuse, we first conduct a systematic literature review (SLR). We then observe that some of the claims are qualitative and lack quantitative evi- dence. We identify quantifiable metrics assoiated with those claims, and measure in order to substantiate these claims. Results: From our SLR, we identify 12 claims about PTM reuse on the HuggingFace platform, 4 of which lack quantitative validation. We successfully test 3 of these claims through a quantitative analysis, and directly compare one with traditional software. Our findings corroborate qualitative claims with quantitative measurements. Our two most notable findings are: (1) PTMs have a significantly higher turnover rate than traditional software, indicating a dynamic and rapidly evolving reuse environment within the PTM ecosystem; and (2) There is a strong correlation between documentation quality and PTM popularity. Conclusions: Our findings validate several qual- itative research claims with concrete metrics, confirming prior qualitative and case study research. Our measures show further dynamics of PTM reuse, motivating further research infrastructure and new kinds of measurements. 

Background: Software Package Registries (SPRs) are an integral part of the software supply chain. These collaborative platforms unite contributors, users, and packages, and they streamline pack- age management. Much engineering work focuses on synthesizing packages from SPRs into a downstream project. Prior work has thoroughly characterized the SPRs associated with traditional soft- ware, such as NPM (JavaScript) and PyPI (Python). Pre-Trained Model (PTM) Registries are an emerging class of SPR of increasing importance, because they support the deep learning supply chain. Aims: A growing body of empirical research has examined PTM registries from various angles, such as vulnerabilities, reuse processes, and evolution. However, no existing research synthesizes them to provide a systematic understanding of the current knowledge. Furthermore, much of the existing research includes unsupported qualitative claims and lacks sufficient quantitative analysis. Our research aims to fill these gaps by providing a thorough knowledge synthesis and use it to inform further quantitative analysis. Methods: To consolidate existing knowledge on PTM reuse, we first conduct a systematic literature review (SLR). We then observe that some of the claims are qualitative and lack quantitative evidence. We identify quantifiable metrics associated with those claims, and measure in order to substantiate these claims. Results: From our SLR, we identify 12 claims about PTM reuse on the HuggingFace platform, 4 of which lack quantitative validation. We successfully test 3 of these claims through a quantitative analysis, and directly compare one with traditional software. Our findings corroborate qualitative claims with quantitative measurements. Our two most notable findings are: (1) PTMs have a significantly higher turnover rate than traditional software, indicating a dynamic and rapidly evolving reuse environment within the PTM ecosystem; and (2) There is a strong correlation between documentation quality and PTM popularity. Conclusions: Our findings validate several qual- stative research claims with concrete metrics, confirming prior qualitative and case study research. Our measures show further dynamics of PTM reuse, motivating further research infrastructure and new kinds of measurements. 

Abstract Metabolism in stream ecosystems alters the fate of organic carbon (OC) received from surrounding landscapes, but our understanding of in‐stream metabolic processes in boreal ecosystems remains limited. Determining the factors that regulate OC metabolism will help predict how the C balance of boreal streams may respond to future environmental change. In this study, we addressed the question: what controls OC metabolism in boreal headwater streams draining catchments with discontinuous permafrost? We hypothesized that metabolism is collectively regulated by OC reactivity, phosphorus availability, and temperature, with discharge modulating each of these conditions. We tested these hypotheses using a combination of laboratory experiments and whole‐stream ecosystem metabolism measurements throughout the Caribou‐Poker Creeks Research Watershed (CPCRW) in Interior Alaska, USA. In the laboratory experiments, respiration and dissolved OC (DOC) removal were both co‐limited by the supply of reactive C and phosphorus, but temperature and residence time acted as stronger controls of DOC removal. Ecosystem respiration (ER) was largely predicted by discharge and site, with some variance explained by gross primary production (GPP) and temperature. Both ER and GPP varied inversely with watershed permafrost extent, with an inverse relationship between temperature and permafrost extent providing one plausible explanation. Our results provide some of the first evidence of a functional response to permafrost thaw in stream ecosystems and suggest the role of metabolism in landscape C cycling may increase as climate change progresses. 

Abstract Regional and global trade of live animals can contribute to the spread and emergence of novel pathogens, including several important pathogens of amphibians. However, understanding the spread or even frequency of infections in large, complex amphibian trade networks has been difficult, in part because businesses tend to be reluctant to participate in surveillance programs. Thus, we developed a novel approach to surveillance in which anonymous participating businesses were sent surveillance kits through a trusted trade advocacy partner, samples were returned to researchers via anonymous prepaid envelopes, and results were provided via a secure website with access regulated by a unique personal identification number (PIN) created by the business. We tested samples for the amphibian pathogens,Batrachochytrium salamandrivorans(Bsal),Batrachochytrium dendrobatidis(Bd), andRanavirusspp. (Rv), as well as the beneficial microbe,Janthinobacterium lividum(Jliv), using quantitative real‐time polymerase chain reaction (qPCR). Out of 120 businesses invited to complete an anonymous socioeconomic survey, 24 volunteered to participate in pathogen surveillance, of which 14 were sent surveillance kits. Eight of these businesses returned samples consisting of swabs collected from amphibians in 78 terrestrial habitats and water filters from 49 aquatic habitats. Copies of a highly conserved vertebrate gene (EBF3N), quantified using qPCR, were consistently low (<100 copies) in returned samples, but similar to those collected by researchers, indicating comparable sample quality. Three samples (from two facilities) had detectable levels ofBdDNA;Bsal,Rv, andJlivwere not detected. This pilot study provides evidence that information about pathogens in pet trade networks can be acquired by developing partnerships with industry, and business participation might be enhanced by ensuring anonymity and inclusion of a trade advocacy partner. 

We tested the impact of a 15-minute VR training on spatial skills and performance on a geoscience task with a control group. The VR group improved more on the Water Level Task-a measure of understanding of horizontal (B = 0.68, p=0.008). Both groups performed equally on the geology task, except for an orientation rule not well instructed in the VR module (B = -1.33, p=0.0057). In the post-survey, the VR group reported higher ability to link knowledge (X2=4.45, p=0.035) and more interest than in past activities (X2=8.47, p=0.004). This is encouraging, given the brevity of the VR lesson. 

The development and training of deep learning models have become increasingly costly and complex. Consequently, software engineers are adopting pre-trained models (PTMs) for their downstream applications. The dynamics of the PTM supply chain remain largely unexplored, signaling a clear need for structured datasets that document not only the metadata but also the subsequent applications of these models. Without such data, the MSR community cannot comprehensively understand the impact of PTM adoption and reuse.This paper presents the PeaTMOSS dataset, which comprises metadata for 281,638 PTMs and detailed snapshots for all PTMs with over 50 monthly downloads (14,296 PTMs), along with 28,575 open-source software repositories from GitHub that utilize these models. Additionally, the dataset includes 44,337 mappings from 15,129 downstream GitHub repositories to the 2,530 PTMs they use. To enhance the dataset’s comprehensiveness, we developed prompts for a large language model to automatically extract model metadata, including the model’s training datasets, parameters, and evaluation metrics. Our analysis of this dataset provides the first summary statistics for the PTM supply chain, showing the trend of PTM development and common shortcomings of PTM package documentation. Our example application reveals inconsistencies in software licenses across PTMs and their dependent projects. PeaTMOSS lays the foundation for future research, offering rich opportunities to investigate the PTM supply chain. We outline mining opportunities on PTMs, their downstream usage, and cross-cutting questions.Our artifact is available at https://github.com/PurdueDualityLab/PeaTMOSS-Artifact. Our dataset is available at https://transfer.rcac.purdue.edu/file-manager?origin_id=ff978999-16c2-4b50-ac7a-947ffdc3eb1d&origin_path=%2F. 

The development and training of deep learning models have become increasingly costly and complex. Consequently, software engineers are adopting pre-trained models (PTMs) for their downstream applications. The dynamics of the PTM supply chain remain largely unexplored, signaling a clear need for structured datasets that document not only the metadata but also the subsequent applications of these models. Without such data, the MSR community cannot comprehensively understand the impact of PTM adoption and reuse. This paper presents the PeaTMOSS dataset, which comprises metadata for 281,638 PTMs and detailed snapshots for all PTMs with over 50 monthly downloads (14,296 PTMs), along with 28,575 open-source software repositories from GitHub that utilize these models. Additionally, the dataset includes 44,337 mappings from 15,129 downstream GitHub repositories to the 2,530 PTMs they use. To enhance the dataset’s comprehensiveness, we developed prompts for a large language model to automatically extract model metadata, including the model’s training datasets, parameters, and evaluation metrics. Our analysis of this dataset provides the first summary statistics for the PTM supply chain, showing the trend of PTM development and common shortcomings of PTM package documentation. Our example application reveals inconsistencies in software licenses across PTMs and their dependent projects. PeaTMOSS lays the foundation for future research, offering rich opportunities to investigate the PTM supply chain. We outline mining opportunities on PTMs, their downstream usage, and cross-cutting questions. Our artifact is available at https://github.com/PurdueDualityLab/PeaTMOSS-Artifact. Our dataset is available at https://transfer.rcac.purdue.edu/file-manager?origin_id=ff978999-16c2-4b50-ac7a-947ffdc3eb1d&origin_path=%2F.