Search for: All records

Groundwater discharge is an important mechanism through which fresh water and associated solutes are delivered to the ocean. Permafrost environments have traditionally been considered hydrogeologically inactive, yet with accelerated climate change and permafrost thaw, groundwater flow paths are activating and opening subsurface connections to the coastal zone. While warming has the potential to increase land-sea connectivity, sea-level change has the potential to alter land-sea hydraulic gradients and enhance coastal permafrost thaw, resulting in a complex interplay that will govern future groundwater discharge dynamics along Arctic coastlines. Here, we use a recently developed permafrost hydrological model that simulates variable-density groundwater flow and salinity-dependent freeze-thaw to investigate the impacts of sea-level change and land and ocean warming on the magnitude, spatial distribution, and salinity of coastal groundwater discharge. Results project both an increase and decrease in discharge with climate change depending on the rate of warming and sea-level change. Under high warming and low sea-level rise scenarios, results show up to a 58% increase in coastal groundwater discharge by 2100 due to the formation of a supra-permafrost aquifer that enhances freshwater delivery to the coastal zone. With higher rates of sea-level rise, the increase in discharge due to warming is reduced to 21% as sea-level rise decreased land-sea hydraulic gradients. Under lower warming scenarios for which supra-permafrost groundwater flow was not established, discharge decreased by up to 26% between 1980 and 2100 for high sea-level rise scenarios and increased only 8% under low sea-level rise scenarios. Thus, regions with higher warming rates and lower rates of sea-level change (e.g. northern Nunavut, Canada) will experience a greater increase in discharge than regions with lower warming rates and higher rates of sea-level change. The magnitude, location and salinity of discharge have important implications for ecosystem function, water quality, and carbon dynamics in coastal zones.

 

Background Conventional diagnosis of COVID-19 with reverse transcription polymerase chain reaction (RT-PCR) testing (hereafter, PCR) is associated with prolonged time to diagnosis and significant costs to run the test. The SARS-CoV-2 virus might lead to characteristic patterns in the results of widely available, routine blood tests that could be identified with machine learning methodologies. Machine learning modalities integrating findings from these common laboratory test results might accelerate ruling out COVID-19 in emergency department patients. Objective We sought to develop (ie, train and internally validate with cross-validation techniques) and externally validate a machine learning model to rule out COVID 19 using only routine blood tests among adults in emergency departments. Methods Using clinical data from emergency departments (EDs) from 66 US hospitals before the pandemic (before the end of December 2019) or during the pandemic (March-July 2020), we included patients aged ≥20 years in the study time frame. We excluded those with missing laboratory results. Model training used 2183 PCR-confirmed cases from 43 hospitals during the pandemic; negative controls were 10,000 prepandemic patients from the same hospitals. External validation used 23 hospitals with 1020 PCR-confirmed cases and 171,734 prepandemic negative controls. The main outcome was COVID 19 status predicted using same-day routine laboratory results. Model performance was assessed with area under the receiver operating characteristic (AUROC) curve as well as sensitivity, specificity, and negative predictive value (NPV). Results Of 192,779 patients included in the training, external validation, and sensitivity data sets (median age decile 50 [IQR 30-60] years, 40.5% male [78,249/192,779]), AUROC for training and external validation was 0.91 (95% CI 0.90-0.92). Using a risk score cutoff of 1.0 (out of 100) in the external validation data set, the model achieved sensitivity of 95.9% and specificity of 41.7%; with a cutoff of 2.0, sensitivity was 92.6% and specificity was 59.9%. At the cutoff of 2.0, the NPVs at a prevalence of 1%, 10%, and 20% were 99.9%, 98.6%, and 97%, respectively. Conclusions A machine learning model developed with multicenter clinical data integrating commonly collected ED laboratory data demonstrated high rule-out accuracy for COVID-19 status, and might inform selective use of PCR-based testing. 

Surface effects of sea‐level rise (SLR) in permafrost regions are obvious where increasingly iceless seas erode and inundate coastlines. SLR also drives saltwater intrusion, but subsurface impacts on permafrost‐bound coastlines are unseen and unclear due to limited field data and the absence of models that include salinity‐dependent groundwater flow with solute exclusion and freeze‐thaw dynamics. Here, we develop a numerical model with the aforementioned processes to investigate climate change impacts on coastal permafrost. We find that SLR drives lateral permafrost thaw due to depressed freezing temperatures from saltwater intrusion, whereas warming drives top‐down thaw. Under high SLR and low warming scenarios, thaw driven by SLR exceeds warming‐driven thaw when normalized to the influenced surface area. Results highlight an overlooked feedback mechanism between SLR and permafrost thaw with potential implications for coastal infrastructure, ocean‐aquifer interactions, and carbon mobilization.

 

We present a machine learning (ML) approach for the prediction of galaxies’ dark matter halo masses which achieves an improved performance over conventional methods. We train three ML algorithms (XGBoost, random forests, and neural network) to predict halo masses using a set of synthetic galaxy catalogues that are built by populating dark matter haloes in N-body simulations with galaxies and that match both the clustering and the joint distributions of properties of galaxies in the Sloan Digital Sky Survey (SDSS). We explore the correlation of different galaxy- and group-related properties with halo mass, and extract the set of nine features that contribute the most to the prediction of halo mass. We find that mass predictions from the ML algorithms are more accurate than those from halo abundance matching (HAM) or dynamical mass estimates (DYN). Since the danger of this approach is that our training data might not accurately represent the real Universe, we explore the effect of testing the model on synthetic catalogues built with different assumptions than the ones used in the training phase. We test a variety of models with different ways of populating dark matter haloes, such as adding velocity bias for satellite galaxies. We determine that, though training and testing on different data can lead to systematic errors in predicted masses, the ML approach still yields substantially better masses than either HAM or DYN. Finally, we apply the trained model to a galaxy and group catalogue from the SDSS DR7 and present the resulting halo masses.

 

Major disturbances can temporarily remove factors that otherwise constrain population abundance and distribution. During such windows of relaxed top‐down and/or bottom‐up control, ungulate populations can grow rapidly, eventually leading to resource depletion and density‐dependent expansion into less‐preferred habitats. Although many studies have explored the demographic outcomes and ecological impacts of these processes, fewer have examined the individual‐level mechanisms by which they occur. We investigated these mechanisms in Gorongosa National Park, where the Mozambican Civil War devastated large‐mammal populations between 1977 and 1992. Gorongosa’s recovery has been marked by proliferation of waterbuck (Kobus ellipsiprymnus), an historically marginal 200‐kg antelope species, which is now roughly 20‐fold more abundant than before the war. We show that after years of unrestricted population growth, waterbuck have depleted food availability in their historically preferred floodplain habitat and have increasingly expanded into historically avoided savanna habitat. This expansion was demographically skewed: mixed‐sex groups of prime‐age individuals remained more common in the floodplain, while bachelors, loners, and subadults populated the savanna. By coupling DNA metabarcoding and forage analysis, we show that waterbuck in these two habitats ate radically different diets, which were more digestible and protein‐rich in the floodplain than in savanna; thus, although individuals in both habitats achieved positive net energy balance, energetic performance was higher in the floodplain. Analysis of daily activity patterns from high‐resolution GPS‐telemetry, accelerometry, and animal‐borne video revealed that savanna waterbuck spent less time eating, perhaps to accommodate their tougher, lower‐quality diets. Waterbuck in savanna also had more ectoparasites than those in the floodplain. Thus, plasticity in foraging behavior and diet selection enabled savanna waterbuck to tolerate the costs of density‐dependent spillover, at least in the short term; however, the already poorer energetic performance of these individuals implies that savanna occupancy may become prohibitively costly as heterospecific competitors and predators continue to recover in Gorongosa. Our results suggest that behavior can provide a leading indicator of the onset of density‐dependent limitation and the likelihood of subsequent population decline, but that reliable inference hinges on understanding the mechanistic basis of observed behavioral shifts.