Engineering: Journal Articles

Permanent URI for this collection

https://ir.canterbury.ac.nz/handle/10092/2

Browse

Now showing 1 - 20 of 1673

Open Access
A versatile capillaric circuits microfluidic viscometer
(2023) Meffan, Robert Claude; Menges , Julian; Mak , Daniel; Dolamore, Fabian; Fee , Conan; Nock, Volker; Dobson, Renwick C J
A novel capillary-action microfluidic viscometer has been designed that can measure the relative viscosity of a sample compared to a control liquid. Using capillary action circuits, the viscosity of a sample is transformed into an easy-to-read comparative bargraph format while using no external instrumentation. This implementation is made practical through capillaric Field Effect Transistors, and the automated valving they enable. Using poly-ethylene glycol solutions as viscous samples, we demonstrate that the device can transduce the relative viscosity of a sample and control material consistently to within 2% in less than 5 minutes of measurement time. In addition, we show that the device can partially account for temperature through internal referencing. In the future, this device could be applied to medical care, quality assurance, and scientific measurements.
Open Access
TDOA source positioning in the presence of outliers
(Institution of Engineering and Technology (IET), 2019) Ma , Fuhe; Yang, Le; Zhang, Min; Guo , Fu-Cheng
Source localisation using time difference of arrival (TDOA) measurements has drawn considerable attention in the past few decades. The presence of outliers in TDOAs could deteriorate the localisation performance significantly. Under the reasonable assumption that outliers are sparse and they do not dominate in the raw measurement set, a computationally efficient outlier-robust TDOA localisation method is proposed in this study. It integrates the half-quadratic minimisation and reweighted least absolute shrinkage and selection operator to iteratively identify the outliers and find the source location estimate using the TDOA inliers only. Both additive and multiplicative forms of the proposed method are established. Simulation results demonstrate the computational efficiency and effectiveness of the developed algorithms.
Open Access
A hydrological and nutrient load balance for the Lake Clearwater catchment, Canterbury, New Zealand
(2013) Wadworth-Watts , Henry D; Caruso, Brian S; O'Sullivan, Aisling Dominique; Clucas , Rosemary
Nitrogen and phosphorus entering waterways from diffuse agricultural sources is a major problem in New Zealand and internationally. This problem is well documented for lowland areas but little is known about nutrient enrichment from farming in high country areas. The Lake Clearwater catchment, in the Canterbury high country of New Zealand, has a native ecosystem that has adapted to low-nutrient conditions. The Department of Conservation's Arawai Kākāriki Wetland Restoration Programme identifies wetlands in the catchment as one of three key endemic wetland types. Uncertainty regarding diffuse nutrient load from agriculture into the lake and wetland is limiting effective management of this unique catchment. This study investigated the hydrological regimes and nitrogen and phosphorus concentrations and loads in five key surface waterways at ten surface water sites and three groundwater seeps for two years. It aims to improve knowledge of nutrient sources, characteristics and loads from agricultural land use in this 46 km2 high country catchment. Additionally, nutrient load predictions from the Catchment Land Use for Environmental Sustainability (CLUES) model were compared to measured nutrient loads to assess the applicability of the model in high country catchments. The CLUES model was developed, primarily for lowland areas, to predict changes in water quality and nutrient loads from land-use change. The total nitrogen concentrations down- stream of farmland were typically above the Australia and New Zealand Environment and Conservation Council (ANZECC) water quality guideline and the median concentration for upland Canterbury waterways. Specifically, the nitrate concentration (0.19-0.29 g m-3) in farmland subsurface runoff was elevated, compared to streams in the Lake Clearwater catchment with unfarmed catchments, and was estimated to contribute 52% of total nitrogen yield from farmland. The total nitrogen yield (1.96-2.94 kg ha-1 year-1) for farmed land was comparable to minimum values for New Zealand pastoral land use reported in the literature. The total estimated nitrogen export from Lake Clearwater (2518 kg year-1) was 83% greater than the estimated diffuse input from all land in the catchment (1375 kg year-1). This indicated an additional source of nitrogen into the lake and seasonal nitrogen saturation. Total phosphorus yields (0.093-0.123 kg ha-1 year-1) downstream of farmland were well below yields for New Zealand pastoral land 116 use reported in the literature. Total estimated phosphorus export from the lake (58 kg year-1) was 24% less than total estimated diffuse loads into the lake (76 kg year-1). The ratio of total nitrogen to total phosphorus in Lake Clearwater (49:1) indicated that phosphorus is the limiting nutrient and that nitrogen loads into the lake are above natural levels. Total nitrogen loads predicted by the CLUES model in the Lake Clearwater catchment were reasonable, providing land-use area inputs are accurate and nutrient loads exit catchments in surface water. However, CLUES greatly overestimated phosphorus loads from farmed and unfarmed land. © New Zealand Hydrological Society (2013).
Open Access
A taxonomy of common engineering activities and competencies
(Informa UK Limited) Crossin Enda; Richards Jessica I; Dart Sarah; Naswall Katharina
In this paper, we address the lack of a unified approach to understanding engineering practice by developing and presenting a taxonomy of common engineering activities. The taxonomy consists of 86 common engineering activities linked to 17 engineering competencies and the 11 International Engineering Alliance graduate attributes. The list of activities was developed using a six-step process, including multiple systematic literature searches and surveying engineers. The taxonomy provides a critical foundation for better understanding what engineers do, particularly in the Australian and New Zealand context. The taxonomy has potential utility in both engineering practice research and engineering education curriculum reform.
Open Access
Enhanced Degradability, Mechanical Properties, and Flame Retardation of Poly(Lactic Acid) Composite with New Zealand Jade (Pounamu) Particles
(2023) Lin , Lilian; Dang, Quang Anh; Park, Heon Ewan
Plastic pollution has become a global concern, demanding urgent attention and concerted efforts to mitigate its environmental impacts. Biodegradable plastics have emerged as a potential solution, offering the prospect of reduced harm through degradation over time. However, the lower mechanical strength and slower degradation process of biodegradable plastics have hindered their widespread adoption. In this study, we investigate the incorporation of New Zealand (NZ) jade (pounamu) particles into poly(lactic acid) (PLA) to enhance the performance of the resulting composite. We aim to improve mechanical strength, flame retardation, and degradability. The material properties and compatibility with 3D printing technology were examined through a series of characterization techniques, including X-ray diffraction, dispersive X-ray fluorescence spectrometry, scanning electron microscopy, energy-dispersive X-ray spectroscopy, thermogravimetric analysis, 3D printing, compression molding, pycnometry, rheometry, tensile tests, three-point bending, and flammability testing. Our findings demonstrate that the addition of NZ jade particles significantly affects the density, thermal stability, and mechanical properties of the composites. Compounding NZ jade shows two different changes in thermal stability. It reduces flammability suggesting potential flame-retardant properties, and it accelerates the thermal degradation process as observed from the thermogravimetric analysis and the inferred decrease in molecular weight through rheometry. Thus, the presence of jade particles can also have the potential to enhance biodegradation, although further research is needed to assess its impact. The mechanical properties differ between compression molded and 3D-printed samples, with compression-molded composites exhibiting higher strength and stiffness. Increasing jade content in composites further enhances their mechanical performance. Th results of this study contribute to the development of sustainable solutions for plastic pollution, paving the way for innovative applications and a cleaner environment.
Open Access
Simulating deployment of genetic gain in a radiata pine breeding program with genomic selection
(2023) McLean, Duncan; Apiolaza, Luis; Paget , Mark; Klápště , Jaroslav
Genomic selection (GS) is currently being used in the New Zealand radiata pine (Pinus radiata D. Don) breeding program to accelerate genetic gain. GS also has the potential to accelerate the deployment of genetic gain to the production forest through early selection. The increased rate of genetic gain in the breeding cycle will need to be transferred more quickly to realise that gain in the deployment population. GS selections will have lower accuracies than selections based on phenotypic data as currently practised; however, it is unknown how this will affect the genetic gain from GS-based deployment. Moreover, census size and turnover rate need to be optimised to cope with the influx of new marker-based selected material into a commercial orchard. We utilised a stochastic simulation approach to investigate these concepts, comparing three deployment scenarios: half-sib open-pollinated orchards (OP), full-sib control-pollinated orchards (CP) and clonal deployment through somatic embryogenesis. When accounting for time, genomic selection in OP, CP and clonal deployment pathways increased genetic gain by 9.5%, 15.9% and 44.6% respectively compared to phenotypic selection. The optimal orchard scenario would be genomic-selected control-pollination with a low census size (n = 40, males and females combined), low female turnover (5%) and a high male turnover (15–25%). This scheme balances high genetic gain with high seed yield while moderating the rate of inbreeding.
Open Access
A Systematic Review of Individual Tree Crown Detection and Delineation with Convolutional Neural Networks (CNN)
(Springer Science and Business Media LLC, 2023) Zhao , Haotian; morgenroth, justin; Pearse , Grant; Schindler , Jan
Purpose of Review Crown detection and measurement at the individual tree level provide detailed information for accurate forest management. To efficiently acquire such information, approaches to conduct individual tree detection and crown delineation (ITDCD) using remotely sensed data have been proposed. In recent years, deep learning, specifcally convolutional neural networks (CNN), has shown potential in this field. This article provides a systematic review of the studies that used CNN for ITDCD and identifies major trends and research gaps across six perspectives: accuracy assessment methods, data types, platforms and resolutions, forest environments, CNN models, and training strategies and techniques. Recent Findings CNN models were mostly applied to high-resolution red–green–blue (RGB) images. When compared with other state-of-the-art approaches, CNN models showed significant improvements in accuracy. One study reported an increase in detection accuracy of over 11%, while two studies reported increases in F1-score of over 16%. However, model performance varied across different forest environments and data types. Several factors including data scarcity, model selection, and training approaches affected ITDCD results. Summary Future studies could (1) explore data fusion approaches to take advantage of the characteristics of different types of remote sensing data, (2) further improve data efficiency with customised sample approaches and synthetic samples, (3) explore the potential of smaller CNN models and compare their learning efficiency with commonly used models, and (4) evaluate impacts of pre-training and parameter tunings.
Open Access
On the Effectiveness of Using Virtual Reality to View BIM Metadata in Architectural Design Reviews for Healthcare
(2023) Buchanan , Emma; Loporcaro , Giuseppe; Lukosch, Stephan
This article reports on a study that assessed whether Virtual Reality (VR) can be used to display Building Information Modelling (BIM) metadata alongside spatial data in a virtual environment, and by doing so determine if it increases the effectiveness of the design review by improving participants’ understanding of the design. Previous research has illustrated the potential for VR to enhance design reviews, especially the ability to convey spatial information, but there has been limited research into how VR can convey additional BIM metadata. A user study with 17 healthcare professionals assessed participants’ performances and preferences for completing design reviews in VR or using a traditional design review system of PDF drawings and a 3D model. The VR condition had a higher task completion rate, a higher SUS score and generally faster completion times. VR increases the effectiveness of a design review conducted by healthcare professionals.
Open Access
Spatial Information Science in 2023
(Journal of Spatial Information Science, 2023) Adams, Benjamin; Dodge S; Purves R
Open Access
Prediction of Needle Physiological Traits Using UAV Imagery for Breeding Selection of Slash Pine
(American Association for the Advancement of Science (AAAS), 2023) Niu X; Song Z; Xu, Cong; Wu C; Luan Q; Jiang J; Li Y
Leaf nitrogen (N) content and nonstructural carbohydrate (NSC) content are 2 important physiological indicators that reflect the growth state of trees. Rapid and accurate measurement of these 2 traits multitemporally enables dynamic monitoring of tree growth and efficient tree breeding selection. Traditional methods to monitor N and NSC are time-consuming, are mostly used on a small scale, and are nonrepeatable. In this paper, the performance of unmanned aerial vehicle multispectral imaging was evaluated over 11 months of 2021 on the estimation of canopy N and NSC contents from 383 slash pine trees. Four machine learning methods were compared to generate the optimal model for N and NSC prediction. In addition, the temporal scale of heritable variation for N and NSC was evaluated. The results show that the gradient boosting machine model yields the best prediction results on N and NSC, with R2 values of 0.60 and 0.65 on the validation set (20%), respectively. The heritability (h2) of all traits in 11 months ranged from 0 to 0.49, with the highest h2 for N and NSC found in July and March (0.26 and 0.49, respectively). Finally, 5 families with high N and NSC breeding values were selected. To the best of our knowledge, this is the first study to predict N and NSC contents in trees using time-series unmanned aerial vehicle multispectral imaging and estimating the genetic variation of N and NSC along a temporal scale, which provides more reliable information about the overall performance of families in a breeding program.
Open Access
Dismissal informs the priorities of endometriosis patients in New Zealand
(Frontiers Media SA, 2023) Ellis , Katherine; Munro, Deborah; Wood , Rachael
Introduction. Endometriosis is a common condition with average delays to diagnosis in New Zealand of almost 9 years. Methods. In total, 50 endometriosis patients participated in anonymous, asynchronous, online group discussions about their priorities, and their experiences with the development of symptoms, seeking a diagnosis, and receiving appropriate treatment. Results. Higher subsidy of care was the top change endometriosis patients wanted, followed by more research funding. When asked to choose whether research should be focused on improving diagnosis or improving treatment methods, the results were evenly split. Within this cohort, patients highlighted that they did not know the difference between normal menstrual discomfort and pathological endometriotic pain. If, upon seeking help, medical practitioners classified their symptoms as “normal,” these dismissals could instill doubt in patients, which made it more difficult for them to continue to seek a diagnosis and effective treatments. Patients who did not express dismissal had a significantly shorter delay from symptom onset to diagnosis of 4.6 ± 3.4 years vs. 9.0 ± 5.2 years. Conclusion. Doubt is a frequent experience for endometriosis patients in New Zealand, which was reinforced by some medical practitioners who were dismissive of their pain and thus prolonged the patient's delay to diagnosis.
Open Access
Three-dimensional nonlinear dynamics of a thin liquid film on a spinning ellipsoid
(AIP Publishing, 2023) Duruk, Selin; Shepherd, Ross Geoffrey; Boujo, Edouard; Sellier, Mathieu
The present work investigates the three-dimensional flow of a thin liquid film distributed on the outer surface of an ellipsoid, rotating around the vertical axis at constant angular velocity. The lubrication approximation expressing the evolution of the film thickness, originally developed for stationary curved substrates, has been re-derived by including the non-inertial forces associated with the rotation. This comprehensive model, which incorporates the gravitational, centrifugal, and capillary forces, is employed for a parametric investigation via numerical simulations. The results validate and extend the conclusions of our former study covering the axisymmetric case and bring about an advanced understanding by exploring non-axisymmetric effects. The parametric analysis sheds light on the significance of rotation on a non-constant curvature substrate by comparing the thickness profiles with the static case.
Open Access
Indigenous forest classification in New Zealand – A comparison of classifiers and sensors
(Elsevier BV, 2021) Ye N; morgenroth, justin; Xu, Cong; Chen N
Understanding the composition and the changes of New Zealand's woody vegetation communities is important for effective management. However, past national-scale mapped classifications emphasised mature rather than seral vegetation communities and forests were mapped in relative coarse spatial resolution. The integration of Sentinel-2 and PlanetScope imagery provides an opportunity for forest mapping with low cost and high accuracy. This study aims to investigate the feasibility of the integrated image for detailed forest mapping. Free satellite data (Sentinel-2, PlanetScope, fused data) were compared with commercial data (WorldView-2, and WorldView-2 resampled to Sentinel-2 and PlanetScope spatial resolutions) by conducting pixel-based classification with three machine learning classifiers (Support Vector Machine radial basis function kernel, Random Forest, Artificial Neural Network). The combinations of imagery type and classifier were assessed on their potential for mapping nine land cover classes in podocarp forest in New Zealand's central north island, including: conifer, low layer vegetation, broadleaf evergreen, highland softwood, wetland vegetation, water, dead tree, lowland softwood, and low-density vegetation and bare soil. Spectral features (single bands and indices), textural features, and an 8 m resolution digital terrain model (DTM) were used in classifications; the relative importance of these input features was also assessed. In this study, it was found that the overall classification accuracy was dependent on the combination of classifier and imagery, with different combinations resulting in a range of accuracies between 0.669 and 0.956. The best overall accuracy was achieved by integrating Sentinel-2 and PlanetScope imagery (0.956) which was even greater than that of WorldView-2 (0.951). The digital terrain model was the most important feature for all scenarios; Gray-Level Co-Occurrence Matrix-Mean was the most important texture variable for WorldView-2 and integrated images. Original bands, as well as GI, Norm-G, and SR-NIRR, were also crucial for vegetation classification.
Open Access
Evaluation of short-term probabilistic eruption forecasting at Whakaari, New Zealand
(Springer Science and Business Media LLC, 2022) Kempa-Liehr AW; Ardid A; Li A; Orenia S; Singh J; Tyler AJ; Cronin SJ; Dempsey, David
Phreatic explosions at volcanoes are difficult to forecast but can be locally devastating, as illustrated by the deadly 2019 Whakaari (New Zealand) eruption. Quantifying eruption likelihood is essential for risk calculations that underpin volcano access decisions and disaster response. But estimating eruption probabilities is notoriously difficult for sudden onset eruptions. Here, we describe two retrospectively developed models for short-term (48 h) probabilistic forecasting of phreatic eruptions at Whakaari. The models are based on a pseudo-prospective analysis of seven Whakaari eruptions whose precursors were identified by time series feature engineering of continuous seismic data. The first model, an optimized warning system, could anticipate six out of seven eruptions at the cost of 14 warning days each year. While a warning is in effect, the probability of eruption is about 8% in 48 h, which is about 126 times higher than outside the warning. The second model used isotonic calibration to translate the output of the forecast model onto a probability scale. When applied pseudo-prospectively in the 48 h prior to the December 2019 eruption, it indicated an eruption probability up to 400 times higher than the background. Finally, we quantified the accuracy of these seismic data-driven forecasts, alongside an observatory expert elicitation that used multiple data sources. To do this, we used a forecast skill score that was benchmarked against the average rate of eruptions at Whakaari between 2011 and 2019. This exercise highlights the conditions under which the three different forecasting approaches perform well and where potential improvements could be made.
Open Access
Assessing the energy storage potential of electric hot water cylinders with stochastic model-based control
(Informa UK Limited, 2023) Williams B; Bishop, Daniel; Docherty, Paul
As electric hot water cylinders (HWCs) have a large capacity for thermal storage, they are well-suited for Demand Side Management (DSM). This paper compares different methods of HWC temperature control and presents a methodology to assess the amount of thermal storage available in HWCs for demand side management based on use behaviour in different household types. Simple stochastic methods for domestic hot water (DHW) demand prediction were employed to design a smart controller that produced lower rates of unmet DHW demand and higher available storage than setpoint and ripple controllers. The average storage available for DSM from the use of this smart controller is predicted to be between 3.63 and 7.20 kWh per household. These results indicate the use of HWCs for thermal storage is a low-cost viable option for peak-shaving of power system load and could decrease power system greenhouse gas (GHG) emissions in countries such as Aotearoa New Zealand, where GHG-emitting electricity generation is primarily used to meet peak loads.
Open Access
A Luminance-Based Lighting Design Method: A Framework for Lighting Design and Review of Luminance Measures
(MDPI AG, 2023) Bishop, Daniel; Chase, Geoff
Imaging photometers and ray-tracing software packages have made it possible to capture and model high-resolution and accurate luminance maps. However, luminance map measurement is rarely seen in professional practice, despite its ability to evaluate visual parameters accurately and directly, such as contrast, visual, size, and target brightness. Two barriers to the uptake of luminance measurement and associated design measures include (1) lack of knowledge of the range of measures available, and (2) difficulty in assessing whether a luminance-based lighting design method is a sufficient and justifiable replacement for the current illuminance-based practice. This paper reviews current practice and presents alternative luminance design measures and human needs for lighting to construct a framework for designing and comparing lighting design methods. It concludes by presenting a new luminance-based lighting method in the context of this framework to show that it is more accurate and comprehensive than current practice and can be enabled by emerging low-cost and increasingly accessible luminance measurement technologies. The overall outcomes provide the metrics and framework to bring more complete and effective luminance-based lighting design into practice.
Open Access
Mapping fireline intensity and flame height of prescribed gorse wildland fires
(Elsevier BV, 2023) Valencia A; Melnik KO; Kelly RJ; Jerram TC; Wallace H; Aguilar-Arguello S; Katurji M; Pearce HG; Gross S; Strand T
Wildfires currently represent an imminent danger to communities around the world due to their increasing severity and frequency. It is critical to expand current knowledge on wildfire behaviour to improve risk management practices, firefighting operation and engineering design in the Wildland Urban Interface (WUI). This paper reports 2D high resolution measurements of Byram’s fireline intensity generated during a 150 × 200 m2 gorse shrub burn generated from visual and infrared video footage of the prescribed fire, and Light-Detectionand-Ranging (LiDAR) data of the vegetation canopy, both acquired using Unmanned Aerial Vehicles (UAV) technology. The results show that the fireline intensity was highly variable across the plot, reaching localized peak values over 40,000 kW/m and an averaged intensity of 14,300 kW/m approximately. An empirical correlation for estimating the average flame height was derived from the fireline intensity maps and from experimental measurements carried out during the prescribed burn. The correlation was used to produce a 2D detailed map of the average flame height of the prescribed fire, achieving a 11% of error compared to in-fire measurements. The findings from this research suggests that non-linear variabilities of fireline intensity should be carefully considered to adequately describe wildifire behavior of shrubland fires.
Open Access
Hydraulic fracturing fluid migration in the subsurface: A review and expanded modeling results
(American Geophysical Union (AGU), 2015) Birdsell DT; Rajaram H; Viswanathan HS; Dempsey, David
Understanding the transport of hydraulic fracturing (HF) fluid that is injected into the deep subsurface for shale gas extraction is important to ensure that shallow drinking water aquifers are not contaminated. Topographically driven flow, overpressured shale reservoirs, permeable pathways such as faults or leaky wellbores, the increased formation pressure due to HF fluid injection, and the density contrast of the HF fluid to the surrounding brine can encourage upward HF fluid migration. In contrast, the very low shale permeability and capillary imbibition of water into partially saturated shale may sequester much of the HF fluid, and well production will remove HF fluid from the subsurface. We review the literature on important aspects of HF fluid migration. Single-phase flow and transport simulations are performed to quantify how much HF fluid is removed via the wellbore with flowback and produced water, how much reaches overlying aquifers, and how much is permanently sequestered by capillary imbibition, which is treated as a sink term based on a semianalytical, one-dimensional solution for two-phase flow. These simulations include all of the important aspects of HF fluid migration identified in the literature review and are performed in five stages to faithfully represent the typical operation of a hydraulically fractured well. No fracturing fluid reaches the aquifer without a permeable pathway. In the presence of a permeable pathway, 10 times more fracturing fluid reaches the aquifer if well production and capillary imbibition are not included in the model.
Open Access
Collective properties of injection-induced earthquake sequences: 1. Model description and directivity bias
(American Geophysical Union (AGU), 2016) Suckale J; Dempsey, David
Induced seismicity is of increasing concern for oil and gas, geothermal, and carbon sequestration operations, with several M > 5 events triggered in recent years. Modeling plays an important role in understanding the causes of this seismicity and in constraining seismic hazard. Here we study the collective properties of induced earthquake sequences and the physics underpinning them. In this first paper of a two-part series, we focus on the directivity ratio, which quantifies whether fault rupture is dominated by one (unilateral) or two (bilateral) propagating fronts. In a second paper, we focus on the spatiotemporal and magnitude-frequency distributions of induced seismicity. We develop a model that couples a fracture mechanics description of 1-D fault rupture with fractal stress heterogeneity and the evolving pore pressure distribution around an injection well that triggers earthquakes. The extent of fault rupture is calculated from the equations of motion for two tips of an expanding crack centered at the earthquake hypocenter. Under tectonic loading conditions, our model exhibits a preference for unilateral rupture and a normal distribution of hypocenter locations, two features that are consistent with seismological observations. On the other hand, catalogs of induced events when injection occurs directly onto a fault exhibit a bias toward ruptures that propagate toward the injection well. This bias is due to relatively favorable conditions for rupture that exist within the high-pressure plume. The strength of the directivity bias depends on a number of factors including the style of pressure buildup, the proximity of the fault to failure and event magnitude. For injection off a fault that triggers earthquakes, the modeled directivity bias is small and may be too weak for practical detection. For two hypothetical injection scenarios, we estimate the number of earthquake observations required to detect directivity bias.
Open Access
Insights into interconnections between the shallow and deep systems from a natural CO2 reservoir near Springerville, Arizona
(Elsevier BV, 2014) Keating E; Newell D; Pawar R; Dempsey, David
If carbon dioxide (CO2) sequestration into deep geologic reservoirs is to be accepted by the public and environmental regulators, the possibility of upward leakage into shallow groundwater should be acknowledged and those processes well-understood. Studies of natural CO2 reservoirs and their connection (or lack thereof) with the shallow subsurface is one way to explore these issues. A natural reservoir near Springerville, Arizona has leaked CO2 to the surface along a fault zone for thousands of years, creating large travertine deposits. In recent times, the CO2 leak rates have declined significantly yet the shallow aquifer is still highly enriched in CO2. In this study, using water level data and simulations we demonstrate that the fault zone likely provides hydrologic communication between the shallow aquifer and the deeper reservoir. It is reasonable to assume, therefore, that the source of the CO2 in wells completed within the fault zone is the deeper CO2 reservoir. We present water chemistry data to demonstrate the geochemical impact of this CO2 on shallow groundwater quality. Interestingly, arsenic concentrations are elevated, but other trace metals concentrations are not. Arsenic and chloride concentrations co-vary, suggesting perhaps an external source of both elements rather than an in situ release of As due to CO2 attack on shallow aquifer sediments. Observations at this site demonstrate that hydraulic communications between shallow and deep layers and upward CO2 migration does not preclude long-term viability of a substantial CO2 reservoir at depth. We present multi-phase flow simulations to illustrate possible mechanisms trapping the CO2 at depth. Collectively, these analyses show that some degree of upward CO2 leakage may not be necessarily incompatible with the overarching goals of sequestering CO2 and protecting shallow groundwater. © 2014.