DDDavid Dempsey

David Dempsey

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


David Dempsey has a total of 138 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    A Physics‐Based, Lumped‐Parameter Model for Attributing Natural and Anthropogenic CO 2 Emissions From Producing Geothermal Systems

    by David Dempsey

    Although geothermal energy is a low‐carbon resource, some powerplants can coproduce and release , which is then subject to emissions monitoring and sometimes financial penalty. Accurate attribution of these emissions is difficult because exploitation of the reservoir alters hydrological flows that in turn modify surface fluxes. Hence, emissions metered through the powerplant do not represent the true anthropogenic impact. This study develops a lumped‐parameter model of a geothermal reservoir that tracks pressure evolution, dissolved and emissions through the powerplant and the Earth's surface. Closed‐form analytical solutions are obtained for reservoir degassing, outflow reversal, delayed reinjection returns and mineralization reactions, and these are calibrated against emissions data from the Rotokawa and Ngā Tamariki geothermal fields in New Zealand. The interplay of characteristic timescales for pressure equilibration, loss, reservoir residence, breakthrough delay and mineralization are shown to control whether a field's emissions decline, stabilize or rebound over decades. Calibrated models indicate that plant‐metered emissions at Rotokawa overcount the true anthropogenic impact by 17–45 kt , because production suppresses surface emissions that are not credited to the operator. storage in geothermal reservoirs is shown to be fundamentally different from conventional carbon capture and storage, as it is a flow‐through system in which dissolved only resides temporarily. Long‐term sequestration requires either mineralization or reinjection at sufficient distance to delay returns. Field‐specific models and repeat flux surveys are needed to constrain these effects and support emissions accounting based on true anthropogenic impact.

  • Open Access

    Socio-economic value of data-driven eruption forecasts to balance false alarms against catastrophic loss

    by Alberto Ardid, David Dempsey, Garry McDonald, Shane Cronin

    The adoption of new forecasting methods for volcanic eruptions typically emphasizes predictive accuracy, often overlooking their potential to reduce overall life and economic losses. We introduce a Potential Economic Value (PEV) framework that evaluates forecast utility by balancing precautionary actions against avoidable catastrophic losses (such as mass casualties). Using machine-learning forecasts from continuous seismic data at five volcanoes, we show that non-forecasted eruptions (missed) have disproportionate consequences, compared to false alarms, which generate recurring and manageable disruption. Retrospective analyses of the 2019 Whakaari (New Zealand) and 2014 Ontake (Japan) eruptions, along with three additional volcanoes, indicate that losses could have been reduced by 30–90%, despite generating numerous false alarms. Across all case studies, effective forecasting prioritizes reducing missed eruptions over maximizing accuracy. This supports the use of more precautionary warning thresholds, provided that the resulting increase in alert frequency is managed in ways that maintain public compliance and trust.


  • Open Access

    A Machine‐Learning Classification for Torlesse Composite Terrane Petrofacies and its Application to Sediment Provenance

    by Matthew O. Parker, David E. Dempsey, Matthew W. Sagar, Kari N. Bassett, Alexander R. L. Nichols, Greg H. Browne

    The Torlesse Composite Terrane (TCT) forms many of the mountain ranges in Aotearoa New Zealand and has provided enormous quantities of coarse‐grained sediment to Te Riu‐a‐Maui/Zealandia's basins since the mid‐Cretaceous. Tracing the provenance of these sediments to certain regions of the TCT can indirectly reconstruct exhumation patterns associated with the development of the Australian–Pacific plate boundary. Supplemented by new analyses, we are now able to include one of the new constituent terranes of the TCT (the Kaweka Terrane), which had not been recognised when the previous geochemical classification was developed. We train four different machine learning algorithms using samples from Te Waipounamu/the South Island to classify TCT‐derived sandstone conglomerate clasts into four different petrofacies/terranes of the TCT using log‐transformed trace element ratios. Balanced accuracies for the four algorithms used are 77% (linear support vector machine), 69% (k‐nearest neighbours), 75% (random forest), and 77% (neural network). This rises to 80% when an equal voting system is used. These accuracies are similar to existing classifications for the TCT but include the Kaweka Terrane as a classification option. We then show an example of how this updated classification can be applied to late Cenozoic TCT‐derived conglomerate clasts.


  • Open Access

    Physics-based Accounting for Natural and Anthropogenic CO2 Emissions from Producing Geothermal Systems

    by David Dempsey

    Geothermal systems are frequently used as a source of low-emissions energy. However, reservoirs with high amounts of dissolved CO2 can produce substantial power plant emissions, exceeding 100 kt/yr in some cases. Accurately accounting for the net anthropogenic emissions at geothermal systems is difficult because these systems also naturally flux CO2, sometimes at a magnitude similar to plant emissions, and this natural flux may increase or decrease due to plant operations. Furthermore, recent efforts in Iceland and New Zealand to capture and reinject geothermal emissions can further alter CO2 fluxes and raise the amount of CO2 stored in the subsurface.In the absence of direct monitoring data, mathematical models can be used to estimate geothermal CO2 emissions. Here, I describe a lumped parameter model of CO2 flux through a liquid geothermal reservoir. The model parameterises CO2 influx due to magmatic degassing, CO2 loss from vertical migration through caprock or lateral outflow, degassing due to pressure or boiling driven solubility changes, CO2 extraction and reinjection through wells, and permanent storage through mineralisation reactions. Under constant mass extraction and suitable simplifying assumptions, the model can be solved exactly yielding exponential approximations of emissions rates (natural and plant), and reservoir CO2 content.Calibration of this model to ten years of plant emissions and pressure decline data at Rotokawa and Ngā Tamariki geothermal fields (New Zealand), suggest that degassing trends are largely driven by CO2 dilution of the liquid reservoir. Furthermore, both depressurisation and dilution substantially lower the natural CO2 outflow, an effect not presently accounted for in greenhouse gas inventories. This means that measured plant emissions may exceed the true anthropogenic impact on geothermal emissions by up to a factor of three, which has substantial financial implications for geothermal plant operators.

  • Open Access

    Geochemical Modelling for Carbon Dioxide Removal Applications

    by Ting Hu, David Dempsey, Zhencheng Zhao, Jie Dong, Zhenhua Rui

    Water-rock and aqueous reactions affect CO2 geological storage in several ways, including through carbonate mineralization, dissolution and reprecipitation, silicate dissolution, and acid-base buffering. However, complex water chemistry compositions and multiple rock mineral types make the quantitative characterization of these reactions difficult. Here, predictive models of geochemical reactions were developed taking place within strong to moderately reactive storage formations where pH-sensitive silicate dissolution and carbonate precipitation dominate. To do this, the TOUGHREACT thermal-hydrological-chemical multiphysics subsurface reactive transport simulator was used to develop well-calibrated models based on field monitoring data.This study first benchmarked a model against a single-well CO2 push-pull field test conducted in a pH 11.02, shallow peridotite formation in Oman, described in Matter et al. (2025). The simulation included the 13.7-hour carbonated water injection, the subsequent 45-day shut-in, and then 11.2 days of pumping. Calibration of the porosity, permeability and formation mineral assemblage primarily occurs against recorded ion concentrations during the pumping period. The model suggests calcite precipitation dominated at the margins of the 6.4 m radius mineralization zone, with dolomite at intermediate distances and magnesite in the immediate vicinity of the well. Magnesite precipitation is associated with lower pH conditions near the well where there is sufficient available Mg2+ dissolved from the host rock, whereas dolomite and calcite are deposited at higher pH and sufficient available Ca2+. During the storage period, our model underpredicts mineralization (52%) compared to that inferred by Matter et al. (88%), likely due to underprediction of dolomite or magnesite. The precipitated carbonates remain stable upon re-equilibration of the groundwater.The model was then applied to a hypothetical doublet storage operation in a CO2-rich hydrothermal system at Ohaaki, New Zealand. The goal was to predict CO2 phase evolution subject to long-term geochemical reactions as well as boiling of the fluid phase. Simulations show that ions primarily controlled by a single mineral (Ca2+, Na+, K+, and Fe2+) all reach their peak concentrations within five years, whereas subsequent geochemical evolution is influenced by the dynamic equilibrium of aqueous complexes, such as CaSO4(aq), NaCl, MgHCO3+, and FeCl+. Driven by the injection of aqueous solutions with high carbonic acid concentrations, the mineral volume fraction at the injection well changes at a rate 2–11 times greater than that observed in the rest of the simulation domain. Under high-temperature and low-pressure conditions of the production well, a CO2 boiling zone forms in the reservoir, with the peak gas saturation of CO2 exsolved from the liquid phase reaching 7.6 wt% over the simulation period. This research shows that the geochemical reaction simulation holds significant scientific value for CO2 storage applications in strong to moderately reactive storage formations.

  • Open Access

    The Research and Impact of the Works by Don Nield

    by Sid Becker, David Dempsey, Dan Zhao

    This is a commemoration of the work of Don Nield. His theoretical and applied mathematical models have had a tremendous impact in many fields including convection heat and mass transfer, stability analysis, nanofluid enhanced convection, and flows in porous media. Here the major contributions of Don Nield’s Scopus-indexed outputs are reviewed. His research outputs and related impact are presented chronologically. The citation counts are presented with respect to each work published in that year. The work and impact of Don Nield as single author and in his collaborations are presented.

  • Open Access

    Quantifying Interannual Variations in Groundwater Fluxes Within a Braidplain Aquifer Using Active‐Distributed Temperature Sensing

    by A. J. Sai Louie, L. K. Morgan, D. Dempsey, S. Wilson, E. W. Banks

    Braided rivers are a significant contributor to groundwater flow in braidplain aquifer (BPA) systems, yet the spatiotemporal variations in flow exchange remain poorly understood. This study presents an extensive active‐distributed temperature sensing data set collected over a 3‐year period from two 100‐m long fiber optic cables installed up to ∼5 m depth below a braided river channel in New Zealand to quantify specific discharge (i.e., parafluvial flow) within the BPA at high spatial resolution. Despite river flow fluctuating by two orders of magnitude, specific discharge within the BPA showed moderate variation, with values ranging from 3.5 to 5.9 m d−1 on HDD1 (20.4% variance relative to the median) and from 2.4 to 4.5 m d−1 on HDD2 (13.3% variance relative to the median) over the study period. Geomorphology influences hydraulic conductivity, thereby affecting the hydraulic gradient. Following a flood event, specific discharge increased beneath the river margins over time, with the largest increase corresponding with areas of greatest geomorphic change, and changes to the subsurface flow dynamics. Surveys conducted after the flood event, where river flow peaked at 225 m3 s−1 (compared to a median of 2.51 m3 s−1), showed a marked increase in specific discharge. These subsurface changes occurred gradually, with no obvious seasonal influence, compared to rapid geomorphic surface alterations. This translates to variable specific discharge within the BPA in both space and time caused by changing bedform and connectivity to preferential flowpaths. Geomorphic changes following a flood event had the largest impact on specific discharge.


  • Open Access

    Forecasting Eruptions at Steamboat Geyser: Time Scales, Differentiability, and Detectability of Seismic Precursors Through Data‐Driven Methods

    by Alberto Ardid, Anna Barth, David Dempsey, Michael Manga, Shane J. Cronin

    Geyser eruptions provide a test bed for using geophysical data to forecast eruptions and to understand heat and mass transport in hydrothermal systems. We used time series analyses of seismic data at Steamboat Geyser, Yellowstone National Park, to identify short‐term precursors that are recurrent, detectable in real time, distinctly identifiable, as well as being rare during non‐eruptive periods. We analyzed seismic data from March to December 2018 to identify patterns that occurred before 31 eruptions. Four seismic amplitude measures and 700 time‐series features were computed from the seismic data. A template matching analysis identified an optimal 18‐hr window for detecting precursors. We applied a random forest to classify pre‐eruptive and non‐eruptive data for out‐of‐sample eruptions (eruptions that were not included in the model's training data), showing ability to distinguish between the two. This model performed better than a simpler amplitude‐based approach. Seismic features with the most predictive power include autocorrelations, longest strike above the mean, and change quantiles, particularly within the 4.5–16 Hz frequency range. We applied isotonic regression, a method that converts raw model outputs into calibrated probabilities, to improve the interpretability of eruption forecasting outputs. The likelihood of an eruption reaches 12.6% within 18 hr prior to the event, representing a marginal increase over the static 8% probability derived solely from eruption intervals. Unlike the interval‐based approach, our model does not rely on the time since the last eruption, instead using real‐time seismic features to detect precursory signals. Our study advances Machine Learning methodologies in eruption forecasting by integrating calibrated probability estimation through isotonic regression, which has advantages over traditional approaches for geysers with highly irregular eruption intervals.


  • Open Access

    Ergodic seismic precursors and transfer learning for short term eruption forecasting at data scarce volcanoes

    by Alberto Ardid, David Dempsey, Corentin Caudron, Shane Cronin, Ben Kennedy, Társilo Girona, Diana Roman, Craig Miller, Sally Potter, Oliver D. Lamb, et al.

    Seismic data recorded before volcanic eruptions provides important clues for forecasting. However, limited monitoring histories and infrequent eruptions restrict the data available for training forecasting models. We propose a transfer machine learning approach that identifies eruption precursors—signals that consistently change before eruptions—across multiple volcanoes. Using seismic data from 41 eruptions at 24 volcanoes over 73 years, our approach forecasts eruptions at unobserved (out-of-sample) volcanoes. Tested without data from the target volcano, the model demonstrated accuracy comparable to direct training on the target and exceeded benchmarks based on seismic amplitude. These results indicate that eruption precursors exhibit ergodicity, sharing common patterns that allow observations from one group of volcanoes to approximate the behavior of others. This approach addresses data limitations at individual sites and provides a useful tool to support monitoring efforts at volcano observatories, improving the ability to forecast eruptions and mitigate volcanic risks.