Geology and Geophysics Research Services

Geological and geophysical knowledge underpins foundations, tunnels, slopes, groundwater, resource exploration, hazard assessment and environmental studies. Both disciplines infer subsurface conditions from indirect observations, so their results are inherently uncertain and non-unique. Good research makes this explicit by presenting alternatives, quantifying data quality and stating what remains unconstrained. Our support covers data analysis, numerical modelling, geostatistics, inversion workflows and manuscript preparation, not site investigation reports or professional geological certification.

Seismic reflection and refraction, electrical resistivity tomography, ground-penetrating radar, gravity and magnetics each respond to different physical properties. Interpretation depends on processing choices, resolution limits and the quality of acquisition. Inversion converts measurements to property models, but multiple models can fit the data equally well, so regularisation and prior information shape the result.

We help structure processing and inversion workflows, assess resolution and uncertainty and compare results across methods in a transparent way.

Engineering decisions rely on soil and rock properties derived from boreholes, in-situ tests and laboratory testing. Correlations between index and strength parameters are site-specific and have large scatter. Statistical treatment of test data, spatial variability and a clear separation of measured from inferred values are necessary for credible analysis.

We support statistical analysis, probabilistic characterisation and the use of results in numerical models, with the assumptions stated.

Hazard assessment combines mapping, monitoring and models. Susceptibility mapping with machine learning requires careful treatment of spatial autocorrelation and class imbalance, and validation on areas or events not used in training. Seismic hazard calculations involve source models, ground-motion models and logic trees, with epistemic uncertainty that must be conveyed.

We help design validation strategies, interpret performance metrics and write hazard studies with appropriate caveats. Official hazard assessments are outside the scope.

Groundwater studies require hydraulic conductivity, recharge, storage and boundary conditions that are rarely well known. Model calibration should use independent datasets for validation, and parameter uncertainty should be evaluated, for example with ensembles or Bayesian methods.

We support model set-up review, calibration strategy and uncertainty analysis, and the communication of results to non-specialists.

Satellite imagery, digital elevation models and GIS support mapping of lithology, structure, deformation and land cover. InSAR measurements of surface displacement need atmospheric correction and unwrapping checks. Classification results need accuracy assessment with independent reference data.

We help with data processing workflows, accuracy assessment and cartographic presentation of research results.

Kriging, indicator methods and conditional simulation represent spatial variability and uncertainty. Variogram models should be justified and cross-validated, and trends should be handled explicitly. Comparison of interpolation methods on withheld data provides evidence about which performs better for the dataset.

We support analysis and communication of spatial uncertainty.

Reviewers expect maps with scale, coordinates and legends; stratigraphic and structural context; clearly described data acquisition and processing; and uncertainty discussion. Interpretations should be separated from observations in the text, and figures should show raw data as well as interpreted results.

Our manuscript services help structure geoscience papers, check consistency and refine figures, without altering data or interpretations.

Tell us the site or dataset type, the question and the output you need. We will outline an analysis plan and indicate where field validation or professional sign-off must be arranged by you.

Geophysical inversion is ill-posed: different property models produce nearly identical data. Deterministic inversion with regularisation gives one smooth model whose features depend on the regularisation choice. Probabilistic approaches, such as Bayesian sampling or ensemble methods, return distributions of models but at higher computational cost and with assumptions about priors and noise.

A credible paper states the regularisation or prior, tests the inversion on synthetic models with known answers, includes noise and shows resolution or sensitivity maps. We support design of synthetic tests, analysis of data misfit and presentation of uncertainty.

The usefulness of geophysical and geological data depends on acquisition geometry, instrument calibration, positioning accuracy and environmental noise. Recording these details, along with weather, cultural noise sources and processing parameters, allows later users to assess data quality.

We help organise field metadata, perform quality control and document processing in manuscripts and data reports.

Deep learning has been applied to seismic interpretation, facies classification, earthquake detection and lithology prediction. Geoscience data are spatially correlated, imbalanced and often poorly labelled. Training and test sets should be separated by area or time, and performance should be reported per class. Models trained on one basin may not transfer to another.

We help design validation, interpret errors and communicate limitations.

Interpretations vary between experts. Presenting alternative scenarios, ensembles of models and clear statements about confidence helps decision makers and fellow researchers. Colour maps should be perceptually uniform and accessible.

We assist with writing and figure design for communicating uncertainty to technical and non-technical audiences.

Laboratory testing provides parameters for models: porosity, permeability, elastic moduli, strength, thermal and electrical properties. Sample disturbance, stress path and saturation affect results. Several specimens per unit and a description of preparation allow assessment of variability.

We support statistical analysis of laboratory datasets and their translation to model inputs with uncertainty.

The paper should begin with the geological question and why geophysics can address it. The survey design section lists method, line spacing, sampling interval and instrument settings. The processing section gives each step and parameter in order, because small choices alter images. Interpretation presents the geophysical model next to independent evidence such as boreholes or outcrop, with disagreements discussed rather than hidden.

Finally, the paper states what the data can and cannot resolve, for example the minimum feature size and the depth beyond which sensitivity falls. This framework applies equally to seismic, electrical and potential-field studies.

Our manuscript support follows this logic and helps authors assemble the evidence and the figures to match.

Monitoring of ground deformation, microseismicity, groundwater level and temperature shows how the subsurface changes with time. Long records require attention to instrument stability and environmental influences such as barometric pressure and tides. Event detection and catalogue completeness influence statistical conclusions about rates and triggers.

We support time-series analysis, catalogue statistics and the evaluation of relationships between operations and observed responses, with careful distinction between correlation and causation.

Geoscience findings must be understood by engineers, planners and communities. Clear maps, cross-sections with vertical exaggeration stated, and plain-language summaries help. Terminology differs across disciplines, so a short glossary can prevent confusion.

We help with figures and plain-language text for papers, reports and presentations.

Reviewers frequently ask whether the interpretation is unique, whether the data support the claimed resolution, whether processing parameters are given, whether independent evidence confirms the model, and whether the conclusions are limited to the area studied. Addressing these questions in the first draft shortens review. We help authors prepare a point-by-point self-assessment before submission and revise figures and text accordingly.

Another frequent concern is mixing observation and interpretation in the same figure caption. Separating them helps readers judge the evidence.

A short list of additional boreholes, profiles, laboratory tests or monitoring that would reduce the largest uncertainty is valuable in a discussion. It signals awareness of limits and gives readers and funders a clear next step. We help authors identify the uncertainty that matters most and phrase this section constructively.

It separates observation from interpretation, tests inversions on synthetic models, quantifies resolution and uncertainty, uses independent evidence such as boreholes and reports what the data cannot constrain.

We help with processing workflows, statistics and manuscripts, while site investigation and professional sign-off remain with qualified geoscientists.