Forward Modeling Steady-State Free Precession in Surface NMR

Griffiths, Matthew P.; Grombacher, Denys; Liu, Lichao; Vang, Mathias; Larsen, Jakob Juul

doi:10.1109/tgrs.2022.3221624

Cited by 6 publications

(14 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since the primary results of the study is a regional water table mapping, our smooth inversion scheme is not optimal. A layered inversion is currently not implemented with the steady-state modeling methodology since layer boundaries are fixed using the fast-mapping approach (Griffiths et al, 2022) Through the case studies, pulse sequences have been optimized by heuristically inspecting the data and sensitivity functions.…”

Section: Discussionmentioning

confidence: 99%

“…The steady-state pulse trains are defined by the following parameters, pulse duration τ , repetition time, style, offset, Q's, and stack number (Griffiths et al, 2022). The duration of each pulse in the pulse train, τ , is between 10 ms to 100 ms.…”

Section: Data Collectionmentioning

confidence: 99%

“…The style of the pulse can be either "regular", where the polarity of the pulse is the same for the full pulse train, or "alternating" where the polarity changes for each pulse. The offset, δf T , can be used to project the signal away from noisy frequency bins (Grombacher et al, 2022). The number of pulse moments (Q's) defines how dense the current range is sampled.…”

Section: Data Collectionmentioning

confidence: 99%

“…The SNMR data at each site are inverted for water content, T * 2 , and T 2 . we utilize a fast-mapping approach, where data kernels are calculated in advance (Griffiths et al, 2022). The kernels are discretized by a 26-layer model with increasing thickness at depth, to a total depth of 50 m. We used the same starting model for all three field campaigns.…”

Section: Inversionmentioning

confidence: 99%

“…Furthermore, the steady-state approach can also shift the NMR signal away from narrow band noise sources Grombacher et al (2022). This is particularly useful in cases where the Larmor frequency is close to or coincides with a power line harmonic frequency.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Technical note: High density mapping of regional groundwater tables with steady-state surface nuclear magnetic resonance – three Danish case studies

Vang

Grombacher

Griffiths

et al. 2022

Preprint

View full text Add to dashboard Cite

Abstract. Groundwater is an essential part of the water supply worldwide and the demands on this water source can be expected to increase in the future. To satisfy the needs and ensure sustainable use of resources, increasingly detailed knowledge of groundwater systems is necessary. However, it is difficult to directly map groundwater with well-established geophysical methods, as these are sensitive to both lithology and pore fluid. Surface nuclear magnetic resonance (SNMR) is the only method with direct sensitivity to water and it is capable of non-invasively quantifying water content and porosity in the subsurface. Despite these attractive features SNMR has not been widely adopted in hydrological research, the main reason being an often-poor signal-to-noise ratio, which leads to long acquisition time and high uncertainty on results. Recent advances in SNMR acquisition protocols based on a novel steady-state approach has demonstrated the capability of acquiring high quality data much faster than previously possible. In turn, this have enabled high-density groundwater mapping with SNMR. We demonstrate the applicability of the new steady-state scheme in three field campaigns in Denmark where more than 100 SNMR soundings with approximately 30 m depth of investigation were conducted. We show how the SNMR soundings enables us to track water level variations at the regional scale and we demonstrate a high correlation between water levels obtained from SNMR data and water levels measured in boreholes. We also interpret the SNMR results jointly with independent transient electromagnetics (TEM) data, which allows us to identify regions with water bound in small pores. Field practice and SNMR acquisition protocols where optimized during the campaigns, and we now routinely measure high-quality data on eight to ten sites per day with a two-person field crew. Together, the results from the three surveys demonstrate that with steady-state SNMR it is now possible to map regional variations in water levels with high quality data and short acquisition times.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Data Collectionmentioning

confidence: 99%

Section: Data Collectionmentioning

confidence: 99%

Section: Inversionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Technical note: High density mapping of regional groundwater tables with steady-state surface nuclear magnetic resonance – three Danish case studies

Vang

Grombacher

Griffiths

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

A surface NMR forward in a dot product

Griffiths

Grombacher

Kass

et al. 2023

Geophysical Journal International

View full text Add to dashboard Cite

Summary The computation required to simulate surface nuclear magnetic resonance (SNMR) data increases proportionally with the number of sequences and number of pulses in each sequence. This poses a particular challenge to modelling steady-state SNMR, where suites of sequences are acquired, each of which require modelling 10-100s of pulses. To model such data efficiently, we have developed a reformulation of surface NMR forward model, where the geometry of transmit and receive fields are encapsulated into a vector (or set of vectors), which we call B1-Volume-Receive (BVR) curves. Projecting BVR curve(s) along complimentary magnetization solutions for a particular sequence amounts to computing the full SNMR forward model. The formulation has the additional advantage that computations for increased transmitter current amounts to a relative translation between the BVR and magnetization solutions. We generate 1D kernels using BVR curves and standard integration techniques and find the difference is within 2 per cent. Using BVR curves, a typical suite of steady-state kernels can be computed two orders of magnitude faster than previous approaches.

show abstract