Are Elemental Salinity Proxies Worth Their Salt?

Graphical abstract

Salinity is a fundamental variable of watermass chemistry, often varying strongly in coastal, estuarine, lagoonal, restricted-marine basinal, and non-freshwater lacustrine systems (Figure 1). Salinity variation commonly controls other watermass properties (e.g., redox, temperature, pH, and chemical composition) as well as nutrient levels (and thus bioproductivity). Despite its importance, paleoenvironmental research has long focused mainly on reconstruction of redox conditions with more limited attention given to other variables. In many studies, salinity has been completely ignored, sometimes upon the assumption that a given paleodepositional system was fully marine, but mostly because no direct geochemical proxy was available to assess paleosalinity conditions. Although various tools are available to constrain paleo-watermass salinity, they tend to be highly specialized in their application, e.g., requiring the presence of specific biotic or mineral components or involving complexities such as independent assessment of paleotemperature (as for salinity analyses based on O isotopes). What has been lacking to date are geochemical proxies that can be applied to bulk sediment samples for interpretation of paleosalinity conditions.

Over the past few years, a quiet revolution in paleoenvironmental research has taken place through demonstration of the utility of elemental proxies for salinity analysis, including boron/gallium (B/Ga), strontium/barium (Sr/Ba), and sulfur/total organic carbon (S/TOC). The underlying principles for use of B/Ga and Sr/Ba as salinity proxies are: (1) B and Sr are present in low concentrations in freshwater and high concentrations in seawater, (2) both are readily taken up by clay minerals in the sediment in approximate proportion to their aqueous concentrations, and (3) Ga and Ba serve as normalizing agents for detrital content, being in the same elemental groups and exhibiting similar ionic radii and crustal concentrations as the elements being normalized (i.e., B and Sr) (Figure 2). The S/TOC proxy is based on the presence of aqueous sulfate in marine and brackish facies versus its near-absence in freshwater facies; in the former, sulfate is reduced to H₂S through linked oxidation of organic matter and then fixed in the sediment as iron sulfides (Figure 2). Although these proxies were originally proposed in the 1960s, they failed to gain traction at that time but have since experienced a renaissance through the work of Profs. Wei Wei and Thomas Algeo, of the China University of Geosciences-Wuhan, and their research colleagues. Wei and Algeo published a key paper in Geochimica et Cosmochimica Acta in 2020 that calibrated these proxies in modern fine-grained siliciclastic facies across a salinity spectrum from freshwater to fully marine conditions (Figure 3).

Since the foundational work of Wei and Algeo (2020), these proxies have been widely applied in paleoenvironmental research. Subsequent work has shown that B/Ga is generally the most robust of these proxies in terms of both fidelity to depositional salinity and resistance to post-depositional alteration. The potential applications of these paleosalinity proxies are diverse and expanding. Salinity is an important control on nearly all other watermass properties, so depositional systems characterized by significant salinity variation typically exhibit strong covariation of salinity with redox, productivity, and hydrographic restriction proxies, among other environmental parameters. Salinity-redox covariation is a significant feature of many paleodepositional systems, including the Cryogenian Nanhua Basin of South China (Cheng et al., 2021), the Late Devonian Appalachian Basin (Gilleaudeau et al., 2023), and the Early Jurassic Cleveland Basin of the UK (Remírez and Algeo, 2020). Salinity was shown to be a major control on phytoplankton community composition in the Late Devonian Illinois Basin, which is marked by a shift in relative dominance from red algae and bacteria to green algae in conjunction with watermass freshening (Song et al., 2021). In marginal marine basins, salinity variation has been linked to sea-level elevation and seawater exchange rates (Remírez and Algeo, 2020; Cheng et al., 2021; Remírez et al., 2025a). The utility of these proxies has now been tested in units ranging in age from the Holocene (Remírez et al., 2024) to the Paleoproterozoic (Liu et al., 2025), demonstrating their wide temporal applicability with potential for assessment of salinity variation even in stratigraphic units of great antiquity.

Paleosalinity reconstruction is of greatest value in coastal depositional systems in which watermass salinity is likely to have varied as a consequence of sea-level change, shifts in shorelines or river mouths, or changes in hydrology linked to climate or other controls. Marine facies generally yield high and relatively invariant salinity proxy values, as would be expected of depositional systems without significant salinity variation (Wei and Algeo, 2020; Wei et al., 2025a). In contrast, substantial variation in salinity proxies is encountered in paleodepositional systems representing coastal settings (Remírez et al., 2025b), marginal marine basins (Remírez and Algeo, 2020; Remírez et al., 2025a), deep epicratonic basins (Cheng et al., 2021; Gilleaudeau et al., 2023), and shallow epeiric seas (Song et al., 2021; Algeo et al., 2025a) (Figure 1). Elemental salinity proxies have occasionally yielded complete surprises, as in the case of the Late Triassic–Early Jurassic Neuquén Basin of Argentina, which had long been regarded as marine in character on the basis of a sparse bivalve biota but which elemental proxies demonstrated unambiguously to have been a freshwater lake with, at most, infrequent and limited seawater incursions (Remírez et al., 2024).

Salinity proxies have not yet been widely applied to continental depositional systems. Freshwater lakes, rivers, and swamps are expected to yield low proxy values, consistent with the low concentrations of B, Sr and SO₄^2– present in such environments (Wei and Algeo, 2020). However, lakes can vary widely in salinity as a function of the climate belt in which they are located, with elevated salinity associated with climatic aridity (Figure 1), a signal recorded by sediment B/Ga ratios in the Laguna Mar Chiquita of Argentina (Remírez et al., 2025b, c). Epeiric seas located in arid climate belts can also exhibit elevated salinities, resulting in B/Ga values far in excess of those of marine facies, as in the Late Devonian–Early Mississippian Williston Basin of North America (Remírez et al., 2025c). Given that the B/salinity ratios of continental waters are known to vary considerably and to diverge frequently from that of the modern ocean, inferring absolute salinity conditions from sediment B/Ga data of saline lakes will be difficult, and use of elemental salinity proxies in such systems may be limited primarily to assessing relative salinity variation.

Progress in the development of boron-based salinity proxies has accelerated recently. Specifically, several key studies advancing the methodology and/or application of these proxies have just been published or are in press as part of a special issue of the journal Chemical Geology:

(1) Wei et al. (2025b) tested four pretreatment methods for analysis of B by ICP-MS and determined that the alkaline fusion method is superior, yielding near-quantitative recovery of B in both siliciclastic and carbonate samples (>98%), whereas conventional methods making use of strong acids result in partial loss of B through volatilization during sample digestion, yielding recoveries ranging only from 67% to 93%.

(2) Algeo et al. (2025b) investigated the relationship between the B/Ga proxy and clay-mineral assemblages, establishing that, although illite-rich sediments tend to take up more B than illite-poor sediments, this effect is generally weak, and variations in clay-mineral assemblage only rarely have a measurable influence on salinity facies interpretations.

(3) Wei et al. (2025a) assessed whether elemental salinity proxies are applicable to carbonate sediments (note that, until now, these proxies were calibrated and tested only in fine-grained siliciclastic sediments, i.e., muds and shales). Although shales make up 60-70% of the sedimentary reservoir, carbonates represent another 20-25% of that reservoir, and development of an elemental salinity proxy applicable to such facies would be a major boon to paleoenvironmental research. This study, which examined carbonates and marls from a range of modern and ancient environments, determined that the B/Ga proxy is robust for samples with Al >2% (i.e., all marls and shales) but not for samples with Al < 2% (i.e., limestones) because of rising B/Ga values at low Al contents (this is due to Ga concentrations approaching zero together with the presence of a small non-clay B fraction in all samples). Significantly, this study proposes a new proxy known as “excess B” (calculated as B_xs = B_total – 6 × Ga) that can robustly estimate the salinity facies of carbonate sediments, with brackish carbonates typically having B_xs of <0 ppm and marine carbonates B_xs of 5-30 ppm (Figure 3). Development of an elemental salinity proxy for carbonates makes possible salinity analysis of mixed siliciclastic and carbonate successions such as Carboniferous cyclothems, as shown in the same special issue by Algeo et al. (2025a).

(4) Remírez et al. (2025c) evaluated the application of the B/Ga proxy to hypersaline facies, examining both siliciclastic and carbonate sediments in marine and lacustrine settings. This study found that B/Ga values in hypersaline siliciclastic facies are > 12, ranging up to 30–40 in the Devonian-age Bakken Shale. Although B/Ga cannot be applied to carbonates, this study found that, compared to B_xs values of ~0–20 ppm in marine carbonates, hypersaline carbonates yield B_xs values of > 40 ppm (n.b., 20 ppm–40 ppm is a fuzzy threshold between marine and hypersaline conditions; Figure 3). This work serves to extend the utility of boron-based salinity proxies beyond marine conditions.

The theoretical framework for the use of elemental salinity proxies, especially boron-based proxies such as B/Ga and excess B, has now been firmly established, so elemental salinity proxies can be considered to be “worth their salt”. In the future, serious paleoenvironmental research will need to routinely make use of these proxies in order to evaluate depositional salinity conditions and their relationship to other watermass properties (e.g., redox, productivity, and hydrographic restriction). There remain areas in which further work is urgently needed, however: (i) the influence of burial diagenesis, i.e., how do time, temperature, and fluids potentially alter boron-based salinity proxies? (cf. Liu et al., 2024), and (ii) are bulk-sediment boron isotopes controlled by salinity, pH, or other watermass parameters? (cf. Wei et al., 2022).

References

[1] Algeo, T. J., Wei, W., Sweet, D. E., 2025a. Salinity Variation in Carboniferous Cyclothemic Successions. Chemical Geology, 683: 122760. https://doi.org/10.1016/j.chemgeo.2025.122760

[2] Algeo, T. J., Wei, W., Remírez, M. N., et al., 2025b. Influence of Clay-Mineral Assemblages on the B/Ga Salinity Proxy. Chemical Geology, 688: 122837. https://doi.org/10.1016/j.chemgeo.2025.122837

[3] Cheng, M., Zhang, Z. H., Algeo, T. J., et al., 2021. Hydrological Controls on Marine Chemistry in the Cryogenian Nanhua Basin (South China). Earth-Science Reviews, 218: 103678. https://doi.org/10.1016/j.earscirev.2021.103678

[4] Gilleaudeau, G. J., Wei, W., Remírez, M. N., et al., 2023. Geochemical and Hydrographic Evolution of the Late Devonian Appalachian Seaway: Linking Sedimentation, Redox, and Salinity across Time and Space. Geochemistry, Geophysics, Geosystems, 24(8): e2023GC010973. https://doi.org/10.1029/2023GC010973

[5] Liu, Z. H., Algeo, T. J., Arefifard, S., et al., 2024. Testing the Salinity of Cambrian to Silurian Epicratonic Seas. Journal of the Geological Society, 181(3): jgs2023–jgs2217. https://doi.org/10.1144/jgs2023-217

[6] Liu, Z. H., Algeo, T. J., Brocks, J. J., et al., 2025. Salinity Reconstruction in Proterozoic Depositional Systems. Geological Society of America Bulletin, 137(1/2): 447–464. https://doi.org/10.1130/b37489.1

[7] Remírez, M. N., Algeo, T. J., 2020. Paleosalinity Determination in Ancient Epicontinental Seas: A Case Study of the T-OAE in the Cleveland Basin (UK). Earth-Science Reviews, 201: 103072. https://doi.org/10.1016/j.earscirev.2019.103072

[8] Remírez, M. N., Algeo, T. J., Shen, J., et al., 2024. Low-Salinity Conditions in the “Marine” Late Triassic-Early Jurassic Neuquén Basin of Argentina: Challenges in Paleosalinity Interpretation. Palaeogeography, Palaeoclimatology, Palaeoecology, 646: 112216. https://doi.org/10.1016/j.palaeo.2024.112216

[9] Remírez, M. N., Schwarz, E., Gilleaudeau, G. J., et al., 2025a. Seas to Lakes and Vice-versa: A Test of Facies versus Elemental Salinity Proxies. Chemical Geology, 678: 122663. https://doi.org/10.1016/j.chemgeo.2025.122663

[10] Remírez, M. N., Gilleaudeau, G. J., McBride, R., et al., 2025b. Calibrating Elemental Salinity Proxies in Holocene Sedimentary Environments. Chemical Geology, 678: 122664. https://doi.org/10.1016/j.chemgeo.2025.122664

[11] Remírez, M. N., Gilleaudeau, G. J., Sahoo, S. K., et al., 2025c. Beyond Evaporites: Assessing Hypersalinity Using Boron-Based Sedimentary Proxies. Chemical Geology. In press

[12] Song, Y., Gilleaudeau, G. J., Algeo, T. J., et al., 2021. Biomarker Evidence of Algal-Microbial Community Changes Linked to Redox and Salinity Variation, Upper Devonian Chattanooga Shale (Tennessee, USA). GSA Bulletin, 133(1/2): 409–424. https://doi.org/10.1130/B35543.1

[13] Wei, W., Algeo, T. J., 2020. Elemental Proxies for Paleosalinity Analysis of Ancient Shales and Mudrocks. Geochimica et Cosmochimica Acta, 287: 341–366. https://doi.org/10.1016/j.gca.2019.06.034

[14] Wei, W., Yu, W. C., Algeo, T. J., et al., 2022. Boron Proxies Record Paleosalinity Variation in the North American Midcontinent Sea in Response to Carboniferous Glacio-Eustasy. Geology, 50(5): 537–541. https://doi.org/10.1130/G49521.1

[15] Wei, W., Algeo, T. J., Meyer, D., et al., 2025a. Utility of the B/Ga Salinity Proxy in Carbonate and Marly Sediments. Chemical Geology, 682: 122751. https://doi.org/10.1016/j.chemgeo.2025.122751

[16] Wei, W., Algeo, T. J., Chen, L., et al., 2025b. Evaluation of Sample Pretreatment Methods for Boron Concentration Analysis. Chemical Geology, 687: 122838. https://doi.org/10.1016/j.chemgeo.2025.122838

RIGHTS & PERMISSIONS

China University of Geosciences (Wuhan) and Springer-Verlag GmbH Germany, Part of Springer Nature