Start with the question that a pyrogram can actually answer
Did you know that one heating curve can tempt a reader into claiming three different things: where organic matter came from, how far it has matured, and how much petroleum it might generate? Rock-Eval pyrolysis is useful precisely because its peaks summarize a controlled thermal experiment. But those claims are not interchangeable. An instrument measures products released from a prepared sediment aliquot as temperature follows a defined programme; it does not watch a buried ecosystem or a petroleum reservoir operating in nature.
When a marine sediment report shows S1, S2, S3, total organic carbon, hydrogen index, oxygen index and Tmax, translate each entry as a measurement with units and a method. Then translate the pattern as an interpretation that may depend on carbonate content, thermal history, mineral matrix, organic matter type and analytical limits. This distinction is the difference between a useful reading guide and a confident story unsupported by the specimen. The examples below use invented teaching data, not observations from a particular ocean core.
A fifty-second summary for the first pass
Rock-Eval heats a small dry sediment sample under a programmed sequence. The S1 peak reflects hydrocarbons readily released at an early heating stage. S2 reflects additional hydrocarbons generated during further pyrolysis of organic material; its peak temperature is Tmax. S3 represents carbon dioxide produced during the relevant stage, although instrument generations differ in how they detect and partition signals. TOC expresses organic carbon per mass of sample. Hydrogen index is conventionally S2 multiplied by 100 and divided by TOC expressed as weight percent; oxygen index uses S3 in the same manner.
Do not translate HI into a percentage of hydrogen atoms, or OI into the oxygen concentration of the sediment. Do not translate Tmax into the maximum temperature the sediment once experienced. They are derived indices and a laboratory peak location, respectively. Weak signals, weathering, contamination, carbonate-derived carbon dioxide and method-specific calibration can undermine a tidy classification. Read the laboratory protocol before comparing one expedition with another.
The physical specimen before the numbers
A core interval is not a homogeneous bottle of material. Mud, biogenic carbonate, plant fragments, volcanic grains, pyrite, shell debris and small organic-rich laminae can coexist over centimetres. The laboratory typically dries, grinds and subsamples material, but the documented treatment matters. A result represents the tested aliquot, not automatically every millimetre between the recorded upper and lower depths. If one half of a split contains a dark lamina and the other contains a shell, even careful analysts may obtain different dry-weight concentrations.
Write down core identifier, section, depth convention, sample mass, moisture treatment, pretreatment and date. Document whether the reported number is on dry whole sediment, carbonate-free sediment, or extracted organic matter. A percentage can look the same across tables even when the denominator changes. That is a fundamental translation error because an apparent downcore trend may instead be a change in preparation or mass basis.
Where the S1 signal comes from
S1 is commonly reported as milligrams of hydrocarbon per gram of dry rock or sediment. In a classic Rock-Eval implementation, readily volatilized or released compounds emerge during an initial lower-temperature stage. The peak can include native mobile hydrocarbons, retained bitumen and material introduced during sampling or laboratory handling. It is not an automatic measurement of free oil in a commercial reservoir, and a large S1 in a young marine mud should provoke questions rather than a dramatic conclusion.
A solvent residue, lubricating oil, plastic handling contamination or storage effects can elevate the early signal. Conversely, volatile loss during drying or prolonged storage can reduce it. Before comparing a freshly frozen core with a sample archived for years, ask whether identical storage and preparation were used. A reported S1 of 0.20 mg HC/g and one of 0.40 mg HC/g differ numerically by a factor of two, but only controlled blanks and replicate results justify calling that difference geological.
What S2 measures and what it cannot mean by itself
S2 is conventionally milligrams of hydrocarbon-like products generated per gram of dry sediment during programmed pyrolysis. It is associated with cracking and thermal decomposition of kerogen and other pyrolyzable organic matter. A larger S2 often indicates more remaining hydrocarbon-generating potential in an appropriate source-rock context, but the peak depends jointly on amount, composition, preservation and thermal history. It is not a direct census of past plankton abundance.
Imagine sample A has 0.8 mg HC/g S2 with 0.4 wt% TOC while sample B has 1.6 mg HC/g S2 with 2.0 wt% TOC. B produces twice the S2 per gram of sediment, yet its S2 per unit organic carbon is much smaller. The two readings answer different questions. The concentration controls how much material emerges per gram of whole sample; a carbon-normalised index helps explore the composition of the surviving organic matter.
What S3 measures and why carbonates matter
S3 is conventionally reported as milligrams of carbon dioxide per gram of dry sediment, associated with the instrument’s specified heating programme and carbon dioxide measurement window. It is used in calculating the oxygen index. The challenge in carbonate-rich marine sediments is that carbon dioxide can emerge from more than organic matter under some conditions or analytical programmes. Mineral decomposition and acid pretreatment choices may contaminate the tidy relationship that readers expect between S3 and organic oxygen functionality.
The laboratory’s protocol may report separate mineral carbon and organic carbon signals, or use different thermal stages from an older instrument. Never assume two methods with the same column label are identical. Where carbonate reaches substantial proportions, treat a high OI with caution and inspect independent carbonate, TOC and instrument data. In particular, the oxygen index is not a dissolved-oxygen record of ancient bottom water.
Read Tmax as a peak position, not an ancient thermometer
Tmax is the temperature at which the S2 pyrolysis peak reaches its maximum under the analytical ramp, usually reported in degrees Celsius. It is sensitive to organic composition and maturation, but the laboratory heating rate, instrument design, signal strength and overlapping peaks matter. A value of 430 °C is not evidence that the seabed ever reached 430 °C. It indicates the peak of a laboratory response; an ancient burial temperature would have to be inferred with suitable thermal-history models and independent constraints.
Conventional maturation intervals can be useful broad heuristics for appropriate source-rock systems, but young marine sediments often behave badly within such templates. Weak S2 peaks produce unstable Tmax estimates, and immature organic matter, oxidised fractions or mixed sources can alter shape. A data table that lists a precise Tmax to one decimal place does not guarantee one-decimal-place geological information. Precision in display is not precision in interpretation.
The denominator that makes HI easy to misread
Hydrogen index is commonly HI = 100 × S2 / TOC when S2 is recorded in mg hydrocarbon per gram sediment and TOC in weight percent. The resulting customary unit is mg HC per gram of total organic carbon. This is not a literal weight fraction of elemental hydrogen. The factor of 100 converts percent carbon to grams of organic carbon per hundred grams of rock, producing a coherent ratio. Readers who omit the factor understate HI by one hundred.
With S2 = 3.0 mg HC/g sediment and TOC = 1.5 wt%, HI = 200 mg HC/g TOC. If someone divides 3.0 by 1.5 and reports HI = 2, their arithmetic has silently mixed percent and fraction conventions. Restate the denominator every time you teach this calculation. The ratio can be especially volatile at very low TOC because a modest absolute error in carbon becomes a large relative error.
The oxygen index uses a different product
Oxygen index is commonly OI = 100 × S3 / TOC, with S3 in mg CO2/g sediment and TOC expressed as wt%. Thus S3 = 0.9 mg CO2/g and TOC = 1.5 wt% give OI = 60 mg CO2/g TOC. The symbol OI can resemble a chemical oxygen-content ratio, but it is an operational index of carbon dioxide released under the defined method. It cannot simply be converted into the atomic O/C ratio without additional analysis and validated relationships.
On a plot of HI against OI, broad organic-matter groupings may suggest marine algal, terrestrial higher-plant, highly oxidised or thermally altered contributions. Yet mixing and preservation complicate the picture. A high OI could reflect genuine organic composition, contamination by carbonates, or altered measurement conditions. The correct translation is ‘consistent with under these constraints’, not ‘proves the source was’.
TOC is a separate measurement with a separate uncertainty
Total organic carbon may be reported from combustion analysis, coulometry after separating carbonate, or an instrument-specific estimate from pyrolyzed and residual carbon. Do not silently treat all TOC columns as measured by the same procedure. An independent elemental analyser has its own blanks, calibration standards, carbonate-removal issues and lower quantification limit. If a sample contains a high fraction of inorganic carbonate, an error in removing or estimating that carbonate can dominate the resulting organic carbon.
Suppose total carbon is 3.20 wt% and carbonate-derived inorganic carbon is 2.70 wt%. Their difference is TOC = 0.50 wt%. If each large component has a moderate analytical uncertainty, the uncertainty on the difference may be proportionally large. When HI or OI divides by that difference, its reliability worsens. Report the carbon method and whether uncertainties and replicates support the final digits.
Production index: a helpful ratio with multiple possible stories
Production index is conventionally PI = S1 / (S1 + S2). Both signals need the same mass basis, and the denominator cannot be zero. With S1 = 0.3 and S2 = 2.7 mg HC/g, PI = 0.10. This can help describe the proportion of early-released hydrocarbons relative to the total measured early-plus-pyrolyzable hydrocarbon signal. Its usual petroleum-maturity interpretation is contingent, not universal.
Consider a shallow sediment with a small S2 but contaminated S1. The ratio rises even without meaningful thermal maturation. In a mature source rock PI may increase as generated hydrocarbons become mobile, but migration can also add or remove material. If PI looks extreme, investigate the absolute peak sizes. A fraction of 0.8 calculated from two near-detection-limit numbers is much less informative than the same fraction from robust signals.
Read the axes before interpreting a crossplot
Some reports plot hydrogen index on the vertical axis and oxygen index on the horizontal axis. Others plot Tmax against HI or use a Van Krevelen-style diagram with inferred chemical atomic ratios. These are not interchangeable plots. Before giving the axis a familiar name, check symbols, units, logarithmic scaling, detection thresholds, colour groups and whether values were censored. A sloping point cloud can originate from a common denominator rather than a genuine geochemical evolution.
Suppose S2 and S3 change little but TOC halves. Both HI and OI double because they divide by the same smaller quantity. A diagonal shift in HI–OI space therefore does not by itself prove a compositional change. It may reflect dilution, measurement uncertainty or a real change in carbon pool. Examine the underlying S2, S3 and TOC values together instead of interpreting only the derived indices.
An original worked example: the normalisation reversal
Two teaching samples come from neighbouring invented core intervals. Sample U has TOC 0.60 wt% and S2 1.80 mg HC/g. Sample V has TOC 1.80 wt% and S2 3.60 mg HC/g. Their hydrogen indices are 100 × 1.80 / 0.60 = 300 and 100 × 3.60 / 1.80 = 200 mg HC/g TOC. V releases more hydrocarbon per gram of sediment while U releases more per gram of organic carbon. Both statements are true.
Check the calculation independently by imagining 100 g dry sediment. U contains 0.60 g organic carbon and releases 180 mg hydrocarbon-equivalent S2 products; 180/0.60 = 300 mg/g. V contains 1.80 g organic carbon and releases 360 mg; 360/1.80 = 200 mg/g. The different denominators explain the reversal. The interpretation is not that U contained more organic life. Its surviving organic fraction has a different pyrolysis response.
An original worked example: uncertainty becomes a ratio problem
An invented sample has S2 = 1.00 ± 0.08 mg HC/g and TOC = 0.40 ± 0.05 wt%, with uncertainties treated as independent for this teaching calculation. HI = 250 mg HC/g TOC. For small independent errors, the approximate relative standard uncertainty of a ratio is the square root of the sum of squared relative uncertainties: sqrt[(0.08/1.00)^2 + (0.05/0.40)^2] = 0.148. The approximate uncertainty in HI is 37 mg HC/g TOC.
Thus a defensible rounded expression is about 250 ± 37 mg HC/g TOC under those assumptions, not an implied exact value of 250.000. If S2 and TOC share sample-preparation errors, independence may not hold. In a real paper use the actual calibration, replicate and covariance information where available. Propagation does not rescue a biased extraction or an incorrectly determined carbon denominator.
An original worked example: a misleading S3–OI ranking
Sample W has S3 = 0.90 mg CO2/g and TOC = 3.00 wt%; sample X has S3 = 0.60 mg CO2/g and TOC = 0.60 wt%. W has higher raw carbon dioxide release, but OI is 30 mg CO2/g TOC. X has lower raw S3 and OI is 100 mg CO2/g TOC. The ranking changes because X contains much less organic carbon per gram sediment. That is a transparent mathematical result, not evidence that X’s bottom water once held more oxygen.
A reader should ask whether both results used the same carbon and S3 methods, whether carbonate interfered, and whether the small TOC denominator was measured reliably. To perform a transfer test, replace X’s TOC by 1.20 wt% while holding S3 constant: OI falls to 50. The change arises entirely from normalisation, illustrating why index trajectories must be checked against their underlying measurements.
An original worked example: the PI numerator
Invented sample Y has S1 = 0.15 mg HC/g and S2 = 2.85 mg HC/g. PI = 0.15/(0.15 + 2.85) = 0.05. Sample Z has S1 = 0.20 and S2 = 0.30, so PI = 0.40. The high PI of Z does not imply more readily released hydrocarbon mass than Y by an order of magnitude: S1 rises only from 0.15 to 0.20. Most of the fraction change is the collapse of S2.
Now ask whether Z’s S2 is above the reliable quantification range, whether it contains oxidised carbon, and whether it belongs to a young sediment whose maturity is not captured by source-rock heuristics. If S2 were reported as zero because it was below detection, PI may be mathematically 1 but scientifically uninformative. Where measurements are censored, say so instead of using a dramatic ratio.
What an S2 shoulder or double peak can mean
An analyst may see an S2 curve with a shoulder or more than one maximum. Mixtures of organic fractions can contribute, but instrument response, overlapping mineral signals and sample heterogeneity are also candidates. A single Tmax obtained from the tallest point collapses that complexity. If a report permits inspection of the full pyrogram, look at peak width, asymmetry, baseline and repeated analyses. The secondary feature can make a nominal ‘maturity’ number unstable.
For an exercise, imagine two apparent S2 maxima around 418 and 443 °C. If small changes in baseline correction shift the chosen maximum from one to the other, reporting a single 443 °C maturity label is misleading. The appropriate response is to describe the curve, inspect controls and consider fraction-specific methods. A thermal curve is evidence; a classification chart is a model layered on the evidence.
Why the carbonate fraction can distort a trend
Many marine cores alternate between carbonate-rich biogenic layers and terrigenous clays. A high carbonate layer dilutes organic carbon on a whole-sediment basis even if the input of organic material per unit area is unchanged. The same layer can affect interpretation of carbon dioxide signals depending on method. Therefore a decrease in TOC and an increase in OI may not represent a sudden switch from productive marine ecology to oxidised terrestrial debris.
If a layer contains 80 g carbonate and 20 g other solids per 100 g sediment, every constituent restricted to the other solids will appear five times more concentrated when expressed per gram of the noncarbonate fraction rather than per gram of whole sediment. Neither denominator is inherently wrong. One answers the bulk sediment question; the other explores the composition of a separated fraction. Label them and do not mix.
Why ash and clay complicate the thermal response
Clay mineral surfaces can adsorb organic molecules and influence pyrolysis products and apparent peak shapes. Sulfides, carbonates, hydrated minerals and reactive phases may also produce thermal features. Laboratories sometimes use different matrix corrections or calibration checks for unusual sediments. Do not assume a generic Rock-Eval workflow validates every mineral-rich specimen. Describe the mineralogy, inorganic carbon and instrument model when interpretation depends upon a subtle thermal feature.
A sample might have the same actual organic carbon mass as another but a different clay-to-carbonate ratio. The two may respond differently in the instrument because the matrix affects released compounds and carbon dioxide. Cross-check with independent TOC, microscopy, mineralogical information, or solvent extraction when a major paleoenvironmental claim hangs on a small change. Rock-Eval is one window on a complex sample, not a magical elemental analyser.
The sample mass and the detection-floor problem
Instrument signals are generated from a finite amount of material. Very small organic peaks may approach blanks and lower quantification limits, especially where TOC is low. A laboratory can display an S2 or S3 number even when its uncertainty is a large share of that number. That is why old rules of thumb restricting Rock-Eval interpretation in low-TOC samples should be understood as warnings about signal quality, not as universal physical boundaries.
Ask for sample mass, control material, blank subtraction, peak integration rule and whether the instrument flags a value as below quantification. Suppose S2 is 0.06 mg HC/g with a reporting uncertainty of 0.04. That is not equivalent to a robust 0.60 ± 0.04. Dividing the fragile 0.06 by a tiny TOC can produce an impressive HI that has almost no diagnostic power. Stable interpretation starts with robust primary measurements.
Use standards and duplicates to look for drift
Standards with known reference behaviour help track instrument response and calibration. Laboratory duplicates test sample preparation and measurement repeatability, while procedural blanks reveal background contamination. A check standard measured before and after a sequence helps detect drift. But a duplicate that agrees is not a proof of geological accuracy if both aliquots were prepared from a biased or unrepresentative part of the core.
A useful data audit has separate columns for analytical replicate, field or core replicate, blank, reference material, instrument batch and correction rule. A spike recovery answers yet another question: whether an added material can be recovered from the matrix. If a report gives only final HI and Tmax, the reader cannot independently evaluate drift. State that limitation rather than assuming an elaborate quality system was followed.
Do not infer depositional oxygen from OI alone
The oxygen index is often misunderstood because the word oxygen evokes water-column oxygenation. Sediment redox and bottom-water oxygen influence organic preservation in some settings, but OI measures the pyrolysis-released carbon dioxide normalised to organic carbon. Different source materials, alteration histories, carbonates and matrix artefacts can shape it. It is not a direct proxy for dissolved oxygen concentration in millilitres per litre, micromoles per kilogram or percent saturation.
If you want to discuss ancient water-column oxygen, independent evidence may include iron speciation, trace-metal behaviour, benthic assemblages, sedimentary structures and pore-water redox processes. Even those need careful assumptions and geological context. Rock-Eval OI can be part of an organic-matter story, but translating it as ‘the bottom water became oxygen rich’ without corroboration changes an analytical index into an unsupported environmental observation.
Do not confuse organic matter type with original ecosystem fraction
A high hydrogen index can be compatible with lipid-rich aquatic organic matter, while lower values may be associated with land-derived, oxidised or mature material. Yet the organic fraction that survives burial is selectively preserved. Much of the original biomass may have been respired or chemically transformed before sampling. The sample is not an unbiased archive of whole organisms, and the different carbon pools may release very different pyrolysis products.
A sediment may combine plankton-derived molecules, vascular plant waxes, bacterial biomass, reworked fossil carbon and contaminants. Mixing does not necessarily produce an obvious midpoint because endmembers have different carbon concentrations and response factors. A claim such as ‘70% terrestrial’ requires a validated quantitative mixing model, not a casual visual reading of HI and OI. Prefer ‘consistent with increased terrigenous contribution’ when that is the limit of evidence.
Maturation depends on time and temperature, not Tmax alone
Organic maturation is driven by thermal history over geological time, influenced by pressure, mineral environment and chemistry. Rock-Eval Tmax is one empirical indicator in suitable systems, not an equation that uniquely yields burial depth, elapsed time or maximum paleotemperature. A marine deposit can experience heating from burial, geothermal anomalies or later tectonic events. Different pathways may yield similar present-day thermal signals.
Where maturity is central, compare with vitrinite reflectance where available, biomarker isomerisation, independent thermal models and stratigraphic constraints. A shallow modern mud can have low or irregular Tmax despite a complicated degradation history, while a reworked organic particle may be more mature than the enclosing sediment. The word ‘maturity’ must identify what material and analytical concept is being described.
An original case: low TOC does not equal low accumulation
Two hypothetical layers each receive 1.0 g organic carbon per square metre per day, but one receives much more mineral sediment. The organic carbon weight percentage can be smaller in the rapidly accumulating layer due to dilution. Conversely, a high TOC layer may have low total carbon burial if very little sediment accumulates. Translate concentration and accumulation as different dimensions. A concentration of 1 wt% is one gram organic carbon per 100 g dry sediment, not a flux through the seabed.
To estimate burial flux, combine a suitable dry mass accumulation rate with a concentration expressed as a mass fraction. If dry sediment accumulation is 100 g/m²/year and TOC is 2 wt%, organic carbon burial is 2 g C/m²/year, before considering postdepositional loss and other constraints. Rock-Eval S2 can also be converted to a mass-accumulation proxy by multiplying its concentration by dry mass accumulation, but that does not make it a direct primary productivity measure.
An original case: water content changes the denominator
Suppose a wet core sample weighs 5.0 g and contains 2.0 g water. It has 3.0 g dry solid. If pyrolysis produces 6 mg S2-equivalent hydrocarbons from the aliquot, the dry-basis concentration is 2.0 mg HC/g dry sediment, not 1.2 mg/g wet sediment. The latter number uses a different denominator. If one database silently combines wet-basis and dry-basis results, an apparent geographic or temporal signal can be created from moisture.
Marine sediment often contains pore-water salt that remains after drying. Some analytical programmes correct dry mass for salt, others specify an operational dry-weight basis. These conventions matter most in wet, unconsolidated muds. State whether salt correction occurred, especially when comparing near-surface samples, and do not mix ‘per gram dry bulk sample’ with ‘per gram salt-corrected dry sediment’ without reconciliation.
An original case: carbonate-free data are not bulk data
Imagine an aliquot of 100 g dry sediment with 40 g carbonate. After carefully removing carbonate, the remaining dry solid weighs 60 g. A compound measured at 3 mg per gram of the carbonate-free residue corresponds to 1.8 mg per gram of original bulk sediment if no losses or additions occurred. The numerical result is different because the denominator changed, not because chemistry changed. Real acid pretreatment may alter some organic constituents, adding another reason for caution.
When comparing Rock-Eval and independent TOC results, track whether the analyses used bulk powder or different pretreated subsamples. If the process damages thermally labile molecules, multiplying by a mass conversion factor may not reconstruct the original signal. The proper translation may be ‘results on acid-treated material only’. A method description is part of the scientific claim, not paperwork detached from the results.
A full reading exercise with four linked indices
Consider invented sample Q with S1 = 0.24 mg HC/g, S2 = 2.16 mg HC/g, S3 = 0.54 mg CO2/g, TOC = 1.20 wt%, and a clean S2 maximum at 422 °C. Calculate HI = 180 mg HC/g TOC; OI = 45 mg CO2/g TOC; PI = 0.24/(2.40) = 0.10. S2 is nine times S1. Each number is dimensionally consistent and describes a separate aspect of the experiment.
A restrained interpretation says the sediment contains a measurable pyrolyzable organic fraction and a smaller readily released fraction, with derived indices that may be compared to method-matched reference materials. It does not automatically assign a source type or maturity zone. To make the answer stronger, ask for replicate precision, inorganic carbon, detection limits, mineral matrix, core context and independent biomarkers. If the S2 peak is broad, even the meaning of the recorded Tmax requires additional inspection.
Transfer exercise one: find the wrong factor
A worksheet claims that with S2 = 4.8 mg HC/g and TOC = 2.0 wt%, HI is 2.4. Diagnose the error before reading the answer. The student has divided the two displayed numbers without converting the percent convention to a mass ratio. The conventional HI is 100 × 4.8/2.0 = 240 mg HC/g TOC. Explain the calculation in words: 100 g dry sediment holds 2 g organic carbon and releases 480 mg hydrocarbon-equivalent S2 products, so the ratio is 240 mg/g.
The error matters because a factor of one hundred can send a point into the wrong part of a reference diagram and corrupt any downstream interpretation. Do not ‘fix’ it simply by memorising a formula if the lab report actually stores TOC as a decimal fraction instead of percent. First read the column units; then use the appropriate equation. The same symbols are not a licence to ignore a data dictionary.
Transfer exercise two: a fake maturation signal
An analyst observes PI rising from 0.05 to 0.35 across two layers and announces faster geological maturation. Work backwards. The early layer has S1 = 0.10 and S2 = 1.90 mg/g, yielding 0.05. The later layer has S1 = 0.14 and S2 = 0.26 mg/g, yielding 0.35. The denominator collapses while S1 changes little. Very low pyrolyzable mass, contamination, mixed sources or measurement-limit effects may explain the pattern.
The repaired claim is: the measured proportion of early-released hydrocarbons relative to S1+S2 increased. That statement is specific, testable and does not confuse an instrumental ratio with a geologic process. Investigation should compare absolute peaks, TOC, sample handling, pyrograms and maturation proxies. If both samples come from young seabed sediment, standard petroleum system rules may be particularly inappropriate.
Transfer exercise three: same HI, different deposits
Sediment A has S2 = 1 mg/g and TOC = 0.5 wt%. Sediment B has S2 = 8 mg/g and TOC = 4 wt%. Both have HI = 200 mg HC/g TOC. They can differ profoundly in organic carbon concentration, net sediment accumulation, preservation, mineral composition and amount of pyrolyzable carbon per unit bed mass. Identical carbon-normalised indices do not establish identical ecosystems or equal petroleum potential per unit mass of rock.
The correct answer is that HI alone cannot choose between the deposits. State the different bulk concentrations and investigate why they vary. A reader should also check whether a very low TOC sample has robust S2 and carbon measurements. In a school or university transfer test, the aim is not merely two correct divisions; it is recognising that normalisation removes information that may be scientifically essential.
Transfer exercise four: interpret a suspect Tmax
A pyrogram has a weak S2 peak that barely rises above the baseline and a software-generated Tmax of 457 °C. A colleague translates this as proof of overmature sediment. Diagnose the weak link: the peak temperature is unreliable when peak identification and integration are uncertain. If the heating curve has noise or an overlapping matrix response, a high reported maximum can be an algorithmic artefact.
The stronger response is to inspect raw traces, blank and standard runs, repeated measurements and independent maturity indicators. A software output is not equivalent to a verified specimen property. It can be appropriate to report ‘Tmax not interpreted due to weak S2’ rather than to place a dramatic number on a crossplot. Scientific translation sometimes means preserving an explicit missing or unusable result.
Interpreting a depth profile responsibly
A continuous core figure may display TOC, HI, OI, Tmax and lithology against metres below seafloor. Before reading it as a time series, establish the age–depth relationship. One metre of deposition can represent decades in a rapidly accumulating environment or far longer in a condensed interval. Hiatuses, slumps, turbidites and bioturbation can disrupt the correspondence between depth and time. Rock-Eval measurements add chemical information but do not solve the chronology.
Next compare trends with changes in carbonate, grain size, density and water content. A TOC decrease at a turbidite base might reflect dilution by mineral sediment; a high HI layer might concentrate a particular organic facies; a Tmax jump could be driven by poor peak quality. Triangulate rather than taking any one curve as the story. A good figure caption records mass basis, sampling interval, method, uncertainty and excluded values.
What cross-laboratory comparison requires
Different Rock-Eval generations, ramp programmes, detector calibration, sample sizes and processing conventions can yield values that are not strictly interchangeable. One laboratory may apply a protocol for sedimentary rocks while another uses a modern organic-matter approach for young marine mud. Numerical similarity can be accidental when the operational definitions differ. Before pooling data, inspect method sections and reference materials, then consider a paired subset analysed by both methods.
A scientifically useful harmonisation table includes instrument model, heating sequence, aliquot size, dryness, carbonate method, TOC method, blank, certified or in-house standard, peak integration, censoring convention and retained decimals. If you cannot reconcile a key field, do not simply append datasets and plot a trend. Flag an intermethod boundary and test whether the finding survives analysing each subset separately.
Claims to avoid in public explanations
Do not write ‘S2 is the amount of oil in the seabed’, ‘Tmax records the temperature of burial’, ‘OI measures sea oxygen’, or ‘HI tells us the percentage of algae’. Each converts a limited measurement into a direct environmental claim it does not support. Also avoid reporting index categories as immutable natural types. Organic matter is a mixture, and preservation, degradation and thermal transformation can move its response through a diagram over time.
Accurate wording can still be clear and inviting. Say ‘S2 measures hydrocarbons generated during the programmed laboratory heating’; ‘Tmax marks the strongest part of the S2 peak’; ‘HI compares that output with measured organic carbon’; and ‘an interpretation of source or maturity needs method-matched context’. These sentences are shorter and more accurate than embellished alternatives. Readers can use them as a disciplined translation glossary.
Questions a reader should ask the laboratory
Was the sample dried or freeze-dried, homogenised and carbonate-treated? Were results reported on bulk dry sediment or a corrected denominator? Which Rock-Eval instrument and temperature programme were used? How were S1, S2, S3 and TOC calibrated? Did low S2 make Tmax unreliable? What were the detection and quantification limits? Which blanks, duplicates and reference standards were run? Were traces inspected rather than only summary columns exported?
These questions are not an accusation against competent analysts. They are the ordinary path from a table to an auditable interpretation. A good laboratory report often answers many of them in a methods appendix. If the information is genuinely unavailable, mark the inference provisional. The limitations section should not be an apologetic footnote; it tells the reader precisely how far the numbers can support the statement.
A practical diagnosis-and-repair reading workflow
First identify the raw variables and their units. Second establish the sample basis and analytical protocol. Third recompute HI, OI and PI for a few rows to catch transcription or conversion errors. Fourth examine the primary peak magnitudes and any censoring before discussing ratios. Fifth inspect source and maturity hypotheses against mineralogy, TOC and independent evidence. Sixth decide which claims are robust, which are conditional and which should be withheld.
Imagine a chart showing high OI in a carbonate-rich layer with low TOC. A careless translation attributes it to strong ancient ocean oxygenation. The repair begins by recognising S3/TOC as a carbon dioxide response ratio, checking carbonate interference and low-TOC uncertainty, then comparing with genuine redox proxies. The useful conclusion may simply be that the organic response is poorly constrained. That is progress, not a failure to tell a story.
FAQ: Is there one universal HI threshold?
No. Reference ranges can suggest organic matter types for defined geological settings, but they depend on preservation, mixture, alteration, methods and source material. A single threshold cannot certify ‘marine’ versus ‘terrestrial’ without supporting observations. A highly degraded marine contribution may show a low HI; a land-derived lipid-rich contribution may behave differently than expected. The strongest explanation presents the range as a contextual heuristic and names what other evidence could change the interpretation.
Start with what the instrument demonstrably measured, then say what endmember or environmental process is compatible. Compare nearby lithology, organic petrography, carbon isotopes and diagnostic molecular biomarkers if available. The purpose of the index is to focus a hypothesis, not to replace the field and laboratory evidence needed to test it.
FAQ: Can I compare HI when the TOC method changes?
Only with care. If one campaign calculates TOC by difference between total and inorganic carbon while another uses a Rock-Eval estimate, shared errors may not be the same. The derived HI can change simply because of the denominator. An ideal check analyses overlapping reference or core samples with both workflows and reports bias and dispersion rather than assuming equivalence.
If no overlap exists, keep the methods labelled, examine trends within each dataset, and perform sensitivity tests over plausible carbon uncertainty. The question is not whether one method is always right; it is whether the comparison supports the specific size and direction of the claim being made. Honest captions preserve that distinction.
FAQ: Does higher TOC mean a richer ancient ocean?
Not necessarily. TOC in sediment is the result of organic supply, degradation, preservation, dilution and sediment accumulation. High primary productivity can coexist with low preserved organic carbon where remineralisation or mineral dilution is intense. Low productivity can coexist with a relatively high concentration in a slowly depositing environment. The mass accumulation rate and depositional context are therefore crucial.
Translate TOC as organic carbon concentration on the defined dry mass basis. If discussing export production, use independent age control and sedimentation rates as well as proxies appropriate to the setting. A single high TOC point is not an ocean productivity meter. The reader learns more by separating physical deposition from biological production than by memorising a simplistic sign.
FAQ: Why can an instrument produce apparently precise but implausible values?
Software integrates peaks and performs algebra even when the input data are poor. A baseline can shift, a low S2 signal can produce an unstable Tmax, and division by a tiny TOC can yield a huge HI or OI. The computer is consistent about calculation, not necessarily about the existence of a meaningful geological signal. Analytical acceptance limits and raw-trace review keep calculated precision in proportion to evidence.
An implausible value should trigger an inspection, not automatic deletion. Preserve the original measurement and its flag; determine whether contamination, preparation, instrument drift, mislabelled units or genuine unusual material explains it. Document the decision so later readers can distinguish a measured outlier from an excluded one. Reproducibility means preserving the path between raw data and published interpretation.
FAQ: Can a Rock-Eval profile be translated directly into an age?
No. The method provides chemical and thermal-response variables, not a calendar date or a unique elapsed burial time. A downcore profile can be compared with independently dated horizons, but changes in peak position or indices alone do not establish a chronology. Even burial maturity inference needs thermal context and suitable calibrations. A sediment from one environment may mature differently from a specimen with another organic composition and temperature history.
When a paper aligns Rock-Eval curves to climate events, verify the independent time scale: biostratigraphy, radiometric markers, magnetic stratigraphy or a probabilistic age–depth model. Do not infer an exact year from the shape of an organic carbon curve. Keep the chronology evidence separate from the geochemical explanation.
Independent transfer problems with answers
Problem A: A bulk dry sediment has TOC 2.5 wt%, S2 5.0 mg HC/g and S3 1.25 mg CO2/g. Calculate HI and OI. Answer: HI = 100 × 5.0 / 2.5 = 200 mg HC/g TOC; OI = 100 × 1.25 / 2.5 = 50 mg CO2/g TOC. Note that the different numerator units must be retained. Both are instrument-defined ratios rather than chemical atomic fractions.
Problem B: S1 is 0.50 mg HC/g and S2 is 1.50 mg HC/g. PI = 0.50/(0.50 + 1.50) = 0.25. If S2 falls to 0.50 with S1 unchanged, PI becomes 0.50. This ratio change does not independently prove maturation because loss or mismeasurement of S2 also changes the denominator. The appropriate explanation distinguishes the computation from a geological hypothesis.
Problem C: A listed Tmax is 462 °C but the associated S2 is barely distinguishable from a blank. Answer: do not assign a categorical maturity label on that number alone. Examine original traces and controls, replicate where possible, and treat the value as unreliable if its peak position cannot be established robustly. The most important answer is sometimes a request for better evidence rather than a category.
How to put the explanation into plain English
To translate a dense table for a colleague, begin with the purpose: the scientists heated dry sediment and recorded what came off at different steps. Next explain the numerator and denominator: hydrocarbon products per gram of sediment, then per gram of organic carbon. Only after that introduce the interpretive question, such as whether the measured organic residue resembles particular source and alteration patterns. Keep uncertainty and assumptions visible next to the claim.
A clear summary might say: ‘The sampled interval has measurable organic carbon, a comparatively moderate pyrolyzable hydrocarbon response after carbon normalisation, and a laboratory S2 peak whose temperature may be useful if peak quality is good. Its source and alteration history remain conditional on independent evidence.’ That statement is useful to a non-specialist without pretending the method contains an entire ocean history. Precision of language protects precision of science.
Primary method references and useful reading routes
IODP Expedition 344 Rock-Eval data report.
IODP Expedition 302 organic geochemistry methods.
Ocean Drilling Program technical note: Rock-Eval pyrolysis.
Google Search Central: people-first content.
How to Translate Easily: series index.
How English Works: Vocabulary and Meaning.