Dear RG members, IUGS is planning a new edition of the classical "Le Maitre" book devoted to the classification and nomenclature of igneous rocks. A group of 17 igneous petrologists (hereafter TGIR - Task Group on Igneous Rocks) is working for three years to update specific definitions or proposing entirely new sections.
As the Chair of the TGIR, I would like to start a discussion with all the interested people that want to give help concerning this task. I and the other members of the TGIR will start posting a series of arguments that will greatly benefit from your comments, so I hope to receive stimulating feedback.
The IUGS book on the classification and nomenclature of igneous rocks (Le Bas) 2002 does not report any indication on two terms often reportd in igneous petrology studies, i.e., tholeiitic rock series and calc-alkaline (or calcalkaline) rock series.
Le Maitre (2002) only reports little comments on "tholeiitic basalt", simply defined as "A common variety of basalt composed of labradorite, augite, hypersthene (= enstatite) or pigeonite, with olivine (often showing a reaction relationship) or quartz, and often with interstitial glass. The Subcommission recommends that this term should be used instead of tholeiite.". IUGS distinguises various types of basalts (e.g., alkali basalt, high-alumina basalt, island arc basalt, mid-ocean ridge basalt, olivine basalt, olivine tholeiite, subalkali basalt, tholeiite, tholeiitic basalt, transitional basalt), but unfortunately without providing any key to distinguish among them.
In the definition of "Quartz dolerite", Le Maitre (2002) reports: "The rock has tholeiitic affinities and its pyroxenes are usually subcalcic augite accompanied by pigeonite or orthopyroxene.", but it is a pity that there is no definition for "tholeiitic affinity".
The new version of the IUGS book has to fill this gap. Several scientific articles have dealt with this topic, and two excellent reviews are written by Sheth et al. (2002) and Arculus (2003). Both the articles underline that the articles published in the last 40-50 years completely changed the original meaning of the term "calc-alkaline".
1. Peacock (1931) coined this term using a simple plot reporting both CaO vs SiO2 (usually with negative correlation) and (Na2O+K2O) vs SiO2 (usually with positive correlation). Depending on the SiO2 value where the two trends intersect, the rock series were defined as alkalic (61 wt% SiO2). You understand that the term "calc-alkaline", when defined, had nothing to do with the actual meaning. As the term indicated, it was referring to lime ("Calc") and alkali.
2. Then Kennedy (1933) proposed the existence of a primary magma series evolving towards SiO2-saturated to SiO2-oversaturated compositions, defined it as "tholeiitic".
3. Only twenty years later, other authors started to connect "tholeiitic" and "calc-alkaline" terms referring respectively to series with and without iron enrichment with increasing SiO2 (Nockolds and Allen, 1953).
4. Thirteen years later, Kuno (1966) proposed for the term "calc-alkaline" a meaning of a magma that evolved under oxidized conditions, distinguishing two fields for "tholeiitic" and "calc-alkaline" series in his triangular plot AFM (A = alkali = Na2O+K2O; F = FeO*; M = MgO).
5. Eight years later (Miyashiro, 1974) used a simple equation to distinguish tholeiitic from calc-alkaline series: SiO2 (wt%) = 6.4 × FeO*/MgO + 42.8 (with tholeiitic rocks plotting above this straight line, and calc-alkaline rocks plotting below the line). Note that FeO* is total iron as FeO.
6. One year later, Middlemost (1975) proposed a simple diagram reporting A.I. (Alkali Index = (Na2O+K2O)/[(SiO2-43)*0.17] vs. Al2O3, to distinguish tholeiitic basalts (Note: only basalts) from calc-alkaline (or High-Al) basalts.
7. One year after, Peccerillo and Taylor (1976) published their famous article on Eocene Kastamonu igneous rocks (Pontites, NE Anatolia), proposing the K2O vs. SiO2 diagram, where they distinguished four series, with increasing K2O content and K2O/SiO2: 1) arc tholeiitic series, 2) calc-alkaline series, 3) high-K calc-alkaline series and 4) shoshonite series. These authors summarized a series of articles and inferences raised just after the birth of plate tectonics. Since then the term "calc-alkaline" started to be used as a sort of synonym of "subduction-related".
8. Nearly 10 years later, Middlemost (1985) noted that the typical andesitic rocks emplaced above subduction systems plot in the "calcic" field of Peacock (1931), and Icelandic tholeiites plot in the "calc-alkalic" field, to underline that the original meaning of "calc-alkalic" was completely distorted.
The Kuno (1966) and Miyashiro (1974) diagrams are useful discriminators for tholeiitic and calc-alkaline rock series but, as underlined by Sheth et al. (2002) they do not deal with CaO (Calc), so they refer to a concept that uses different criteria. Another problem of AFM is that evolved tholeiitic and calc-alkaline terms greatly overlap and the three components make up less than 50% of the oxides in the rocks.
This problem was particularly relevant, according to Arculus (2003), who proposed to use Kuno (1966) discrimination diagram [i.e., (FeO*/MgO) vs. SiO2] but instead of using "tholeiitic" and "calc-alkaline" terms, he suggested using High-Fe, Medium-Fe and Low-Fe. While reasonably solid, this proposal got little agreement by the scientific community, who demonstrated a great inertia in changing the two-term definition with a three-term definition.
What should IUGS do? We believe that it is unreasonable to propose to delete the "tholeiitic" and "calc-alkaline" terms, so we feel it is necessary to leave them, however, giving them the petrological significance proposed by Kuno (1966), Miyashiro (1974) and Arculus (2003), i.e., associated to differences in oxygen fugacity conditions and absolute FeO/MgO ratios.
We have to propose that the Peccerillo and Taylor (1976) diagram cannot be used as a reference to distinguish between tholeiitic and calc-alkaline rocks.
What is important, is that the idea of the IUGS TGIR the terms "tholeiitic" and "calc-alkaline" should not have any tectonic significance, i.e. tholeiitic compositions are not necessarily emplaced away from active subduction settings, an calc-alkaline does not mean emplacement above active subduction settings. We have to state that the term "tholeiitic" leaves the original significance of Kennedy (1933), but the term "calc-alkaline" has a completely different significance compared to the original statement of Peacock (1931).
Proposed items to be added in the Glossary of terms:
THOLEIITIC SERIES
Subalkaline rock series characterised by total iron enrichment with differentiation, leading to SiO2-saturated to SiO2-oversaturated compositions. The term should not be used to infer a specific tectonic setting. Tholeiitic series rocks have to plot in the tholeiitic series field of the AFM diagram of Kuno (1966; where A = alkali = Na2O+K2O; F = FeO*; M = MgO) or in the tholeiitic series field diagram of Miyashiro [1974; (FeO*/MgO) vs. SiO2]. In both cases FeO* is total iron expressed in the reduced form. If two iron oxidation states are available, FeO* = FeO + 0.8998*Fe2O3. In case only ferric iron is available, FeO* = 0.8998*Fe2O3. See calc-alkaline series.
CALC-ALKALINE (or CALCALKALINE) SERIES
Subalkaline rock series characterised by total iron depletion with differentiation, leading to SiO2-saturated to SiO2-oversaturated compositions. The term should not be used to infer a specific tectonic setting. Calc-alkaline series rocks have to plot in the calc-alkaline series field of the AFM diagram of Kuno (1966; where A = alkali = Na2O+K2O; F = FeO*; M = MgO) or in the calc-alkaline series field diagram of Miyashiro [1974; (FeO*/MgO) vs. SiO2]. In both cases FeO* is total iron expressed in the reduced form. If two iron oxidation states are available, FeO* = FeO + 0.8998*Fe2O3. In case only ferric iron is available, FeO* = 0.8998*Fe2O3. See tholeiitic series.
This is a first draft, and comments are very welcome. Considering it is a hotly debated topic, I invite all the RG members to be short and directly to the point.
Thanks,
michele
Arculus, 2003: https://academic.oup.com/petrology/article-abstract/44/5/929/1464257
Kennedy, 1933: https://ui.adsabs.harvard.edu/abs/1933AmJS...25..239K/abstract
Kuno, 1966: https://link.springer.com/article/10.1007/BF02597153
Middlemost, 1975: https://www.sciencedirect.com/science/article/pii/0012825275900392
Middlemost, 1985: Magmas and magmatic rocks: An introduction to igneous petrology: London, UK, Longman, 266 pp.
Miyashiro, 1974: https://cir.nii.ac.jp/crid/1574231873970603392
Nockolds and Allen, 1953: https://www.sciencedirect.com/science/article/pii/0016703753900556
Peacock, 1931: https://www.journals.uchicago.edu/doi/abs/10.1086/623788
Peccerillo and Taylor, 1976: https://link.springer.com/article/10.1007/bf00384745
Sheth et al., 2002: https://www.tandfonline.com/doi/epdf/10.2747/0020-6814.44.8.686?needAccess=true
I agree with your new definitions. Fe-enrichment at the early stages of evolution is now the accepted criterion and in my opinion addresses important petrogenitic processes. Sheth is correct in pointing out that using intermediate Fe enrichment to distinguish "calc-alkaline" leaves out "calc" and barely addresses "alkaline." Middlemost is also correct in suggesting using names that reflect the criteria (e.g., "high-Fe"). But, as you say, "inertia" has entrenched the terms tholeiitic and calc-alkaline and attempts to revise at this late date, however appropriate, seem to gain little traction. I believe we get into problems, however, when we attempt to ascribe a tholeiitic or calc-alkaline parent. Fe-enrichment depends more on conditions (e.g., H2O and fO2) than on parent.
Dear all: great discussion, it was about time many old terms should be updated. One real enigma is how to define anatectic leucogranitic rocks found in post-collisional settings. Are they calc-alkaline or not? Accordingly, C-A series are supposedly exclusive or volcanic arc settings, but those leucogranites certainly are neither tholeiitic nor alkaline! This is also a problem of the alphabet classification of granites, which we all know well. Are postcollisional and other anatectic granitoids type "A", "I" or "S", or Th, C-A, or A? These uncertainties should be clarified for good now that we are in the middle of the third decade of the XXI'st Century! Regards, Sebastian.
I hunt for porphyry coppers and have one being drilled so I have a horse in this race. I strongly recommend the petrologic terms somehow reflect the process that leads to mineralization. For example...the so called "adakite" term is supposed to indicate the source to have been influenced by garnet bearing material that influences the oxidation state of iron. It seems the garnet fractionation process might have an influence on iron...an important component of the so called tholeiitic series but how would it play out with petrologists. Similarly...hornblende fractionation is likely to play an important role in sourcing copper and gold...more so than the effect of garnet fractionation. I attach a relatively recent review and an image showing what I am getting at. Tholeiitic Basalt...is a pretty lame term if you ask me.
Hi Steve Johnson. What you write sounds interesting, but is out of the interest of IUGS. What the TGIR wants to produce here is an updated version of the book on the systematics of igneous rocks. Our main aim is to produce a series of recommendations to help young (but also senior) researchers to correctly classify igneous rocks, independently from their economic relevance or their ultimate origin. Petrology has to remain as much as possible out of the book, otherwise it would be certainly impossible to reach a consensus.
In few words: we want to provide the simplest tools to classify a rock as carbonatite or granite (and all the other variants in between). Once having correctly classified a rock in the proper way, each researcher is left free to propose his/her petrogenetic model.
Sorry to read that the term "tholeiitic basalt" is so insignificant for you. It is not so by the great majority of igneous petrologists.
Cheers,
michele
Dear Michele,
ZIMMER et al. 2010 (JOURNALOF PETROLOGY; doi:10.1093/petrology/egq062) summarized the shortcomings of AFM and FeO*/MgO-SiO₂ classification diagrams, highlighting the Fe enrichment and depletion trends in the early stages (MgO = 8-4 wt%) of magmatic evolution as key characteristics of tholeiitic and calc-alkaline series, respectively. This classification scheme is worth considering.
Cheers,
Jun
Dear Jung,
sorry for the delay in replying to your message. I simply forgot to send you a feedback on that. Thanks for evidencing the famous article of Zimmer et al. (2010). They propose a relatively simple index (defined as THI) to distinguish tholeiitic from calcalkalilne rock series. Despite its easy use, I see two-fold problems for our finalities:
1) Our idea is to propose the definition of "tholeiitic" or "calcalkaline" adjective for a single rock, not necessarily for a rock series. The THI parameter requires the knowledge of the FeO* content of a series of igneous rocks at 4 mass% MgO and at 8 mass% MgO. This means that the investigated rocks have to span at least this compositional range, and the THI parameter cannot be used for rocks with MgO >4 mass% or with MgO
Dear Michele,
The definitions are quite clear and emphasise the clear evolution of total iron enrichment in differentiating the series. My concern is with the more evolved terms of the series. As you wrote, in the AFM, the evolved tholeiitic and calc-alkaline terms greatly overlap. So how can we know the magmatic iron trend from the more acidic rocks alone? This question can be even further explored in the case where acidic rocks are thought to be of crustal origin: firstly, I have doubts that they can even be considered part of a series (no basic and intermediate terms); secondly, I don´t think we can fit crustal-derived rocks into the concept of iron enrichment and depletion series, often related to the initial stages of magma evolution. Maybe is worth to add some kind of note, similar to the one that was added for the tectonic setting, saying that for the more evolved rocks it may not be possible to discriminate between calc-alkaline and tholeiitic series or that this terminology can only be applied to mantle-derived rocks or that the iron enrichment and depletion can only be clearly seen up to and including intermediate compositions and dacitic(?) compositions.
Regards
João
Yes Joao, your reasoning is correct. We should take into consideration the possible derivation of rocks not from mantle but from crustal sources. In this case (or where a crustal component is acquired at shallow depths as consequence of AFC-type processes) the ultimate concept of magmatic series loose any significance.
Cheers,
Michele
For better or worse, most of us have come to accept the tholeiitic vs. calc-alkaline distinction as involving Fe enrichment during differentiation. Even if it loses any relationship to the calc-alkaline name. Applying the terms should then be restricted to looking at a suite of reasonably consanguinious rocks from a single area that follow a protracted trend where the presence or lack of Fe enrichment can be evaluated and not to any single samples. Focusing on samples restricted to the primitive end or the evolved end (such as crustal rocks), where the trends converge, is of course problematic for several reasons.
- Badges
- Science method
- Similar topics
- Biological Science
- Ecology