Refocus: The Hotspot-Plume Debate

[Incomplete draft, 2003]
In recent years the ongoing hotspot-plume debate has seemingly increased in intensity, extending beyond refereed publications to sometimes passionate exchanges in opinion and letters sections of organization newsletters and even the popular scientific press. Some of the debate is reminiscent of a political exchange. Assertions are sometimes imprecise or misleading, while counter assertions ignore the initial assertion, providing a response that raises entirely new and different issues. Like ships passing in the night, the debate seems to involve much more miss than hit. Rather than continue this sequence of mixed metaphors, it is preferable to try to refocus the debate. What are the fixed hotspot [1,2] and plume [2,3] hypotheses (while coupled, they can be viewed as separate proposals – either complementary or even competing)? What evidence have we to consider in elaborating and testing the hypotheses? Can we perhaps come to a minimal consensus on these basics, as a basis for progress in future research? This note is an attempt to bring some clarity to the debate while introducing some pertinent observations from both recently published research and earlier, yellowing publications.
1. Hotspots: Some workers have a problem with the term, insofar as it implies anomalously higher temperatures at some depth beneath the lithosphere. Anomalous magmatic centers exist and, therefore, at lithospheric depths, at least, they could be called “hotspots”. For the purpose of this discussion, “hotspot” refers to the source location of a zone or trend of anomalous magmatism.
2. Traces: There is evidence that most inferred hotspots have associated traces that in many (if not all) cases show time-transgressive inception of magmatic activity. [4]
3. Many inferred hotspot traces show recurrent, intermittent volcanic activity after inception [4,5] (e.g., Austral-Cook, Pratt-Welker, and especially the Line Islands). In some cases the recurrence could be interpreted as semi-continuous after initiation (e.g., Yellowstone and Raton). [4,5,6]
4. Plate kinematic/reconstruction models can be constructed which provide moderately impressive fits to the (a) Atlantic-Indian Ocean (“Tristan” set) hotspots and traces7 and, separately, (b) the hotspots and traces (“Hawaiian” set) of the Pacific Ocean [8,9,10]. The two models are not consistent with each other, [4,9,10] however, nor do either fit the Iceland hotspot [4,11,12] and traces, Svalbard, [4] or the putative Eifel trace [4] (collectively, the “Icelandic” set).
5. Three volcanic provinces (East Africa Rift, Eastern Australian Highlands, eastern Brazil) show latitudinal patterns of volcanic migration consistent with hotspot kinematic models, but not along single linear trends.4 Several parallel trends of central volcanics in the Australian example show a propagation rate consistent with either the Tristan or Hawaiian hotspot model. [4,13] The southernmost limit of volcanics in East Africa migrates to the south at the same rate that Africa is inferred to be moving to the north in the Tristan hotspot reference frame4. A second cluster of volcanism in East Africa could be similarly interpreted. [4] Cretaceous to early Cenozoic magmatism in eastern Brazil shows an easterly migration in inception correlative with South American plate motion in the Tristan hotspot frame.
Let’s rethink the hotspot/mantle plume hypothesis. For Wilson, [1] the Hawaiian Islands were evidence for movement of the Pacific crust over a “hotspot” at some undefined depth within the mantle. In the pre-plate tectonic environment of the early Sixties, his interpretation lent additional support to developing ideas of continental drift and seafloor spreading.
Morgan’s [2,3] innovation was two-fold: (1) Hotspots in the mantle formed a fixed (“absolute”) reference frame and (2) the hotspots were a manifestation of deep mantle plumes.
It is easy to overlook the conceptual linkage of the hotspot reference frame and plumes. As part of his elaboration of the two-fold hypothesis, Morgan presented an instantaneous contemporary global model of plate kinematics relative to one another and relative to the hotspots. If such a reference frame existed for more than just the present, a critical paradox emerges. With the elaboration of plate tectonics in the late Sixties, especially the depth extent of subduction zones and the implications of an enlarging Atlantic Ocean at the expense of the plates of the Pacific, that part of the mantle below the lithosphere “entrapped” by subduction zones (from below the asthenosphere to the deepest extent of subduction would have difficulty in maintaining a stable global reference frame. That is, shrinkage of the Pacific Ocean basin requires displacement of shallow mesosphere relative to shallow mesosphere beneath the surrounding continental plates. Subducting slabs form a “wall” that inhibits flow of mesosphere from one side to the other. [14] Therefore, one solution would be to localize the hotspot sources at greater depths than the deepest subduction zones, and allow for rapid vertical transport of “fertile” hotspot source material – thus plumes. This rationale is not at all explicit in Morgan’s formulations, but, it could be argued that it is implicit.
Somehow, the fixed hotspot reference frame as a rationale for plumes has been decoupled from the plume hypothesis. There are two apparent reasons for this. Observation 3 above makes clear that a single hotspot reference frame doesn’t exist. Two reference frames, the Hawaiian and Tristan, appear to be moderately well-established, for the past 80 to 130 m.y., respectively. A third, beneath Iceland is an additional possibility. Interestingly, subduction zones partially separate the three possible reference frames from each other. The neo-classical rationale for plumes, discussed above, has evaporated.
Ironically, numerical and physical plume modeling in the context of whole or partial mantle convection has arrived at a complementary way-station to the implications of the multiple hotspot reference frames. Such models have been unable to produce a globally (or, in some cases, even regionally) consistent fixed hotspot reference frame. The models require either shallow counter-flow or plate-motion-parallel shear of the upper asthenosphere and mesosphere. Thus, some plume advocates how state that moving hotspots (i.e., no global or even local reference frame) are to be expected as a consequence of plume convection.
Nevertheless, and contrary to mantle plume modeling results, the two documented hotspot reference frames appear to exist. So, how good are they? It’s one thing to show a map with data locations and loci, but more is really desirable. Distance versus age plots help, except where there are distinct kinematic changes. Additional visualization tools are desirable.
Herein, a reduced version of Wessel and Kroenke’s [15] hotspotting is applied. Data points along traces are rotated back to their calculated location at their dated age using available hotspot reconstruction models. If the hotspot models were correct and the data points reflecting the age of plate-hotspot encounter, such reconstructed data points would all overlie the inferred hotspot. However, even in the context of the hotspot model, two additional effects are to be anticipated: delayed magmatism within the moving plate and displacement of magmatism from the hotspot due to lithospheric heterogeneities.
The combined deviations should result in elliptical clusters of restored points, with the long axis of the ellipse oriented in the direction of plate motion at the time of magmatic emplacement, while the short axis would represent a combination of deviation from the fixed-hotspot reference hypothesis and errors in the kinematic model itself. The hotspot would be expected to occur at the younger end of the ellipse.
Some traces show an elliptical clustering pattern. These include the Hawaiian-Emperor, Louisville, Tuamotu-Easter-Nazca (mirror-image traces; Pilger and Handschumacher, 1981), Cobb, Society, Foundation, Pitcairn-Gambier, and Juan Fernandez of the Hawaiian set, and the Tristan-Walvis-Rio Grande (another mirror image), Kerguelen-Ninetyeast, Reunion-Chagos-Laccadive, Great Meteor-New England, Canary, and Tasman of the Tristan set. In some cases, however, the present-day inferred hotspot location is displaced from younger end of the elliptical cluster (e.g., Great Meteor). And, for some traces, there are a few outliers on the younger side (implying the data point is “too old” for the model), including Louisville, Reunion, Kerguelen, In some cases, the “outliers” correspond with dates that Baksi [1998] has questioned, based on statistical analysis arguments (especially Great Meteor, Reunion and Kerguelen).
Three volcanic provinces show a remarkable pattern of partial correspondence with the hotspot model. The East Africa Rift and eastern Brazil volcanics show a migration pattern that parallels that of the model, but not along a single locus. A distinct trend in inception of magmatism from north to south is apparent in latitude versus age plots (Figure aa) for East Africa, and from west to east (longitude versus age) for eastern Brazil – two trends, corresponding generally with the Tristan and Trinidade hotspots (Figure bb).
1. Wilson, 1963

2. Morgan, 1971

3. Morgan, 1972

4. Pilger, 2003

5. Pilger and Handschumacher, 1981

6. Suppe et al, 1973

7. Muller et al, 1993

8. Harada and Hamano, 2000

9. Norton, 2000

10. Raymond et al., 2000

11. Molnar and Atwater, 1973

12. Molnar and Francheteau, 1975

13. Pilger, 1982

14. Alvarez, W., 1982, Geological evidence for the geographical pattern of mantle return flow and the driving mechanism of plate tectonics: Journal of Geophysical Research, v. 87, p. 6697-6710, 1982.

15. Wessel, P., and L. W. Kroenke, 1997, A geometric technique for relocating hotspots and refining absolute plate motions, Nature, 387, 365-369

Rex H. Pilger, Jr.
Landmark Graphics Corporation
1805 Shea Center Drive
Suite 400
Highlands Ranch, Colorado 80129 USA

September 16, 2003 (minor corrections 1/29/2011)