Tertiary Fossil Floras of Alaskan National Parks

Theodore Matel, NPS–Paleontological Society Paleontology in the Parks fellow
PhD student, University of Michigan

Introduction

The Paleontology Program of the National Park Service is dedicated to developing a baseline understanding of the natural history of the public lands of the United States and places the NPS at the forefront of paleontological research and resource management. NPS paleontological resource inventory reports, together with field reconnaissance missions, enable the discovery of new localities and rediscovery of historical localities, as well as establish a robust context for the study, interpretation, and management of fossil resources (Santucci et al. 2018).

The NPS paleontological resource inventory program has been particularly fruitful in the Alaska region. The national parks of Alaska comprise 54 million acres, approximately two-thirds of the total acreage managed by the NPS (Figure 1). Alaskan parks contain an astonishing diversity of fossils that span nearly the entire geologic column. One notable area of richness amidst the temporal and taxonomic breadth of these resources is the Cenozoic paleobotanical record (Figure 2). Plant fossils from this era have been known from Alaska for more than 150 years and have been studied by some of the most famous paleobotanists of the previous two centuries. Tertiary plant macrofossils (Paleocene through Pliocene, 66 to 2.58 million years ago), the subject of this article, are known from six Alaskan National Parks (Figure 1, Parks 7, 8, 10, 11, 13, and 14) and were recovered from 14 different formations spanning approximately 60 million years (Figure 2). The story told by these fossils is fundamental to the disciplines of paleobotany and plant geography.

^{Figure 1. Map of Alaskan National Park Units. 1. Cape Krusenstern National Monument (CAKR); 2. Noatak National Preserve (NOAT); 3. Kobuk Valley National Park (KOVA); 4. Gates of the Arctic National Park (GAAR); 5. Bering Land Bridge National Preserve (BELA); 6. Yukon-Charley Rivers National Preserve (YUCH); 7. Denali National Park and Preserve (DENA); 8. Lake Clark National Park and Preserve (LACL); 9. Alagnak Wild River (ALAG); 10. Katmai National Park and Preserve (KATM); 11. Aniakchak National Monument and Preserve (ANIA); 12. Kenai Fjords National Park (KEFJ); 13. Wrangell-St. Elias National Park and Preserve (WRST); 14. Glacier Bay National Park and Preserve (GLBA); 15. Klondike Gold Rush National Historical Park (KLGO); 16. Sitka National Historical Park (SITK).}

^{Figure 2. This composite stratigraphic column illustrates the chronological relationship between the paleobotanically productive formations discussed in the text. Older formations are at the bottom and younger formations are at the top. The temporal span of formations reflects minimum and maximum age estimates for outcrops within their respective parks. The formations discussed in the text have yielded plant fossils from only a small subset of the potential time interval illustrated in the figure. Fm. = Formation.}

History of paleobotanical investigation in Alaska

The first fossil plant specimens from Alaska to be figured in a publication were studied by the Swiss paleobotanist Oswald Heer (1809–1883). This collection was made along the shores of Cook Inlet and on the Kenai and Alaskan peninsulas. In the publication Flora Fossilis Alaskana (1869), Heer concluded that the flora included 50 different taxa, ten of which were new to science. He placed the fossil plant assemblage collected at Coal Cove on the Kenai Peninsula in the Miocene, the age which it is understood to be today. However, he also related the Coal Cove flora to an assemblage now understood as Paleocene in age from Atanikerdluk, Greenland. The Coal Cove flora was recollected in the 1960s and 1970s by paleobotanist Jack Wolfe (1936–2005).

Later paleobotanists questioned Heer’s age assignment of the Alaskan plant fossils. Frank Hall Knowlton (1860–1926) described a fossil assemblage from Kukak Bay in 1904. This assemblage was collected by De Alton Saunders during the Harriman Expedition, an expedition funded by E. H. Harriman (1848–1909) that took place in June and July of 1899. Kukak Bay is now within the boundary of Katmai National Park and Preserve on the Alaskan Peninsula (Figure 3, point 3).

^{Figure 3. Map of Alaskan National Park Units and significant Tertiary plant fossil localities. 1. Aniakchak Crater, Tolstoi Fm. (ANIA); 2. Dumpling Mountain, Copper Lake Fm. (KATM); 3. Kukak Bay, Hemlock Conglomerate (KATM); 4. Chinitna Bay, West Foreland Fm. (LACL); 5. Redoubt Point, Tyonek Formation (LACL); 6. Seldovia Point, Kenai Group (non-NPS); 7. Frederika Glacier, Frederika Fm. (WRST); 8. Yakutat Bay, Kulthieth Fm. (WRST); 9. Cenotaph Island, Yakataga Fm. (GLBA); 10. Dunkie Coal Mine, Healy Creek Formation (DENA).}

Knowlton identified 26 taxa from this assemblage, nine of which represented new species, and referred the flora to the “Arctic Miocene” (Knowlton 1904). Knowlton and other paleobotanists of the era considered plants from the “Arctic Miocene” to be late Eocene in age based on the erroneous assumption that high latitude Eocene floras should resemble temperate Miocene floras of middle latitudes because plants had migrated equatorward in response to climatic cooling between the late Eocene and Miocene (Knowlton 1894).

In the 1930s, the paleobotanist Arthur Hollick (1857­–1933) collected from new localities in the Yukon Valley and published comprehensive systematic papers on the Cretaceous and Tertiary floras of Alaska based on collections housed at the Smithsonian Natural History Museum (Hollick 1930, 1936). Hollick identified more than 300 species from the Tertiary of Alaska and noted similarities between the Alaskan floras and floras of Eocene, Oligocene, and Miocene age in western North America. However, the collections Hollick studied lacked the detailed stratigraphic records necessary to recognize different temporal epochs. As a result, he attributed all the specimens to a single formation—the Kenai Formation—and explained the variable age distribution of these “Kenai” taxa in western North American fossil floras by giving the assemblage a composite late Eocene age. This became the accepted age interpretation of the floras of Cook Inlet and the Alaskan and Kenai peninsulas until Jack Wolfe’s revision of Alaskan fossil floras in the 1960s and 1970s (Wolfe et al. 1966; Wolfe 1977). The erroneous conclusion that the diverse fossil assemblage of this region was essentially isochronous was synthesized into the concept of the Arcto-Tertiary Geoflora.

Arcto-Tertiary Geoflora

The term Arcto-Tertiary flora implies an origin area (Arctic) and time (Tertiary) of a type of fossil flora. The term was first coined by Heinrich Gustav Adolf Engler (1844–1930) in the 19^th century to describe the living vegetation of conifers and broad-leaved deciduous angiosperms that occurs in temperate areas of North America, eastern Asia, and Europe. Some well-known members of this broad category of plants include Abies (firs), Acer (maples), Carya (hickories), Cornus (dogwoods), and Picea (spruces). The paleobotanist Ralph Works Chaney (1890–1971) recognized a strong coniferous and temperate deciduous component to Oligocene and Miocene floras of western North America and suggested that an Arcto-Tertiary Geoflora had maintained itself over epochs with minor changes in composition despite significant alterations to its distribution (Chaney 1959). Chaney also leveraged the erroneous conclusions about the stratigraphic and taxonomic affinities of Alaskan Tertiary floras made during the early 20^th century to identify a high-latitude source area (the Arctic Eocene) for the Oligocene and Miocene floras of middle latitudes in North America. Because Alaska was periodically connected to eastern Asia by the Bering Land Bridge during the Cenozoic, Eocene Alaskan forests were a plausible source area for the modern East Asian Arcto-Tertiary assemblages. Thus, the traditional Arcto-Tertiary Geoflora concept was devised to explain the similarity between modern forests of eastern Asia and eastern North America by identifying their common origin in Eocene forests of Beringia.

Modern revisions to the Tertiary flora of Alaska

In the latter half of the 20^th century, the paleobotanist Jack Wolfe revised the Tertiary fossil floras of Alaska by collecting new specimens, studying historical collections, and applying new methodology. Wolfe was the first paleobotanist to place fossil localities from the Alaskan and Kenai peninsulas, Cook Inlet, and the Gulf of Alaska into a single stratigraphic framework. This framework revealed that the homogenous “Kenai flora”, formerly interpreted as Eocene in age, represented fossil deposits from the early Paleocene to late Miocene, a time interval spanning approximately 60 million years. Furthermore, his work showed that the “Kenai flora” was far from uniform; it includes taxonomically and morphologically diverse plant communities representative of several different forest types and regional floras.

In addition to his meticulous fossil collecting, the methodology by which Wolfe analyzed fossil plant communities played an important role in the recognition of this diversity. Wolfe was one of the earliest proponents of analyzing fossil floras on the basis of both their floristic composition and physiognomic composition. Floristic analysis of a fossil flora involves identifying fossils to the lowest taxonomic level possible and studying the distribution of their nearest living and fossil relatives. Then, by leveraging the assumption that the climatic affinities of a group have remained constant over time, inferences can be made about the climatic and ecological setting of the fossil plant community. Physiognomic analysis of a fossil flora involves evaluating the proportion of taxa within an assemblage which display a certain characteristic that can be linked to climate (e.g., the proportion of taxa with toothed leaves is inversely proportional to mean annual temperature). Then, by comparison of these proportions to those measured in modern forests, inferences can be made about the prevailing climatic conditions of the fossil assemblage and vegetation type it represents. By 1966, Wolfe had gathered enough floristic, vegetational, and stratigraphic evidence to conclude that: “Considered in the framework of the floristic succession in Alaska, the concept of an ‘Arcto-Tertiary Geoflora’ does not appear to be valid” (Wolfe 1966).

In place of the Arcto-Tertiary Geoflora concept, Wolfe proposed that the Tertiary Alaskan fossil floras represented two primary forest types and gradations between the two (Wolfe 1972). The predominant forest type of the early Paleogene was a paratropical rainforest (a rainforest not in the tropics), characterized by abundant lianas (woody climbing plants), species with entire-margined (not toothed) leaves, and many taxa with nearest living relatives restricted to tropical latitudes (Wolfe 1977). The predominant forest type of the late Paleogene and Neogene was a mixed northern hardwood forest, characterized by abundant and diverse broad-leafed deciduous taxa, and a taxonomic composition similar to modern forests at temperate latitudes in east Asia and eastern North America (Wolfe 1966; Wolfe and Tanai 1980). The next sections are devoted to celebrating the contribution of fossils collected from Alaskan National Park lands to this theory.

The Paleogene record: Paratropical rainforests

Many of the Paleogene plant fossil localities which formed the primary body of evidence for Wolfe’s paratropical rainforest came from the Yakutat block in the Gulf of Alaska region. The oldest leaf assemblages from this region are derived from early Eocene deposits of the Kulthieth Formation and include a locality near Yakutat Bay (Figure 3, point 8), within Wrangell-St. Elias National Park and Preserve (WRST). Climate reconstructions based on leaf physiognomy estimate the Kulthieth assemblage as the warmest of any Alaskan Paleogene flora, with a mean annual temperature of 19.4°C (66.9°F) (Wolfe 1994a). Floristic studies indicate that the Kulthieth assemblage is also the most diverse in number of species. Taxa from the WRST locality include leaves of tropical plant families such as Annonaceae (custard apple family), and holotype specimens of plants in the Menispermaceae (moonseed family; Figure 4A) and Icacinaceae (Figure 4D), which probably grew as lianas (Wolfe 1977).

Evidence from a leaf assemblage collected at Aniakchak Crater (Figure 3, point 1) in Aniakchak National Park and Preserve (ANIA) on the Alaskan Peninsula suggests that the climate had cooled somewhat by the middle Eocene and that a greater number of taxa now characteristic of temperate deciduous forests occurred in the region (Wolfe 1994a). The Aniakchak Crater assemblage is derived from the Tolstoi Formation and includes 34 species. Characteristic genera of this assemblage include some that still occur in modern temperate forests of North America, such as Quercus (oaks), Ulmus (elms), Cornus (dogwoods), and Acer (maples). Others are restricted to Asia today, such as Cercidiphyllum (katsura tree) and Pterocarya (Chinese wingnut), although these taxa are commonly cultivated in temperate climes of North America.

The climatic deterioration which occurred at the Eocene–Oligocene boundary had a visible effect on the fossil floras of Alaska. An assemblage from the Tyonek Formation at Redoubt Point (Figure 3, point 3), a locality within Lake Clark National Park and Preserve (LACL), produced the lowest estimate of mean annual temperature in Paleogene Alaskan floras based on leaf physiognomy, 10°C (18°F) cooler than the early Eocene Kulthieth assemblage and 5°C (9°F) cooler than the middle Eocene Aniakchak Crater assemblage. The Redoubt Point flora is also significantly less diverse—representing only 17 different species—and characterized by new dominant taxa, Alnus evidens (an extinct alder species; Figure 4B) and Metasequoia (dawn redwood). Fossil localities of a similar age also occur within the Hemlock Conglomerate at Kukak Bay in Katmai National Park and Preserve (KATM) (Figure 4C), and in the Poul Creek Formation in WRST (Wolfe 1966, 1977). The similarity in age of these three geographically disparate localities is partially attested by the ubiquity of Alnus evidens, a taxon whose holotype was collected at Kukak Bay in KATM (Hollick 1936; Wolfe 1966).

The Neogene record: Mixed northern hardwood forest

The mixed northern hardwood forest is best represented by paleofloras deposited during a warm period of the middle Miocene (Wolfe 1994b). In the biostratigraphic framework of Alaska, Wolfe termed this period the Seldovian, for the rich Seldovian Point flora (Wolfe and Tanai 1980) which occurs on the Kenai Peninsula approximately 40 km (25 mi) west of the southwest corner of Kenai Fjords National Park (Figure 3, point 6). This assemblage is the most diverse post-Eocene flora in Alaska and includes 45 species. It has taxonomic richness in plant families which are now characteristic of temperate deciduous forests such as Salicaceae (willow family), Juglandaceae (walnut family), Betulaceae (birch family), Fagaceae (beech family; Figure 4E), Ulmaceae (elm family), and Aceraceae (maple family).

^{Figure 4. Holotype and other specimens recovered from Tertiary Alaskan floras. A. Paratinomiscium conditionalis (Hollick) Wolfe (USNM 39117), recovered from the Yakutat Bay locality in WRST. Hollick initially placed this specimen in the genus Hampea (Malvaceae, mallow family), but it was revised by Wolfe and placed in Menispermaceae (moonseed family). Photo from Hollick (1936) (Plate 118, figure 1). B. Alnus evidens (Hollick) Wolfe (USNM 38844), recovered from the Kukak Bay locality in KATM. This taxon also underwent taxonomic revision when studied by Wolfe; it was initially placed in the genus Corylus (hazels). Photo from Hollick (1936) (Plate 49, figure 3). C. Populus congerminalis Hollick (USNM 39166), recovered from Kukak Bay in KATM. Photo from Hollick (1936) (Plate 116, figure 2). D. Paleophytocrene elytraeformis (Hollick) Wolfe (USNM 39127A, B), recovered from the Yakutat Bay locality in WRST. This holotype represents impression and compression of the distinctive endocarp (fruit interior) of a member of the Icacinaceae. Photo from Hollick (1936) (Plate 120, figures 3 and 4). E. Fagus antipofi Heer. Hypotype specimen of an extinct species of beech tree collected at Seldovia Point by Wolfe. Photo from Wolfe and Tanai (1980) (Plate 6, figure 4).}

Plant fossil localities representative of the Seldovian stage occur in Glacier Bay National Park (GLBA), in sediments of the Yakataga Formation on Cenotaph Island (Figure 3, point 9), and in WRST, in outcrops of the Frederika Formation along Skolai Creek (Figure 3, point 7). The Skolai Creek assemblage is notable because of the prevalence of fossil needles and seeds with affinity to the Pinaceae (pine family) compared to the lack of Pinaceae macrofossils at other Seldovian localities. An untested hypothesis explaining this difference is that the Skolai Creek locality in WRST was deposited at a higher altitude than the coastal Seldovia Point locality, and that the regional flora varied along an altitudinal gradient.

Conclusion

The National Parks of Alaska hold outstanding paleobotanical resources, and despite having been studied for more than a century, new discoveries remain to be made. As recently as 2010 a new plant fossil locality (Figure 3, point 2) was located in outcrops of the early Eocene Copper Lake Formation in the northwest region of Katmai National Park and Preserve (Parrish et al. 2010). The Tolstoi and West Foreland Formations of a similar age have produced plant fossils within Aniakchak and Lake Clark National Parks, respectively. Increased sampling efforts at localities of these formations within and outside national park lands will increase the geographic range of fossil localities from this time interval, which, in Alaska, is best known from the Gulf of Alaska region. Furthermore, recent discoveries of fossil plants from Upper Cretaceous deposits in Yukon-Charley National Preserve (Fiorillo et al. 2014) and Denali National Park and Preserve (Tomisch et al. 2010) offer potential to link the Tertiary Alaskan floras with earlier plant fossil assemblages of the region. The continued study of these resources is likely to yield well-preserved fossils which will enhance our knowledge of Alaskan fossil floras, a record of natural history which has fascinated scientists for more than 150 years and is just beginning to be understood.

References

• Chaney, R. W. 1959. Miocene floras of the Columbia Plateau. Part I. Composition and interpretation. Carnegie Institute of Washington Contributions to Paleontology 617: 1–134.

• Fiorillo, A. R., F. Fanti, C. Hults, and S. T. Hasiotis. 2014. New ichnological, paleobotanical, and detrital zircon data from an unnamed rock unit in Yukon-Charley Rivers National Preserve (Cretaceous: Alaska): stratigraphic implications for the region. Palaios 29: 16–26.

• Heer, O. 1869. Flora fossilis Alaskana. Kongl. Svenska Vetenskaps Akademiens Handlinger 8.

• Hollick, A. 1930. The Upper Cretaceous floras of Alaska. U.S. Geological Survey, Washington, D.C. Professional Paper 159.

• Hollick, A. 1936. The Tertiary floras of Alaska. U.S. Geological Survey, Washington, D.C. Professional Paper 182.

• Knowlton, F. H. 1894. A review of the fossil flora of Alaska, with descriptions of new species. Proceedings of the United States National Museum 17: 207–240.

• Knowlton, F. H. 1904. Fossil plants from Kukak Bay. 1904. Pages 149–162 in B. K. Emerson, C. Palache, W. H. Dall, E. O. Ulrich, and F. H. Knowlton. Alaska. Volume IV: Geology and paleontology. Doubleday, Page & Company, New York, New York.

• Parrish, J. T., A. R. Fiorillo, B. F. Jacobs, E. D. Currano, and E. A. Wheeler. 2010. The Ketavik Formation: A new stratigraphic unit and its implications for the Paleogene paleogeography and paleoclimate of southwestern Alaska. Palaeogeography, Palaeoclimatology, Palaeoecology 295: 348–362.

• Santucci, V. L., J. S. Tweet, and T. B. Connors. 2018. The paleontology synthesis project and establishing a framework for managing National Park Service paleontological resource archives and data. New Mexico Museum of Natural History and Science Bulletin 79: 589–599.

• Tomsich, C. S., P. J. McCarthy, S. J. Fowell, and D. Sunderlin. 2010. Paleofloristic and paleoenvironmental information from a Late Cretaceous (Maastrichtian) flora of the lower Cantwell Formation near Sable Mountain, Denali National Park, Alaska. Palaeogeography, Palaeoclimatology, Palaeoecology 295: 389–408.

• Wolfe, J. A. 1966. Tertiary plants from the Cook Inlet region, Alaska. U.S. Geological Survey, Washington, D.C. Professional Paper 398-B.

• Wolfe, J. A. 1972. An interpretation of Alaskan Tertiary floras. Pages 201–233 in A. Graham, editor. Floristics and paleofloristics of Asia and eastern North America. Elsevier, Amsterdam, Netherlands.

• Wolfe, J. A. 1977. Paleogene floras from the Gulf of Alaska region. U.S. Geological Survey, Washington, D.C. Professional Paper 997.

• Wolfe, J. A. 1994a. Alaskan Palaeogene climates as inferred from the CLAMP database. Pages 223–237 in M. C. Boulter and H. C. Fisher, editors. Cenozoic plants and climates of the Arctic. Springer-Verlag, Berlin, Germany. NATO ASI Series 127.

• Wolfe, J. A. 1994b. An analysis of Neogene climates in Beringia. Paleogeography, Paleoclimatology, Paleoecology 108: 207–216.

• Wolfe, J. A., and T. Tanai. 1980. The Miocene Seldovia Point Flora from the Kenai Group, Alaska. U.S. Geological Survey, Washington D. C. Survey Professional Paper 1105.

• Wolfe, J. A., D. M. Hopkins, and E. B. Leopold. 1966. Tertiary stratigraphy and paleobotany of the Cook Inlet region, Alaska. U.S. Geological Survey, Washington D. C. Professional Paper 398-A.

Fossils and Paleontology