Flora of the Korean Peninsula

Chin S. Chang and Hui Kim

This project includes specimens conserved in European, American and Japanese herbaria, which this paper identified via the following abbreviations proposed by Index Herbariorum (Thiers 2019): (1) E for the Edinburgh Botanical Garden, (2) P for the Paris Museum Herbarium, (3) LE for the Komarov Botanical Institute of Russian Academy of Science Russia, Saint Petersburg, (4) A for the Harvard University Herbaria, (5) KYO for the Kyoto University Herbarium, (6) TI for the University of Tokyo Herbarium and (7) SNUA for the T.B. Lee Herbarium at Seoul National University. The E (3,017) and TI (1,002) herbaria constitute the largest collections of Taquet, while the KYO (2,714), E (1,475) and P (869) conserve the major collections of Faurie. This is the first attempt by Korean researchers to investigate specimens deposited at various foreign herbaria using a single and uniform protocol. We have visited TI, A, E and KYO, taken photos and recorded them in the database. We have also searched for additional specimens at P, LE and K either from botanical collection papers (Grabovskaya-Borodina et al. 2018) or herbarium websites (Royal Botanic Gardens, Kew 2020, Chagnoux 2020).

This research was supported by a BIFA (the Biodiversity Information Fund for Asia) funded by the Ministry of the Environment (BIFA3_14), Government of Japan.

Since 1945, the Korean Peninsula has been divided into what are now two countries: North Korea (Democratic People’s Republic of Korea) and South Korea (Republic of Korea). In terms of botanical importance, its notable islands include Jeju Island (Quelpaert) and Ulleung Island (Ulleungdo). Korea’s vascular flora includes 4,831 taxa (Chang et al. 2014) constituting a relevant portion of Eastern Asian flora. There are also plants from three principal biogeographic regions (Chang et al. 2017): (a) the Amur flora, characterised by cold temperate forests and shrubs; (b) the Northern China flora, characterised by deciduous temperate forests; and (c) the China–Japan–Korea (CJK) flora, characterised by warm temperate forests with evergreen forest taxa.

E. H. Wilson collected 1,200 plant records representing 51 families that he identified with Alfred Rehder (Briggs 1993). These data describe a specimen dataset of Korean Peninsula woody plants preserved at the Harvard University Herbaria (A). E. H. Wilson visited Korea twice: first in 1917 to Oo-rong-do Island (Ulleung or Degelet Island), Quelpaert Island, Mt. Chiri-san, Pingyang, Keijyo (Seoul), Koryo, north-eastern Korea and Mt. Konggo-san and second, in 1918, he revisited the same provinces. Many books and articles on this collector have been published; however, his botanical itineraries have not been well documented (Kim et al. 2010). Meanwhile, Fr. U. J. Faurie made three extended collecting trips to Korea: the central region in 1901, the central and southern regions in 1906 and Quelpaert Island in 1907 (Hayata 1916, Kakuta 1992, Chang et al. 2004). Fr. E. Taquet, who stayed on Quelpaert Island as a Catholic missionary from 1902 to 1915 (Chang et al. 2015), made extensive collections of vascular plants from the Island (Jeju) from 1907 to 1914.

Both Faurie and Taquet did not number their collections chronologically, based on their collecting activities. They seem to have sorted the collections by genera and they assigned numbers to the taxonomic bundles of dried plants. Some of the collection data, such as locality, date or collection number, were missing. The first set of specimens is at E or P, except for some families. Duplicate specimens were widely distributed and could be found at BM, TI, KYO, A, LE and B. Faurie’s collection of several thousand herbarium specimens is deposited in Paris, with duplicates at the University of Kyoto, the British Museum, Kew and elsewhere (Kitagawa 1979, Koidzumi 1936).

Georeferencing: A wide range of historically used toponyms in Korea have Chinese-character origins and can, therefore, be written the same way (Choo 2016, Tanabe and Watanabe 2014). As a result of 36 years of Japanese colonial occupation, Korean place names used for plant collections have become a toponymic enigma. In many Asian countries, Japanese exonyms are names of places in the Japanese language that differ from those given in their dominant language. Japanese botanists or field guides often transliterated these toponyms into the Japanese pronunciation. This has produced many unresolved botanical exonyms, which have been only found on herbarium labels. These Japanese terms for some place names are now a mystery either because they are quite different from endonyms or because of some other obscure etymology. We have prepared a multilingual gazetteer to resolve the inconsistencies, uncertainties and confusion on botanical exonyms in the Korean Peninsula that foreign explorers and botanical collectors in Korea have used over the past 120 years (Table 1, Chang et al. 2015).

After the identification of place names, the next step is providing a precise coordination to a biological collection. We always aimed for accurate georeferencing for location coordinates, but sometimes this was not possible because of insufficient information in the place names. Thus, in these situations, we used higher geographic area coordinates, such as counties or cities. To minimise errors, enhance data consistency and maintain integrity throughout the georeferencing process, we modified a procedure adopted by the Chinese type collection project (Fig. 1, Lohonya et al. 2020).

Figure 1.  

Flow diagram on how to approach labels in different languages and endonyms as well as exonyms.

Using the BRAHMS system, we set up a database of herbarium records. We compared the geographic queries with the label information for each specimen to resolve geographic information. We detected and corrected two types of errors: typographical errors and erroneously identified records. After updating the database with recent publications and cleaning the data, we obtained the clear collection data that corresponds to this dataset.

Finally, we generated the Darwin Core Archive to incorporate the metadata in this file and published the data on GBIF, using the Integrated Publishing Toolkit.

The Korean Peninsula is located in northeast Asia, between China and Japan. To the northwest, the Amnok River separates Korea from Liaoning Province in northern China and to the northeast, the Duman River separates Korea from Jirin Province in northern China and Far Eastern Russia. Excluding the islands, the Peninsula area covers about 220,847 km². The eastern and northern parts of the Peninsula are characterised by the high mountains. The highest point of the Korean Peninsula is located at Mount Paektu (2,744 m a.s.l.; 41°59N; 128°04E) and stands on the border with China (Fig. 2). The southern area of the Peninsula begins at the Island Marado (33°06N; 126°16E)) at the south of Jeju Island and stretches in an eastwards direction to the islets of Dokdo (37°14N; 131°52E).

Figure 2.  

The total occurrence points of vascular plant specimens were collected by three collectors in the Korean Peninsula. Enlarged maps are shown for the Mt. Konggo-san and Quelpaert, where Wilson, Faurie and Taquet made extensive collections.

33°06'45.0"N and 43°00'42.3"N Latitude; 124°13'22.8"E and 131°52'22.5"E Longitude.

The majority of specimens belong to class Magnoliopsida (6,314 specimens) and Liliopsida (2,198), followed by Filicopsida (765), Lycopodiopsida (33), Equisetopsida (4), Coniferophyta (144) and Psilopsida (1). Our dataset represents 165 families (Fig. 3), of which 16.8% and 11.4% of the specimens belong to the monocot families Poaceae and Cyperaceae and the dicot families Rosaceae and Asteraceae, respectively, followed by Fabaceae (3.9%), Ranunculaceae (3.1%), Apiaceae (2.5%), Dryopteridaceae (2.5%), Polygonaceae (2.3%), Lamiaceae (2.2%), Liliaceae (2.2%), Caprifoliaceae (2.1%) and Fagaceae (1.9%). It further includes 755 genera, with the significant ones being Carex (543), Quercus (151), Persicaria (147), Dryopteris (133), Rubus (116), Prunus (107), Euonymus (100), Salix (91), Acer (91), Viola (86), Thelypteris (80), Vicia (79), Clematis (78), Aster (77), Viburnum (77), Lonicera (77), Ranunculus (73), Lespedeza (70), Fimbristylis (66), Cyperus (63), Elaeagnus (59), Athyrium (56), Adenophora (55) and Setaria (54).

Figure 3.  

Taxonomic distribution of occurrences amongst vascular plant families in the dataset. The figure was prepared with the “treemap” package in R (Tennekes 2014).