Received: 1 April 2023 | Revised: 16 August 2023 | Accepted: 17 August 2023 DOI: 10.1111/jbi.14712 C O M M E N TA R Y Settlement and anthropization of the Azores Islands Valentí Rull1,2 1 Botanic Institute of Barcelona, Spanish National Research Council (CSIC), Barcelona, Spain 2 Institut Català de Paleontologia Miquel Crusafont, Universitat Autònoma de Barcelona, Cerdanyola del Vallès (Barcelona), Spain Correspondence Valentí Rull, Botanic Institute of Barcelona, Spanish National Research Council (CSIC), Pg. Migdia s/n, 08038 Barcelona, Spain. Email: vrull@csic.es; valenti.rull@icp.cat Handling Editor: Jack Williams The landscape of the Azores Islands (Figure 1) is one of the most dense native forests before the Portuguese ­arrival. The critique deeply transformed among oceanic islands, as most of their pre- by Elias et al. (2022) was based on a number of methodological anthropic forests have been replaced by other vegetation types in- and interpretation points, including dating issues and alternative troduced from elsewhere (Dias, 2007). This has been considered a explanations for the occurrence of pollen from ­a nthropogenic large-­scale unintentional ecological and biogeographical experiment plants, faecal lipids and the natural (volcanic) or ­a nthropogenic on the assembly of island biotas and ecosystems (Rull, 2021). These origin of fires, as recorded by sedimentary charcoal. These transformations are known to have been caused by the Portuguese authors did not deny human presence before Portuguese con- colonizers who occupied the archipelago in the mid-­15th century tact, but they doubted that these earlier settlers deeply trans- (Dias, 2007), but whether they were the first settlers and the main formed the Azorean landscape in the way proposed by ­R aposeiro anthropization agents is currently under debate. The discussion is et al. (2021). Finally, Elias et al. (2022) considered the proposal focussed on two main points: the earliest human settlement and the of the Norse settlement an undemonstrated but ‘intriguing pos- first large-­scale landscape transformation (anthropization) across sibility’. ­R aposeiro et al. (2022) replied to these comments and the archipelago. reaffirmed their former conclusions. However, these authors Raposeiro et al. (2021) contended that the Azorean archipelago admitted that a number of questions remain open and need fur- was settled by the early Middle Ages (700–­850 CE), much earlier ther work. They also seemed to relax their former conclusions than the Portuguese occupation in the 15th century. According to about the settlement timing and the origin of early settlers. these authors, the earlier colonizers would have been the Norse, Indeed, they concluded that their multisite and multiproxy anal- and their discovery would have been facilitated by anomalous north- yses ‘…strongly suggest that people had already occupied the easterly winds and warmer temperatures. These conclusions were Azores Archipelago and altered the “pristine” landscape before based on multiproxy analyses from lake sediments, which Raposeiro the official arrival of the Portuguese’, with no mention to a poten- et al. (2021) interpreted as a widespread and irreversible transforma- tial Medieval settlement by Norse cultures. tion of terrestrial and aquatic ecosystems. Extensive deforestation In this paper, I briefly discuss these contrasting views based on by logging and fire for cereal cultivation, as recorded by pollen and my own palaeoecological experience in the archipelago, where I have charcoal analyses, and livestock grazing, as documented by the pres- collaborated with Raposeiro, Elias and their co-­workers in the past ence of spores from fungi living in the dung of herbivores and faecal (Rull, Connor, et al., 2017; Rull, Lara, et al., 2017) but have not par- lipids characteristic of these animals, would have been the main dis- ticipated in the recent papers discussed here. The raw material for turbances. These authors proposed that the Azores landscape was this discussion is the evidence furnished by Raposeiro et al. (2021), as already deeply and extensively modified when the Portuguese colo- Elias et al. (2022) did not provide new data. First, it should be high- nized the archipelago. lighted that the evidence provided by Raposeiro et al. (2021) covers, Elias et al. (2022) disagreed with this view, arguing that the evi- for the first time, the whole archipelago and uses a wide arrangement dence provided by Raposeiro et al. (2021) is insufficient to deny the of proxies useful for integrative climatic, ecological and anthropo- written historical records documenting a ‘pristine’ landscape with genic reconstructions. This should be able to provide a first approach This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2023 The Author. Journal of Biogeography published by John Wiley & Sons Ltd. Journal of Biogeography. 2023;00:1–4.  wileyonlinelibrary.com/journal/jbi | 1 | 13652699, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jbi.14712 by Universitat Autonoma De Barcelona, Wiley Online Library on [06/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 2 RULL F I G U R E 1 Map of the Azores Islands indicating the sites with palaeoecological records discussed in the text (yellow dots). 1, Caldeirão; 2, Alagoinha; 3, Rasa; 4, Funda; 5, Caveiro; 6, Peixinho; 7, Pico; 8, Ginjal; and 9, Azul. Coring localities after Connor et al. (2012, 2013); Rull, Lara, et al. (2017) and Raposeiro et al. (2021). to a holistic view of the Azorean archipelago (Figure 1). This said, Another issue is the inaccuracy of some representations, ­ some aspects of the Raposeiro et al. (2021) reconstruction not raised including the cereal—­Cerealia, not ‘Cerealea’, as repeatedly stated by by Elias et al. (2022) will be analysed here in more detail, with the only Raposeiro et al. (2021)—­pollen, one of the most important proxies aim of contributing to identifying potential issues to be addressed by for anthropogenic impact. Indeed, in the same Lagõa Azul record, future studies, as suggested by Raposeiro et al. (2022). rye (Secale) pollen is represented only in the 13th century in the Raposeiro et al. (2021) presented their results in a synthetic form of scattered occurrences, but it truly occurred between the fashion and the detailed outputs of their analyses are unavail- 13th and 19th centuries (Figure 2). Additionally, other cereal pollen, able, as they have not been published elsewhere or displayed in such as wheat (Triticum), are omitted by Raposeiro et al. (2021) but the supporting material. For example, in the case of landscape were actually present in a continuous and consistent fashion from anthropization, a major point of the debate, pollen records are the 13th century to the present. If correctly represented, cereal the critical evidence, but only summary diagrams are provided, pollen trends would have strongly supported the interpretation of which hinders detailed reconstructions of vegetation change and pre-­Portuguese cultivation, which makes these omissions particu- comparison with other similar records. Particularly significant is larly significant. The lack of detailed diagrams in records from other the lack of differentiation within forest categories, especially be- localities prevents us from knowing whether these inconsistencies tween native and introduced forests but also in other vegetation are more general, including other sites and proxies. types. In one of the records (Lagõa Azul, São Miguel Island), this Of the five records utilized by Raposeiro et al. (2021), only two, separation is possible (Figure 2) because the detailed pollen dia- from the Corvo and Pico islands, extend back to 600 CE, whereas one gram was published some years ago (Rull, Lara, et al., 2017), but begins in 900–­1000 CE (Corvo) and the other two, from São Miguel in others, this cannot be evaluated with the information provided. and Terceira, start in 1300–­1400 CE (Figure 1). Therefore, whatever The same occurs in all other palaeoecological records, but detailed the evidence for human settlement, these records are insufficient to raw evidence remains unavailable for checking. Readers unaware demonstrate that the whole archipelago was settled before 900 CE, of fundamental aspects such as these are unable to evaluate the as only two records from central and western islands are available reliability of interpretations, and this might be valid for other prox- before that date. Additionally, the only significant forest decline re- ies. The solution is obvious: publish the detailed records for each corded before 900 CE corresponds to one of these records (Pico), proxy and their corresponding interpretations. whereas the other is too small to be interpreted in terms of forest | 13652699, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jbi.14712 by Universitat Autonoma De Barcelona, Wiley Online Library on [06/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License RULL 3 human impact in the archipelago, although the intensity of land- scape modification was not comparable to that occurring after Portuguese occupation (Rull, Lara, et al., 2017). Finally, the palaeoclimatic simulations utilized by Raposeiro et al. (2021) to infer the origin and the navigation routes of the purported early settlers between 700 CE and 850 CE do not corre- spond to this time period but to a later phase (850–­9 00 CE), with no overlap with the period of interest. Therefore, inferences about Norse people coming from northern Europe as the early colonists are unwarranted (Rull, 2023). Even in the case that the palaeocli- matic reconstruction utilized coincided with the time interval of interest, this would not constitute a proof for the Norse arrival to the Azores at those times, but just a hypothesis to be demon- strated with future studies, where archaeological and anthropo- logical evidence—­w hich is absent in the Raposeiro et al. (2021) study—­m ust be essential. In summary, according to the data provided by Raposeiro et al. (2021), the first consistent evidence of widespread anthro- pogenic impact on the Azorean landscape seems to occur in the 13th–­14th centuries, roughly a century before Portuguese con- tact, as formerly proposed by Rull, Lara, et al. (2017). This does not deny the ‘intriguing possibility’ (Elias et al., 2022) of more an- cient settlements, but additional evidence is needed to support this proposal. This evidence should be sought in new records con- taining pre-­1200 CE sediments, especially in the islands of Flores, Terceira and São Miguel (Figure 1). Pioneering palaeoecological F I G U R E 2 Selected elements of the pollen diagram from Lagõa works by Connor et al. (2012, 2013) documenting widespread an- Azul (São Miguel Island). The dominant taxa of native forests are in thropogenic impact only after the Portuguese arrival were funda- green, and the introduced tree taxa are in orange. Yellow dots are mental for unravelling the ecological history of the Azores Islands, the occurrences of cereal pollen. The dashed line represents the but new and consistent evidence on former human occupations official date of the Portuguese occupation. Simplified from Rull, Lara, et al. (2017). is emerging that cannot be disregarded. The work by Raposeiro et al. (2021) is very useful in that, contrary to the opinion of Elias et al. (2022), it provides strong and widespread support for pre-­ clearing (Corvo). In summary, the evidence for general landscape Portuguese human colonization of the Azores archipelago and ­anthropization before the arrival of the Portuguese is insufficient. the ensuing landscape impact (Rull, 2023). However, whether this The most significant deforestation events took place between 1200 human occupation took place in early Medieval times by Norse and 1400 CE, as recorded in the other three sequences from Flores, people, facilitated by favourable climatic conditions, remains Pico and São Miguel. undemonstrated. The only pre-­9 00 CE deforestation documented in Pico co- incided with the first consistent evidence for fire (charcoal and AC K N O​W L E​D G E​M E N T S polycyclic aromatic carbon), along with a tephra layer and a local The author is grateful to Rui Elias and an anonymous reviewer for peak of faecal lipids attributed by Raposeiro et al. (2021) to the constructive and helpful comments. No funding was received spe- presence of livestock, as wild herbivores were absent prior to cifically for the development of this paper. No new data have been human settlement. Both coprophilous fungi, a grazing proxy, and used in this paper. No permits have been required for the develop- coprostanol, a faecal lipid present in herbivores, humans and other ment of this work. omnivores, were absent until the 11th–­12th centuries. Under these conditions, it is not possible to attribute the documented C O N F L I C T O F I N T E R E S T S TAT E M E N T fires to either volcanic activity or human influence. Therefore, the The author declares no conflict of interest. only potential evidence of human impact before these dates re- mains debatable. Again, widespread evidence for human presence DATA AVA I L A B I L I T Y S TAT E M E N T is more consistent across all islands analysed between 1200 and Data sharing not applicable to this article as no datasets were 1400 CE. Therefore, these dates emerge as the best candidates for ­generated or analysed during the current study. | 13652699, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/jbi.14712 by Universitat Autonoma De Barcelona, Wiley Online Library on [06/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 4 RULL REFERENCES Aguiar, N., Pereira, C., Ritter, C., Rubio, M. J., … Giralt, S. (2021). Connor, S. E., van der Knapp, O., van Leeuwen, J. F. N., & Kuněs, P. (2013). Climate change facilitated the early colonization of the Azores Holocene palaeoclimate and palaeovegetational on the islands of Archipelago during medieval times. Proceedings of the National Flores and Pico. In J. M. Fernández-­Palacios, et al. (Eds.), Climate Academy of Sciences of the United States of America, 118(41), change perspectives from the Atlantic: Past, present and future (pp. e2108236118. 149–­162). Universidad de La Laguna, Servicio de Publicaciones. Rull, V. (2021). Pristinity, degradation and replacement: The three di- Connor, S. E., van Leeuwen, J. F. N., Rittenour, T. M., van der Knapp, O., mensions of human impact on Island vegetation. Progress in Physical Ammann, B., & Björk, S. (2012). The ecological impact of oceanic Geography, 45, 407–­426. Island colonization –­ A palaeoecological perspective from the Rull, V. (2023). Human settlement and landscape anthropization of re- Azores. Journal of Biogeography, 39, 1007–­1023. mote oceanic islands: A comparison between Rapa Nui (Pacific Dias, E. (2007). Açores e Madeira: A floresta das ilhas. Fundaçao Luso Ocean) and the Azores (Atlantic Ocean). Plants, 12, 2089. Americana. Rull, V., Connor, S. E., & Elias, R. B. (2017). Potential natural vegeta- Elias, R. B., Connor, S. E., Góis-­Marques, C. A., Schaefer, H., Silva, L., tion and pre-­anthropic pollen records on the Azores Islands in a Sequeira, M. M., Moura, M., Borges, P. A. V., & Gabriel, R. (2022). Macaronesian context. Journal of Biogeography, 44, 2437–­2440. Is there solid evidence of widespread landscape disturbance in the Rull, V., Lara, A., Rubio-­Inglés, M. J., Giralt, S., Gonçalves, V., Raposeiro, Azores before the arrival of the Portuguese? Proceedings of the P., Hernández, A., Sánchez-­López, G., Vázquez-­Loureiro, D., Bao, National Academy of Sciences of the United States of America, 119(4), R., Masqué, P., & Sáez, A. (2017). Vegetation and landscape dynam- e2119218119. ics under natural and anthropogenic forcing on the Azores islands: Raposeiro, P. M., Hernández, A., Pla-­Rabes, S., Gonçalves, V., Bao, R., A 700-­year pollen record from the São Miguel Island. Quaternary Sáez, A., Sáez, A., Shanahan, T., Benavente, M., de Boer, E. J., Richter, Science Reviews, 159, 155–­168. N., Gordon, V., Marques, H., Sousa, P. M., Souto, M., Matias, M. G., Aguiar, N., Pereira, C., Ritter, C., … Giralt, S. (2022). Multiproxy evi- dence of widespread landscape disturbance in multiple Azorean lakes before the Portuguese arrival. Proceedings of the National Academy of How to cite this article: Rull, V. (2023). Settlement and Sciences of the United States of America, 119(4), e2120107119. anthropization of the Azores Islands. Journal of Biogeography, Raposeiro, P. M., Hernández, A., Pla-­Rabes, S., Gonçalves, V., Bao, R., Sáez, A., Shanahan, T., Benavente, M., de Boer, E. J., Richter, N., 00, 1–4. https://doi.org/10.1111/jbi.14712 Gordon, V., Marques, H., Sousa, P. M., Souto, M., Matias, M. G.,