Electronic Supplementary Material for: The biogeography of population resilience in lowland South America

## Electronic Supplementary Information for "The biogeography of population resilience in lowland South America" chapter This repository holds the code and data for reproducing the analysis in the chapter "The biogeography of population resilience in lowland South America", forthcoming in the Oxford Handbook of Resilience in Climate History. It also contains supplementary information on the statistical modelling undertaken as part of the aforementioned work. The code comprises five principal sections, plus setup: 0. Setup & data loading1. Data processing and display2. Radiocarbon analysis3. Statistical modelling4. Output5. Supplementary output The code features an extended version of the p2pPerm function: (https://github.com/philriris/p2pPerm) that was first introduced in Riris and De Souza (2021) (https://doi.org/10.3389/fevo.2021.740629), here called the `resmet` (RESilience METrics) function. In addition, the code is accompanied by three datasets: - A table containing archaeological radiocarbon dates from lowland tropical South America - A shapefile of South American ecoregions, original data available here: http://ecologicalregions.info/data/sa/- A table of domesticated Neotropical plants resolved in Amazonian palaeoecological records, after Iriarte et al. (2020)(https://doi.org/10.1016/j.quascirev.2020.106582) Briefly, the georeferenced radiocarbon data used in the paper have been compiled from a wide range of sources, including Goldberg et al. (2016), Riris & Arroyo-Kalin (2019), Napolitano et al. (2019) Arroyo-Kalin & Riris (2021), De Souza & Riris (2021), and Bird et al. (2022). These sources have been extensively cross-checked for duplicate lab codes, variation in site naming conventions, and reported locations, in order to minimise errors arising from these variables. It does not purport to be error-free, although it is adequate for the current analysis. As well as its use in the production of Figure 1, the ecoregions shapefile has been intersected with the radiocarbon date locations to append this information to `rhdata.csv`. An extended description and rationale for its use can be found in the main text. For the convenience of the end-user, an additional file containing the main results (metrics_regular.csv) is included. The data contained in this table is the subject of section 3 of the code, Statistical Modelling. It forms the basis of the discussion in the chapter. Data cleaning was carried out manually on the raw output of the `resmet` function to remove false positives from the table. These "events" are either: a) statistically significant downturns present in periods where, logically, no humans should be present, e.g. in the Greater Antilles before ~6000 cal BP, or: b) downturns where there are no minima in the summed probability distributions of calibrated radiocarbon dates, returning nonsensical resilience metrics. Removing these data rows introduces errors to the variable Cumulative, which counts the cumulative number of downturns detected by the `permTest` function in `rcarbon`. The file version of the output in this repository should be considered authoritative for present purposes, as these counting errors in Cumulative have been manually fixed too. ### References - Arroyo-Kalin, M. and Riris, P. 2021. Did pre-Columbian populations of the Amazonian biome reach carrying capacity during the Late Holocene? *Phil. Trans. R. Soc. B* 376: 20190715 http://doi.org/10.1098/rstb.2019.0715 - Bird, D., Miranda, L., Vander Linden, M., Robinson, E., Bocinsky, R.K., Nicholson, C., Capriles, J.M., Finley, J.B., Gayo, E.M., Gil, A. and d’Alpoim Guedes, J. 2022. p3k14c, a synthetic global database of archaeological radiocarbon dates. *Scientific Data* 9: 1-19. https://doi.org/10.1038/s41597-022-01118-7 - De Souza, J.G. & Riris, P. 2021. Delayed demographic transition following the adoption of cultivated plants in the eastern La Plata Basin and Atlantic coast, South America. *Journal of Archaeological Science*. 125: 105293. https://doi.org/10.1016/j.jas.2020.105293 - Goldberg, A., Mychajliw, A.M. and Hadly, E.A. 2016. Post-invasion demography of prehistoric humans in South America. *Nature* 532: 232-235. https://doi.org/10.1038/nature17176 - Iriarte, J., Elliott, S., Maezumi, S.Y., Alves, D., Gonda, R., Robinson, M., de Souza, J.G., Watling, J. and Handley, J. 2020. The origins of Amazonian landscapes: Plant cultivation, domestication and the spread of food production in tropical South America. _Quaternary Science Reviews_ 248: 106582. https://doi.org/10.1016/j.quascirev.2020.106582 - Napolitano MF, DiNapoli RJ, Stone JH, Levin MJ, Jew NP, Lane BG, O’Connor JT, Fitzpatrick SM. 2019. Reevaluating human colonization of the Caribbean using chronometric hygiene and Bayesian modeling. _Science Advances_. 5: eaar7806. https://doi.org/10.1126/sciadv.aar7806 - Riris, P. and Arroyo-Kalin, M. 2019. Widespread population decline in South America correlates with mid-Holocene climate change. *Scientific Reports* 9: 6850. https://doi.org/10.1038/s41598-019-43086-w - Riris, P. and De Souza, J.G. 2021. Formal tests for resistance-resilience in archaeological time series. _Frontiers in Ecology and Evolution_, 9. https://doi.org/10.3389/fevo.2021.740629# 《低地南美洲种群恢复力的生物地理学》章节电子补充材料 本存储库包含用于复现《低地南美洲种群恢复力的生物地理学》章节分析的代码与数据，该章节即将收录于《气候史恢复力牛津手册》（*Oxford Handbook of Resilience in Climate History*）。本存储库同时收录了前述研究中开展的统计建模相关补充材料。 本代码包含五大核心模块及初始化环节： 0. 初始化与数据加载 1. 数据处理与可视化 2. 放射性碳分析 3. 统计建模 4. 结果输出 5. 补充结果输出 本代码使用了经扩展的`p2pPerm`函数（原始版本首次见于Riris与De Souza 2021年发表的研究，https://doi.org/10.3389/fevo.2021.740629，其GitHub仓库地址为https://github.com/philriris/p2pPerm），本研究中将其命名为`resmet`（恢复力指标，RESilience METrics）函数。 此外，本代码配套包含三类数据集： - 收录低地热带南美洲考古放射性碳测年数据的表格 - 南美洲生态区形状文件（shapefile），原始数据可从以下地址获取：http://ecologicalregions.info/data/sa/ - 基于Iriarte等人2020年的研究（https://doi.org/10.1016/j.quascirev.2020.106582）整理的、见于亚马逊古生态记录的新热带界驯化植物数据表 简而言之，本研究使用的带地理坐标的放射性碳数据整合自多项公开来源，包括Goldberg等人（2016）、Riris与Arroyo-Kalin（2019）、Napolitano等人（2019）、Arroyo-Kalin与Riris（2021）、De Souza与Riris（2021）以及Bird等人（2022）的研究成果。研究团队已对这些数据源进行了全面交叉校验，以排查重复的实验室编号、遗址命名规范差异以及报告位置偏差等问题，尽可能降低此类因素带来的误差。本数据集并非完全无错，但足以支撑当前的分析工作。 除用于绘制图1外，本研究还将南美洲生态区形状文件与放射性碳测年点位进行空间叠加，将生态区信息追加至`rhdata.csv`文件中。关于该形状文件的详细说明与使用依据，可参见正文内容。 为方便终端用户使用，本存储库额外收录了包含主要研究结果的文件`metrics_regular.csv`。该表格中的数据对应代码的第3模块——统计建模，亦是本章节讨论部分的核心依据。 研究团队已对`resmet`函数的原始输出结果进行手动数据清洗，以移除表格中的假阳性结果。此类“异常事件”分为两类：a) 统计学上显著的人口下降时段，但逻辑上该时段并无人类活动存在，例如大安的列斯群岛约公元前6000校准距今年（cal BP）之前的时期；或b) 经校准的放射性碳测年累加概率分布无最小值的下降时段，此时得到的恢复力指标无实际意义。移除这些数据行会对`Cumulative`变量产生误差，该变量用于统计`rcarbon`包中`permTest`函数检测到的人口下降累计次数。本存储库中提供的输出文件版本具备权威性，因为`Cumulative`变量的计数误差也已通过手动方式修正。 ## 参考文献 1. Arroyo-Kalin, M. 与 Riris, P., 2021. 亚马逊生物群系的前哥伦布时期种群是否在全新世晚期达到环境容纳量？*《英国皇家学会哲学学报B辑：生物科学》*（Phil. Trans. R. Soc. B）, 376: 20190715, https://doi.org/10.1098/rstb.2019.0715 2. Bird, D., Miranda, L., Vander Linden, M. 等, 2022. p3k14c：全球考古放射性碳测年合成数据库. *《科学数据》*（Scientific Data）, 9: 1-19, https://doi.org/10.1038/s41597-022-01118-7 3. De Souza, J.G. 与 Riris, P., 2021. 南美洲东部拉普拉塔盆地与大西洋沿岸采用栽培植物后出现的延迟人口转型. *《考古科学杂志》*（Journal of Archaeological Science）, 125: 105293, https://doi.org/10.1016/j.jas.2020.105293 4. Goldberg, A., Mychajliw, A.M. 与 Hadly, E.A., 2016. 南美洲史前人类的入侵后人口动态. *《自然》*（Nature）, 532: 232-235, https://doi.org/10.1038/nature17176 5. Iriarte, J., Elliott, S., Maezumi, S.Y. 等, 2020. 亚马逊景观的起源：热带南美洲的植物栽培、驯化与食物生产传播. *《第四纪科学评论》*（Quaternary Science Reviews）, 248: 106582, https://doi.org/10.1016/j.quascirev.2020.106582 6. Napolitano, M.F., DiNapoli, R.J., Stone, J.H. 等, 2019. 利用测年数据清洗与贝叶斯模型重新评估加勒比地区的人类殖民历史. *《科学进展》*（Science Advances）, 5: eaar7806, https://doi.org/10.1126/sciadv.aar7806 7. Riris, P. 与 Arroyo-Kalin, M., 2019. 南美洲广泛的人口下降与中全新世气候变化相关. *《科学报告》*（Scientific Reports）, 9: 6850, https://doi.org/10.1038/s41598-019-43086-w 8. Riris, P. 与 De Souza, J.G., 2021. 考古时间序列中抗逆性与恢复力的正式检验. *《生态学与进化前沿》*（Frontiers in Ecology and Evolution）, 9, https://doi.org/10.3389/fevo.2021.740629