Ecology

The crowdsourcing campaign was organized as a competition with prizes offered to those who contributed the most, based on a combination of quality and quantity. The Geo-Wiki platform (www.geo-wiki.org), a web platform dedicated to engaging citizens in environmental monitoring, was used as the tool to perform the campaign. A customized user interface was prepared for the campaign (Fig. 2), where participants were shown a random location in the tropics (here broadly defined as the area between 30 degrees of latitude north and south of the equator, i.e., including part of the subtropics), where a blue 1 × 1 km box showed the location to be visually interpreted. The Global Forest Change (GFC) tree loss map (v1.7)10 was overlaid on the imagery to show all areas where tree loss was detected at any point between 2008 and 2019. The tree loss area was shaded in red and the map itself was aggregated to 100 m for fast rendering.Fig. 2Customized Geo-Wiki interface for the ‘Drivers of Tropical Forest Loss’ crowdsourcing campaign showing: (a) Tools available to participants such as the NDVI and Sentinel time-series profiles, visualizing the location on Google Earth and exploring the imagery time-series, reviewing the quick-start guide and exploring examples to identify specific drivers of forest loss as well as contacting IIASA staff via chat or email; (b) country and continent of the location as well as dates of the imagery shown; (c) campaign statistics; (d) available background imagery; and (e) tasks to be undertaken by the participants along with buttons to submit or skip the location.Full size imageThe year 2008 was selected as the start date because the RED states that date as the cut-off year for conversion from high-carbon areas, i.e., forest, to other land uses7. In order to capture the main drivers of forest loss, but also include potential additional drivers such as the existence of roads as precursors of deforestation, the participants were asked to complete three steps: 1) To select the predominant tree loss driver visible inside the tree loss pixels in the blue box from a list of nine specific drivers; 2) to select all other tree loss drivers visible inside the tree loss pixels in the blue box from a list of five more general drivers, and 3) to mark if roads, trails, or buildings were visible in the blue box. The list of specific and general drivers as well as their definitions is shown in Table 1. The Geo-Wiki interface allowed participants to switch between different background imagery such as ESRI, Google Maps, and Bing Maps as well as Sentinel 2 satellite imagery. The different sources of imagery allowed the participants to see the location at different resolutions and in different periods of time. It also provided participants with information about the current country and the continent as well as the dates of the background imagery. Furthermore, it provided the participants with links for displaying NDVI and Sentinel time series, and to see the location and explore the historical imagery using the Google Earth platform. All these tools were meant to help with easier identification of the forest loss drivers by allowing participants to look at the locations during different times and at different spatial resolutions.Table 1 List and description of the available list of tree loss drivers that participants could select for steps 1 and 2 of the campaign.Full size tableAt the beginning of the campaign, each participant was shown a quick start guide of the interface and the tasks requested. As shown in Fig. 2, this quick start guide could be accessed again at any point during the campaign. Figure 2 also shows that the interface had buttons for four further functions. The first was to see the gallery of examples with access to pre-loaded video-tutorials and examples of images describing each driver of forest loss and how to do visual interpretation and selection of each of these (available at https://application.geo-wiki.org/Application/modules/drivers_forest_change/drivers_forest_change_gallery.html). An illustration of the gallery of examples shown to participants is shown in Figure S1. The second function was to ask experts for help, which automatically sent IIASA experts an email regarding a specific location. The third was to join the expert chat, which led participants to a dedicated chat interface on the Discord messaging platform. Here participants could pose questions and interact with staff and other participants directly. Finally, there was a button to see the leader board as well as the aims, rules and prizes of the campaign (available at https://application.geo-wiki.org/Application/modules/drivers_forest_change/drivers_forest_change.html). When the participants started the campaign, they were shown 10 initial practice locations, where they could try out the user interface (UI) with control points, which showed the participants how to identify the different drivers of forest loss. This set of videos, the images and the training points, together with the gallery of images, were developed to train the participants before and during the campaign.Campaign set-up and data qualityAs the aim of the campaign was to determine the drivers of tree loss across the tropics, the sample locations were selected from the GFC tree loss layer10 for the tropics (between 30 degrees north and south of the equator). No stratification was used since a completely random sample across the tropics was deemed to be the fairest representation of tree loss and their corresponding drivers. The previous map of deforestation drivers6 used a 5 K sample of 10 × 10 km grid cells to produce a global map. Here the sample size was largely driven by the estimated capacity of the crowd. Hence, we aimed to validate ca. 150k 1 × 1 km locations across the tropics, which is a considerably larger sample size than that of Curtis et al.6. In order to reduce noise, the GFC tree loss layer10 was first aggregated to a 100 m resolution from the original 30 m, and 150 K centroids were then randomly selected. From these, a sub sample of 5000 random locations were selected for visual interpretation by six IIASA experts (with backgrounds in remote sensing, agronomy, forestry and geography). Due to time constraints, only 2001 locations were evaluated by at least three different experts. In these locations, agreement was discussed and once a consensus was reached, these locations became the final control or expert data set. The control locations were then used to produce quality scores for each participant as the campaign progressed in order to rank them and determine the final prize winners. The list of prizes offered to the top 30 participants is shown in Table S1 in the Supplementary Information (SI), and a list and rank of motivations mentioned by the participants is shown on Figure S2 in the SI.The control locations were randomly shown to the participants at a ratio of approximately 2 control locations to every 20 non-control locations visited. If the participants correctly selected the predominant tree loss driver (in step 1), they were awarded 20 points; if they selected the wrong answer, they lost 15 points. If participants confused pasture and commercial agriculture or wildfire with other natural disturbances, they lost only 10 points instead of 15. Furthermore, they could win 8 additional points by selecting the correct secondary drivers in step 2. If a mixture of correct and incorrect answers were provided in step 2, the participants gained 2 points for every correct choice and lost 2 points for every incorrect one, with a minimum gain/loss of 0 points. Finally, participants could earn 2 additional points by correctly reporting the existence of roads, trails or buildings in step 3. The scoring system was based on previous Geo-Wiki campaign experiences and aimed to promote focus on the primary driver selection. The points were used to produce a leader board with the total number of points by participant. Additionally, a relative quality score (RQS) was derived from the score received by the users and the potential score that could have been obtained if all control points were correctly interpreted. This is shown in Eq. 1.$${rm{RQS}}=(({{rm{NCP}}}^{ast }15+{rm{SumScore}})/{rm{NCP}})/45$$
(1)
where RQS ranges between 0 and 1, NCP is the number of control points visited and SumScore is the number of points obtained.The RQS was crucial in understanding how each participant performed in terms of the quality of their visual interpretations, as this was independent of the number of locations interpreted. Once the campaign ended, an average RQS was used as a minimum criterion for participants to receive a prize, independent of where they were located on the leader board. Additionally, all users who submitted a substantial number of interpretations, i.e., more than 1000 with the minimum required RQS, were invited to become co-authors of the current manuscript, independent of whether they received a monetary prize or not. All these co-authors additionally contributed to the editing and revision of this manuscript. Furthermore, future users of the data set could use the RQS as a key data quality indicator.After the campaign, the data post-processing included eliminating interpretations made by users who broke any of the competition rules. Additionally, during the campaign, some users communicated with IIASA staff using the “Ask Experts” button and pointed out that some control points were mistaken. Consequently, the corresponding points lost were added to the final score of those participants where the correction was made. A total of 18742 validations from 1 participant were removed before the end of the campaign and the user was disqualified since their account was deemed to be shared across several people and computers, which was not allowed. Validations from another user (38,502 out of 40,828) were also removed due to inconsistencies but the user remained in the competition. Before the prizes were awarded to the top 30 users, a questionnaire was administered to all users to gather information about participant characteristics and gauge their motivations. Participation was mandatory for the top 30 users. A summary of the participant backgrounds is provided in Figure S3 in the SI. More

FAO & UNEP. The state of the world’s forests 2020: Forests, biodiversity and people (2020).Brinck, K. et al. High resolution analysis of tropical forest fragmentation and its impact on the global carbon cycle. Nature Communications 8, 1–6 (2017).ADS 

Google Scholar 
Mitchard, E. T. The tropical forest carbon cycle and climate change. Nature 559, 527–534 (2018).ADS 
CAS 
PubMed 

Google Scholar 
Hansen, M. C. et al. High-resolution global maps of 21st-century forest cover change. science 342, 850–853 (2013).ADS 
CAS 
PubMed 

Google Scholar 
Tsujino, R., Yumoto, T., Kitamura, S., Djamaluddin, I. & Darnaedi, D. History of forest loss and degradation in Indonesia. Land use policy 57, 335–347 (2016).
Google Scholar 
Holl, K. D. & Brancalion, P. H. Tree planting is not a simple solution. Science 368, 580–581 (2020).ADS 
CAS 
PubMed 

Google Scholar 
Zhang, Y. et al. Multiple afforestation programs accelerate the greenness in the ‘Three North’ region of China from 1982 to 2013. Ecological Indicators 61, 404–412 (2016).
Google Scholar 
Soliño, M., Oviedo, J. L. & Caparrós, A. Are forest landowners ready for woody energy crops? Preferences for afforestation programs in Southern Spain. Energy Economics 73, 239–247 (2018).
Google Scholar 
Paquette, A. & Messier, C. The role of plantations in managing the world’s forests in the Anthropocene. Frontiers in Ecology and the Environment 8, 27–34 (2010).
Google Scholar 
Zulkefli, F., Syahlan, S. & Aziz, M. F. A. Negatives Impact Faced by Oil Palm Estate Management in managing Foreign Workers: A Case Study. International Journal of Academic Research in Business and Social Sciences 8 (2018).Fitzherbert, E. B. et al. How will oil palm expansion affect biodiversity? Trends in ecology & evolution 23, 538–545 (2008).
Google Scholar 
Vijay, V., Pimm, S. L., Jenkins, C. N. & Smith, S. J. The impacts of oil palm on recent deforestation and biodiversity loss. PloS one 11, e0159668 (2016).PubMed 
PubMed Central 

Google Scholar 
Koh, L. P., Miettinen, J., Liew, S. C. & Ghazoul, J. Remotely sensed evidence of tropical peatland conversion to oil palm. Proceedings of the National Academy of Sciences 108, 5127–5132 (2011).ADS 
CAS 

Google Scholar 
Guillaume, T. et al. Carbon costs and benefits of Indonesian rainforest conversion to plantations. Nature communications 9, 1–11 (2018).CAS 

Google Scholar 
Lucas-Borja, M. E., Hedo, J., Cerdá, A., Candel-Pérez, D. & Viñegla, B. Unravelling the importance of forest age stand and forest structure driving microbiological soil properties, enzymatic activities and soil nutrients content in Mediterranean Spanish black pine (Pinus nigra Ar. ssp. salzmannii) Forest. Science of the Total Environment 562, 145–154 (2016).ADS 
CAS 

Google Scholar 
Besnard, S. et al. Quantifying the effect of forest age in annual net forest carbon balance. Environmental Research Letters 13, 124018 (2018).ADS 

Google Scholar 
Dzikiti, S. et al. Estimating the water requirements of high yielding and young apple orchards in the winter rainfall areas of South Africa using a dual source evapotranspiration model. Agricultural water management 208, 152–162 (2018).
Google Scholar 
Zhang, Y., Yao, Y., Wang, X., Liu, Y. & Piao, S. Mapping spatial distribution of forest age in China. Earth and Space Science 4, 108–116 (2017).ADS 

Google Scholar 
Chen, B., Jin, Y. & Brown, P. Automatic mapping of planting year for tree crops with Landsat satellite time series stacks. ISPRS Journal of Photogrammetry and Remote Sensing 151, 176–188 (2019).ADS 

Google Scholar 
Danylo, O. et al. A map of the extent and year of detection of oil palm plantations in Indonesia, Malaysia and Thailand. Scientific data 8, 1–8 (2021).
Google Scholar 
O’Brien, S. T., Hubbell, S. P., Spiro, P., Condit, R. & Foster, R. B. Diameter, height, crown, and age relationship in eight neotropical tree species. Ecology 76, 1926–1939 (1995).
Google Scholar 
Fichtler, E., Clark, D. A. & Worbes, M. Age and long-term growth of trees in an old-growth tropical rain forest, based on analyses of tree rings and 14C1. Biotropica 35, 306–317 (2003).
Google Scholar 
Zhang, C. et al. Mapping forest stand age in China using remotely sensed forest height and observation data. Journal of Geophysical Research: Biogeosciences 119, 1163–1179 (2014).ADS 

Google Scholar 
Wang, B., Li, M., Fan, W., Yu, Y. & Chen, J. M. Relationship between net primary productivity and forest stand age under different site conditions and its implications for regional carbon cycle study. Forests 9, 5 (2018).
Google Scholar 
Wang, S. et al. Relationships between net primary productivity and stand age for several forest types and their influence on China’s carbon balance. Journal of environmental management 92, 1651–1662 (2011).PubMed 

Google Scholar 
Gupta, N., Kukal, S., Bawa, S. & Dhaliwal, G. Soil organic carbon and aggregation under poplar based agroforestry system in relation to tree age and soil type. Agroforestry Systems 76, 27–35 (2009).
Google Scholar 
Huang, C. et al. An automated approach for reconstructing recent forest disturbance history using dense Landsat time series stacks. Remote Sensing of Environment 114, 183–198 (2010).ADS 

Google Scholar 
Thomas, N. E. et al. Validation of North American forest disturbance dynamics derived from Landsat time series stacks. Remote Sensing of Environment 115, 19–32 (2011).ADS 
MathSciNet 

Google Scholar 
Ye, S., Rogan, J., Zhu, Z. & Eastman, J. R. A near-real-time approach for monitoring forest disturbance using Landsat time series: Stochastic continuous change detection. Remote Sensing of Environment 252, 112167 (2021).ADS 

Google Scholar 
Verbesselt, J., Hyndman, R., Newnham, G. & Culvenor, D. Detecting trend and seasonal changes in satellite image time series. Remote sensing of Environment 114, 106–115 (2010).ADS 

Google Scholar 
Kennedy, R. E., Yang, Z. & Cohen, W. B. Detecting trends in forest disturbance and recovery using yearly Landsat time series: 1. LandTrendr—Temporal segmentation algorithms. Remote Sensing of Environment 114, 2897–2910 (2010).ADS 

Google Scholar 
Cohen, W. B., Yang, Z., Healey, S. P., Kennedy, R. E. & Gorelick, N. A LandTrendr multispectral ensemble for forest disturbance detection. Remote sensing of environment 205, 131–140 (2018).ADS 

Google Scholar 
Vogeler, J. C., Braaten, J. D., Slesak, R. A. & Falkowski, M. J. Extracting the full value of the Landsat archive: Inter-sensor harmonization for the mapping of Minnesota forest canopy cover (1973–2015). Remote sensing of environment 209, 363–374 (2018).ADS 

Google Scholar 
de Jong, S. M. et al. Mapping mangrove dynamics and colonization patterns at the Suriname coast using historic satellite data and the LandTrendr algorithm. International Journal of Applied Earth Observation and Geoinformation 97, 102293 (2021).
Google Scholar 
Harris, N., Goldman, E. D. & Gibbes, S. Spatial database of planted trees (SDPT VERSION 1.0). Technical Note. (2019).Descals, A. et al. High-resolution global map of smallholder and industrial closed-canopy oil palm plantations. Earth System Science Data 13, 1211–1231 (2021).ADS 

Google Scholar 
Li, C. et al. The first all-season sample set for mapping global land cover with landsat-8 data. Science Bulletin 62, 508–515 (2017).ADS 

Google Scholar 
Masek, J. G. et al. A Landsat surface reflectance dataset for North America, 1990–2000. IEEE Geoscience and Remote Sensing Letters 3, 68–72 (2006).ADS 

Google Scholar 
Vermote, E., Justice, C., Claverie, M. & Franch, B. Preliminary analysis of the performance of the Landsat 8/OLI land surface reflectance product. Remote Sensing of Environment 185, 46–56 (2016).ADS 
PubMed 

Google Scholar 
Foga, S. et al. Cloud detection algorithm comparison and validation for operational Landsat data products. Remote sensing of environment 194, 379–390 (2017).ADS 

Google Scholar 
Goulden, M. L. & Bales, R. C. California forest die-off linked to multi-year deep soil drying in 2012–2015 drought. Nature Geoscience 12, 632–637 (2019).CAS 

Google Scholar 
He, T. et al. Evaluating land surface albedo estimation from Landsat MSS, TM, ETM+, and OLI data based on the unified direct estimation approach. Remote Sensing of Environment 204, 181–196 (2018).ADS 

Google Scholar 
Flood, N. Continuity of reflectance data between Landsat-7 ETM+ and Landsat-8 OLI, for both top-of-atmosphere and surface reflectance: a study in the Australian landscape. Remote Sensing 6, 7952–7970 (2014).ADS 

Google Scholar 
Roy, D. P. et al. Characterization of Landsat-7 to Landsat-8 reflective wavelength and normalized difference vegetation index continuity. Remote sensing of Environment 185, 57–70 (2016).ADS 
PubMed 

Google Scholar 
Key, C. & Benson, N. Landscape assessment: remote sensing of severity, the normalized burn ratio and ground measure of severity, the composite burn index. FIREMON: Fire effects monitoring and inventory system Ogden, Utah: USDA Forest Service, Rocky Mountain Res. Station (2005).Guo, J. & Gong, P. The potential of spectral indices in detecting various stages of afforestation over the Loess Plateau Region of China. Remote Sensing 10, 1492 (2018).ADS 

Google Scholar 
Kennedy, R. E. et al. Implementation of the LandTrendr algorithm on google earth engine. Remote Sensing 10, 691 (2018).ADS 

Google Scholar 
Yu, L. et al. A multi-resolution global land cover dataset through multisource data aggregation. Science China Earth Sciences 57, 2317–2329 (2014).ADS 

Google Scholar 
Du, Z. et al. A global map of planting years of plantations. figshare https://doi.org/10.6084/m9.figshare.19070084.v1 (2022).Gong, P. et al. Finer resolution observation and monitoring of global land cover: First mapping results with landsat tm and etm+ data. International Journal of Remote Sensing 34, 2607–2654 (2013).ADS 

Google Scholar 
Huang, H. et al. The migration of training samples towards dynamic global land cover mapping. ISPRS Journal of Photogrammetry and Remote Sensing 161, 27–36 (2020).ADS 

Google Scholar  More

Tagging deployments and study subjectsTable 1 contains an overview of the fields in the metadata file (JWS_metadata.xlsx) providing extensive background details on each of the 79 tag deployments and 63 study subjects. The data in this file give essential contextual information needed to understand the methodological, environmental, and demographic factors surrounding the deployments, which are critical for further examination and hypothesis testing of the sensor data. These metadata fall into several specific categories, but are not limited to, (i) information on the deployed electronic devices (platform, model, Platform Transmitter Terminal identifications), (ii) sharks (unique identifying numbers, sex, length), (iii) capture event (date, location, duration, methodology, interaction type), and (iv) the reporting period (duration, linear surface travel distance).Table 1 Metadata descriptions of the sharks, tagging operations, and deployments for all tags included in the database.Full size tableFigure 1 illustrates a typical C. carcharias tagging operation. This involves a contracted commercial fishing vessel with purpose-built gears to capture sharks (Fig. 1a) and a research crew to handle animals, monitor health (Fig. 1b) and attach electronic tags (Fig. 1c). More details on the tagging program and its methodologies are provided elsewhere14,19,20. Figure. 2 provides summaries of the deployment schedule, geographic locations, devices, and capture operations. Of note, 39.7% (25/64) of all tagging operations involved collaborations with commercial fishery operators (Fig. 2f–h), whose engagement was temporarily impacted (Fig. 2a) during the scientific review process when the population was under consideration for US Endangered Species Act listing. Figure 3 displays the demographic focus on small juvenile C. carcharias, with modest deployment durations and travel distances.Fig. 1Depiction of a typical research operation for capturing and tagging juvenile White Sharks in the Southern California Bight. (a) Aquarium research vessel (RV Lucile) with crew approaching a contracted purse seine vessel containing a captured juvenile white shark. (b) Research crew on the RV Lucile leading the shark into a sling, where it is subsequently transferred to the vessel’s deck for tagging. (c) Successfully applied PAT and acoustic tags each positioned lateral of the dorsal fin, anchored via leaders, and affixed with titanium darts (yellow arrows). All images taken by Steve McNicholas (Great White Shark 3D) for the Monterey Bay Aquarium and used with permission.Full size imageFig. 2Metadata summaries of the field program that deployed biologging tags on juvenile white sharks in the southern California Current. (a) Deployment schedule for 72 electronic tags released on 64 White Sharks from 2001–2020 (b) Tagging activity peaked in the late summer months when the population is most locally abundant. Field operations decreased from 2011–2013 when the population was being considered for listing under the U.S. Endangered Species Act (ESA). (c) Deployments focused on opportunities in the Southern California Bight coastline and included deployments in the nursery area of Bahía Sebastian Vizcaíno, Mexico and releases after exhibition at the Monterey Bay Aquarium. (d) Researchers released a variety of pop-up archival transmitting (PAT, 58 sharks), acoustic (21 sharks), and smart position and temperature (SPOT, 20 sharks) tags. This manuscript only reports the geolocation, temperature and depth data from the PAT and SPOT platforms. (e) Half (35 of 64, 54.7%) of all sharks received multiple tags, primarily to compare their relative performance. (f) Most tags (38 of 64, 60.3%) were deployed during focused scientific research operations. (g) The remainder were joint operations resulting from opportunistic bycatch in commercial fisheries using various gears and (h) Targeting various species. “Jab” gear refers to research operations that uses pole extensions to apply tags to sharks without capturing and handling.Full size imageFig. 3Demographic and deployment summaries from the juvenile white shark tagging program. (a) Total body length (TL) histogram indicates that most individuals tagged were either neonates ( More

Ebola datasetThe 2018–2020 DRC EVD outbreak lasted over 24 months and spread over 3 distinct spatial and temporal waves. Between the emergency declaration of the EVD outbreak in northern DRC on August 1, 2018 and the outbreak’s official end on June 25, 2020, the DRC Ministry of Health has reported a total of 3481 cases (including confirmed and probable), 1162 recoveries, and 2299 deaths16 in the provinces of Northern Kivu, Southern Kivu, and Ituri. The dataset considered here is a large subset of the entire EVD database compiled by the University of Kinshasa School of Public Health, which comprises 3117 total case records (confirmed and probable) recorded between May 3, 2018, and September 12, 2019. The data included partially de-identified but still detailed patient information, such as each person’s location, date of symptom onset and hospitalization, as well as discharge due to recovery or death. These individual records came from the Ebola treatment centers in 24 different health zones, spread out among the three DRC provinces of Northern Kivu, Southern Kivu, and Ituri.Of the 24 health zones, 77.1% of all cases were from only 6: Beni, Butembo, Katwa, Kalunguta, Mabalako, and Mandima. Only 9.7% of cases were under the age of 18. There is also a slightly larger proportion of females contracting the disease, comprising 57.0% of the cases. Approximately 5% of the cases were health care workers. About one-third of the EVD fatalities were not identified until patient’s death and thus not effectively isolated from the time of infection. Although over 170,000 contacts of confirmed and probable Ebola cases had been monitored across all affected health zones for 21 days after their last known exposure by the end of the epidemic, some of the contact tracing was incomplete due to insecurity that prevented public health response teams from entering some communities. The overall case density map is presented in panel (A) of Fig. 1 with the animated version of the map presented in the online appendix in Fig. A.1. Notice that the high-density areas, particularly Butembo, Katwa, and Beni, are all spatially small health zones corresponding to cities or towns with larger populations.Figure 1DRC Ebola dataset. (A) The spatial distribution of 3481 EVD cases across the northern DRC health zones during Ebola 2018–2020 outbreak. (B) The flowchart of personal records available up to September 12, 2019 available for the current analysis. The total number of available individual disease records was 3080. Map created using open software R17 with geospatial data obtained from18.Full size imageFigure 2Daily incidence and removal rates. Daily incidence (grey bars) and removal counts (red dots) during DRC Ebola 2018–2020 outbreak between August 15, 2018 and September 12, 2020 along with their respective trendlines (loess smoothers). The blue trendline above the plot represents daily effective reproduction number (mathcal{R}_t) defined as the ratio of daily number of new infections to new removals. The vertical lines indicate cut-off dates for data collection in each wave as listed in Table 1.Full size imageTable 1 Observed cases by EVD wave.Full size tableCase alerts and definitionsSince early August, 2018, the DRC Ministry of Health has been collaborating with several international partners to support and enhance EVD response activities through its emergency operations center in Goma. To the extent possible given regional security considerations19, the response teams were deployed to interview patients and their suspected contacts using a standardized case investigation form classifying cases as suspected, probable, or confirmed. A suspected case (whether surviving or not) was defined as one with the acute onset of fever (over 100(^{circ })F) and at least three Ebola-compatible clinical signs or symptoms (headache, vomiting, anorexia, diarrhea, lethargy, stomach pain, muscle or joint aches, difficulty swallowing or breathing, hiccups, unexplained bleeding, or any sudden, unexplained death) in a North Kivu, South Kivu, or Ituri resident or any person who had traveled to these provinces during this period and reported the signs or symptoms defined above. A patient who met the suspected case definition and died but from whom no specimens were available was considered a probable case. A confirmed Ebola case was defined as a suspected case with at least one positive test for Ebola virus using reverse transcription polymerase chain reaction (RT-PCR)20 testing. Patients with suspected Ebola were isolated and transported to an Ebola treatment center for confirmatory testing and treatment2.Onset and removalIn our analysis of the DRC dataset, we focused on dates of symptom onset and removal, with removal defined as either a death/recovery at home or transfer to an Ebola treatment center (ETC). It was assumed that, once in the treatment center, the probability of further infection spread by an isolated individual was very small due to the strict safety protocols—and later due also to vaccination of healthcare personnel and family members who were in contact with the suspected Ebola case. As summarized in panel (B) of Fig. 1, we were able to access 3117 out of 3481 individual records of confirmed and probable Ebola cases. Of these 3117 records, 37 were missing both the onset and recovery dates and were removed from further analysis. In about 30% of the remaining records, either their dates of onset or removal were missing. A detailed flow diagram summarizing the amount of missing data and data processing leading to the final dataset is presented in panel (B) of Fig. 1. The distribution of the original and the partially imputed records across the three waves of infection is provided for further reference in Table 1.Spatial and temporal patternsThroughout the pandemic, the incidence rates exhibited strong spatial and temporal patterns that can be summarized as three distinct waves of infections with approximate boundaries marked by vertical lines in Fig. 1. The distribution of weekly reported cases across the most affected health zones listed in Table 1 is provided in the bar plot and in the corresponding animation in the appendix (see Figure A.1). As seen from the bar chart and the animated plot, the epidemic was initially driven largely by infections in the health zones of Beni, Mandima and Mabalako. After several months, the incidence of new cases in these zones subsided, but the epidemic moved south to the health zones of Katwa and Butembo, where the majority of new infections was registered between weeks 22 to 45 of the epidemic (see Panel (A) in Figure A.1 in the online Appendix). In the final spatial shift, around week 49, the epidemic returned to the health zones of Beni, Mandima, and Mabalako, where it was mostly extinguished around week 60 (September 2019). Isolated Ebola incidences occurred sporadically across northern DRC until end of the outbreak was officially declared in June 2020.The empirical patterns of incidence and removal for EVD cases are summarized in Fig. 2 with the bar and the dot plots representing the daily numbers of new infections and removals, respectively. As seen from the plot, these daily counts closely follow a three-wave temporal pattern in Table 1. This is further evident from the black and red trendlines representing the loess smoothers (see21). The daily ratio of new cases and removals may be interpreted as a crude estimate of the effective reproduction number (mathcal{R}_t) defined more formally in (2) in Model for Data Analysis below. In particular, the blue trendline for (mathcal{R}_t) indicates that towards the end of the observed time period, the number of removals outpaced the number of new infections ((mathcal{R}_t 0) and (r_t = 0) where (beta > 0) is the rate of infection, (gamma > 0) is the rate of recovery and (rho > 0) is the initial amount of infection. In particular, the model implies the existence of the basic reproduction number (mathcal{R}_0) (R-naught), which determines the average speed of disease spread11 and is given by the formula$$mathcal{R}_0=beta /gamma .$$If (mathcal{R}_0 > 1), the proportion of infected initially rises and then subsides, with the final proposition of surviving susceptibles given by (s_infty = 1 – tau > 0) where (tau) is know as the epidemic’s final size. In typical statistical analysis, an estimate of (mathcal{R}_0) is obtained by separately estimating the parameters (beta) and (gamma). Another important quantity related to (1) is the effective reproduction number, which is typically defined as$$begin{aligned} mathcal{R}_t= mathcal{R}_0 s_t. end{aligned}$$
(2)
Although equation (1) is typically considered in the context of an average behavior of a large population, for our purposes we interpret it as defining the individual histories of infection and recovery, according to the idea of the dynamic survival analysis (DSA) discussed recently in10 and24 and also briefly summarized in the Appendix. With the DSA approach, we interpret equation (1) as the so-called stochastic master equation25 describing the change in probability of a randomly selected individual being at time t either susceptible, infected, or removed. These respective probabilities are represented by the scaled proportions (s_t/(1+rho )), (iota _t/(1+rho )), and (r_t/(1+rho )) and evolve according to (1). As outlined in10, the DSA-based interpretation of the classical SIR equations has a number of advantages that make it particularly convenient for analyzing epidemic data consisting of individual histories of infection onsets and removals, which is exactly the type of data available in the DRC Ebola dataset. The fact that the model is individual-based implies also that we can vary the parameters (theta =(beta ,gamma ,rho )) to account for individual covariates and changes in the parameter values over time, as different waves of infection sweep through the population. Finally, for the purpose of our analysis, it is also important to note that the DSA model does not require any knowledge of the size of the susceptible population subjected to the epidemic pressure. For the DRC dataset, that assumption would be difficult to justify due to spatial and temporal heterogeneity of the epidemic and the frequent movements of local populations driven by political conflicts and insecurity. Another element complicating the determination of the size of susceptible population was the ring vaccination campaign that has been conducted since 2019 wherever possible in the northern DRC during periods of relative stability, despite local mistrust and supply issues. This campaign ultimately resulted in over 250,000 vaccinations.Note that, because (s_0 = 1), the values of (mathcal{R}_0) and (mathcal{R}_t) coincide for (t = 0). Moreover, (s_t = exp left( -mathcal{R}_0 int _0^t r_u mathrm {d}u right)) is a decreasing function of time and therefore, so is (mathcal{R}_t). However, in practice, this implication is problematic. Rewriting (mathcal{R}_t = – {dot{s}}_t/ {dot{r}}_t) suggests that a crude but sensible way to estimate (mathcal{R}_t) empirically is to take the ratio of daily number of new infections to new removals. The empirical (mathcal{R}_t) thus estimated will not be necessarily monotonically decreasing. In the light of possibly changing parameters and the effective population size, we have adopted this approach to estimating the daily effective reproduction number (mathcal{R}_t) in Fig. 2.Parameter estimationWe assume that, for each of the three waves of the epidemic, we have a separate and independent set of parameters (theta) and that, in each wave, we observe (n_T) histories (records) of infection. The i-th individual history may be represented either by the times of disease onset and removal ((t_i,T_i)) or by (t_i) or (T_i) times alone ((t_i,circ )) or ((circ ,T_i)) ((circ) denoting missing value). We assume that among the available (n_T) histories we have n complete records ((t_i,T_i)), (n_1) incomplete ones ((t_i,circ )) and (n_2) incomplete ones ((circ ,T_i )). The wave-specific DSA likelihood function for n complete data records is (see Appendix)$$begin{aligned} begin{aligned} {mathcal {L}}_C(theta vert t_1ldots ,t_n,T_1,ldots ,T_n,T)=(s_T-1)^{-n}prod _{i=1}^n {dot{s}}_{t_i}gamma ^{w_i}e^{-gamma (T_i wedge T -t_i)} end{aligned} end{aligned}$$
(3)
where T is the available time horizon and (w_i) is the binary variable indicating whether (T_i) is right-censored (that is, (T_iwedge T =T)) in which case (w_i = 0) and otherwise (w_i = 1). For the remaining (n_1+n_2) records that are partially incomplete, the wave-specific DSA likelihood function is$$begin{aligned} begin{aligned} {mathcal {L}}_I(theta vert t_1ldots ,t_{n_1},T_1,ldots ,T_{n_2},T)= (s_T-1)^{-(n_1+n_2)} gamma ^{n_2}prod _{i=1}^{n_1} {dot{s}}_{t_i} prod _{i=1}^{n_2} (rho e^{-gamma T_i }-iota _{T_i}) end{aligned} end{aligned}$$
(4)
where we assume that (T_i1). Given the wave-specific time horizons (T’s), the set of parameters for each epidemic wave was estimated independently using 2 independent chains of 3000 iterations, with a burn-in period of 1000 iterations. The chains’ convergence assessed using Rubin’s R statistic28. The analysis resulted in approximate samples from the posterior distribution of (theta) for each of the three waves of the epidemic (see e.g., Fig. 4).Ethics statement on human subjects and methodsThe research was conducted in accordance with the relevant guidelines and regulations of the US law and OSU Institutional Review Board. The research activities involving human subjects discussed in the paper meet the US federal exemption criteria under 45 CFR 46 and 21 CFR 56. More

Emlen, S. T. & Oring, L. W. Ecology, sexual selection, and evolution of mating systems. Science 197(4300), 215–223 (1977).ADS 
CAS 
Article 

Google Scholar 
Lagos, V. O., Bozinovic, F. & Contreras, L. C. Microhabitat use by a small diurnal rodent (Octodon degus) in a semiarid environment: Thermoregulatory constraints or predation risk? J. Mammal. 76(3), 900–905 (1995).Article 

Google Scholar 
Lagos, V. O., Contreras, L. C., Meserve, P. L., Gutiérrez, J. R. & Jaksic, F. M. Effects of predation risk on space use by small mammals: A field experiment with a neotropical rodent. Oikos 74, 259–264 (1995).Article 

Google Scholar 
Schradin, C. & Pillay, N. Female striped mice (Rhabdomys pumilio) change their home ranges in response to seasonal variation in food availability. Behav. Ecol. 17(3), 452–458. https://doi.org/10.1093/beheco/arj047 (2006).Article 

Google Scholar 
Hayes, L. D., Chesh, A. S. & Ebensperger, L. A. Ecological predictors of range areas and use of burrow systems in the diurnal rodent, Octodon degus. Ethology 113, 155–165. https://doi.org/10.1111/j.1439-0310.2006.01305.x (2007).Article 

Google Scholar 
Brivio, F. et al. Forecasting the response to global warming in a heat-sensitive species. Sc. Rep. 9, 3048. https://doi.org/10.1038/s41598-019-39450-5 (2019).ADS 
CAS 
Article 

Google Scholar 
Santamaría, A. E., Olea, P. P., Vinuela, J. & Garcia, J. T. Spatial and seasonal variation in occupation and abundance of common vole burrows in highly disturbed agricultural ecosystems. Eur. J. Wildl. Res. 65, 52. https://doi.org/10.1007/s10344-019-1286-2 (2019).Article 

Google Scholar 
Kinlaw, A. A review of burrowing by semi-fossorial vertebrates in arid environments. J. Arid Environ. 41, 127–145 (1999).ADS 
Article 

Google Scholar 
Daly, M., Beherends, P. R. & Wilson, M. I. Activity patterns of kangaroo rats—Granivores in a desert habitat. In Activity Patterns in Small Mammals: An Ecological Approach (eds Halle, S. & Stenseth, N. C.) 145–158 (Springer, 2000).Chapter 

Google Scholar 
Mackin-Rogalska, R., Adamczewska-Andrzejewska, K. & Nabaglo, L. Common vole numbers in relation to the utilization of burrow system. Acta Theriol. 31(2), 17–44 (1986).Article 

Google Scholar 
Powell, R. A. & Fried, J. J. Helping by juvenile pine voles (Microtus pinetorum), growth and survival of younger siblings, and the evolution of pine vole sociality. Behav. Ecol. 3, 325–333 (1992).Article 

Google Scholar 
Randall, J. A., Rogovin, K., Parker, P. G. & Eimes, J. A. Flexible social structure of a desert rodent, Rhombomys opimus: Philopatry, kinship, and ecological constraints. Behav. Ecol. 16, 961–973 (2005).Article 

Google Scholar 
Ebensperger, L. A. et al. Burrow limitations and group living in the communally rearing rodent, Octodon degus. J. Mammal. 92(1), 21–30 (2011).Article 

Google Scholar 
Santini, L. The habits and influence on the environment of the old world porcupine Hystrix cristata L. in the northernmost part of its range. In Proc. 9th Vertebrate Pest Conference, Vol. 34, 149–153 (1980).Felicioli, A., Grazzini, A. & Santini, L. The mounting and copulation behaviour of the crested porcupine Hystrix cristata. Ital. J. Zool. 64, 155–161 (1997).Article 

Google Scholar 
Felicioli, A., Grazzini, A. & Santini, L. The mounting behaviour of a pair of crested porcupine H. cristata L.. Mammalia 61(1), 123–126 (1997).
Google Scholar 
Felicioli, A. Analisi spazio-temporale dell’attività motoria in Hystrix cristata L. Dissertation, University of Pisa (1991).Felicioli, A. & Santini, L. Burrow entrance-hole orientation and first emergence time in the crested porcupine Hystrix cristata L.: Space-time dependence on sunset. Pol. Ecol. Stud. 20(3–4), 317–321 (1994).
Google Scholar 
Mori, E., Nourisson, D. H., Lovari, S., Romeo, G. & Sforzi, A. Self-defence may not be enough: Moonlight avoidance in a large, spiny rodent. J. Zool. 294, 31–40 (2014).Article 

Google Scholar 
Corsini, M. T., Lovari, S. & Sonnino, S. Temporal activity patterns of crested porcupine Hystrix cristata. J. Zool. Lond. 236, 43–54 (1995).Article 

Google Scholar 
Coppola, F., Vecchio, G. & Felicioli, A. Diurnal motor activity and “sunbathing” behaviour in crested porcupine (Hystrix cristata L., 1758). Sci. Rep. 9, 14283 (2019).ADS 
Article 

Google Scholar 
Pigozzi, G. Crested porcupines (Hystrix cristata) within badger setts (Meles meles) in the Maremma Natural Park, Italy. Saugetierk. Mitt. 33, 261–263 (1986).
Google Scholar 
Coppola, F. & Felicioli, A. Reproductive behaviour in free-ranging crested-porcupine Hystrix cristata L., 1758. Sci. Rep. 11, 20142. https://doi.org/10.1038/s41598-021-99819-3 (2021).ADS 
CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Monetti, L., Massolo, A., Sforzi, A. & Lovari, S. Site selection and fidelity by crested porcupines for denning. Ethol. Ecol. Evol. 17, 149–159 (2005).Article 

Google Scholar 
Coppola, F., Dari, C., Vecchio, G., Scarselli, D. & Felicioli, A. Co-habitation of settlements among Crested Porcupines (Hystrix cristata), Red Foxes (Vulpes vulpes) and European Badgers (Meles meles). Curr. Sci. 119(5), 817–822 (2020).Article 

Google Scholar 
De Villiers, M. S., Van Aarde, R. J. & Dott, H. M. Habitat utilization by the Cape porcupine Hystrix africaeaustralis in a savanna ecosystem. J. Zool. Lond. 232, 539–549 (1994).Article 

Google Scholar 
Corbet, N. U. & de Aarde, R. J. Social organization and space use in the Cape porcupine in a Southern African savanna. Afr. J. Ecol. 34, 1–14 (1996).Article 

Google Scholar 
Massolo, A., Dani, F. R. & Bella, N. Sexual and individual cues in the peri-anal gland secretum of crested porcupines (Hystrix cristata). Mamm. Biol. 74, 488–496 (2009).Article 

Google Scholar 
Mori, E. & Lovari, S. Sexual size monomorphism in the crested porcupine (Hystrix cristata). Mamm. Biol. 79, 157–160 (2014).Article 

Google Scholar 
Mori, E. et al. Patterns of spatial overlap in a monogamous large rodent, the crested porcupine. Behav. Process. 107, 112–118 (2014).Article 

Google Scholar 
Mukherjee, A., Pilakandy, R., Kumara, H. N., Manchi, S. S. & Bhupathy, S. Burrow characteristics and its importance in occupancy of burrow dwelling vertebrates in Semiarid area of Keoladeo National Park, Rajasthan, India. J. Arid Environ. 141, 7–15 (2017).ADS 
Article 

Google Scholar 
Mukherjee, A., Pal, A., Velankar, A. D., Kumara, H. N. & Bhupathy, S. Stay awhile in my burrow! Interspecific associations of vertebrates to Indian crested porcupine burrows. Ethol. Ecol. Evol. 3(4), 313–328 (2019).Article 

Google Scholar 
Fernandez, N. & Palomares, F. The selection of breeding dens by the endangered Iberian lynx (Lynx pardinus): Implications for its conservation. Biol. Conserv. 94, 51–61 (2000).Article 

Google Scholar 
Ross, S., Kamnitzer, R., Munkhtsog, B. & Harris, S. Den-site selection is critical for Pallas’s cats (Otocolobus manul). Can. J. Zool. 88(9), 905–913. https://doi.org/10.1139/Z10-056 (2010).Article 

Google Scholar 
Libal, N. S., Belant, J. L., Leopold, B. D., Wang, G. & Owen, A. Despotism and risk of infanticide influence grizzly bear den-site selection. PLoS ONE 6(9), e24133. https://doi.org/10.1371/journal.pone.0024133 (2011).ADS 
CAS 
Article 
PubMed 
PubMed Central 

Google Scholar 
Elbroch, L. M., Lendrum, P. E. & Quigley, H. Cougar den site selection in the Southern Yellowstone ecosystem. Mamm. Res. 60, 89–96. https://doi.org/10.1007/s13364-015-0212-6 (2015).Article 

Google Scholar 
Solomon, N. G., Christiansen, A. M., Kirk Lin, Y. & Hayes, L. D. Factors affecting nest location of prairie voles (Microtus ochrogaster). J. Mammal. 86(3), 555–560 (2005).Article 

Google Scholar 
Pereoglou, F. et al. Refuge site selection by the eastern chestnut mouse in recently burnt heath. Wildl. Res. 38(4), 290–298. https://doi.org/10.1071/WR11007 (2011).Article 

Google Scholar 
Grazzini, M. T. Comportamento riproduttivo e accrescimento post-natale in Hystrix cristata L. (Rodentia, Hystricidae). Dissertation, University of Pisa (1992).Capizzi, D. & Santini, L. Hystrix cristata Linnaeus, 1758. In Fauna d’Italia, Mammalia II: Erinaceomorpha, Soricomorpha, Lagomorpha, Rodentia (eds Amori, G. et al.) 695–706 (Edizione Calderini de il Sole 24 Ore, 2008).
Google Scholar 
Coppola, F. New knowledge tools for crested porcupine (Hystrix cristata L., 1758) management in the wild: First census model, new behavioural ecology aspects and preliminary investigation on health status. University of Pisa, PhD thesis (2021).Wood, S. N. Generalized Additive Models: An Introduction with R 2nd edn. (Chapman and Hall/CRC, 2017).Book 

Google Scholar 
Wood, S. N. A simple test for random effects in regression models. Biometrika 100, 1005–1010 (2013).MathSciNet 
Article 

Google Scholar 
Zuur, A. F., Ieno, E. N., Walker, N., Saveliev, A. A. & Smith, G. M. Mixed Effects Models and Extensions in Ecology with R (Springer, 2009).Book 

Google Scholar  More

Martin, P. S. & Klein, R. G. Quaternary extinctions: a prehistoric revolution. (University of Arizona Press, 1984).Waguespack, N. M. & Surovell, T. A. Clovis hunting strategies, or how to make out on plentiful resources. Am. Antiq. 68, 333–352 (2003).
Google Scholar 
Surovell, T. A., Pelton, S. R., Anderson-Sprecher, R. & Myers, A. D. Test of Martin’s overkill hypothesis using radiocarbon dates on extinct megafauna. Proc. Natl. Acad. Sci. 113, 886–891 (2016).CAS 
PubMed 
ADS 

Google Scholar 
Martin, P. S. Prehistoric overkill: the global model. In Quaternary extinctions: a prehistoric revolution (eds. Martin, P. S. & Klein, R. G.) 355–403 (University of Arizona Press, 1984).Barnosky, A. D. & Lindsey, E. L. Timing of Quaternary megafaunal extinction in South America in relation to human arrival and climate change. Quatern. Int. 217, 10–29 (2010).
Google Scholar 
Prescott, G. W., Williams, D. R., Balmford, A., Green, R. E. & Manica, A. Quantitative global analysis of the role of climate and people in explaining late Quaternary megafaunal extinctions. Proc. Natl. Acad. Sci. 109, 4527–4531 (2012).CAS 
PubMed 
PubMed Central 
ADS 

Google Scholar 
Sandom, C., Faurby, S., Sandel, B. & Svenning, J.-C. Global late Quaternary megafauna extinctions linked to humans, not climate change. Proc. R. Soc. B Biol. Sci. 281, 20133254 (2014).
Google Scholar 
Wolfe, A. L. & Broughton, J. M. A foraging theory perspective on the associational critique of North American Pleistocene overkill. J. Archaeol. Sci. 119, 105162 (2020).
Google Scholar 
Berger, J., Swenson, J. E. & Persson, I. L. Recolonizing carnivores and naïve prey: Conservation lessons from pleistocene extinctions. Science 291, 1036–1039 (2001).CAS 
PubMed 
ADS 

Google Scholar 
Brook, B. W. & Bowman, D. M. J. S. The uncertain blitzkrieg of Pleistocene megafauna. J. Biogeogr. 31, 517–523 (2004).
Google Scholar 
Johnson, C. N. Determinants of loss of mammal species during the Late Quaternary ‘megafauna’ extinctions: life history and ecology, but not body size. Proc. R. Soc. London. Ser. B Biol. Sci. 269, 2221–2227 (2002).CAS 

Google Scholar 
Bourgon, N. et al. Trophic ecology of a Late Pleistocene early modern human from tropical Southeast Asia inferred from zinc isotopes. J. Hum. Evol. 161, 103075 (2021).PubMed 

Google Scholar 
Meltzer, D. J. Overkill, glacial history, and the extinction of North America’s Ice Age megafauna. Proc. Natl. Acad. Sci. 117, 28555–28563 (2020).CAS 
PubMed 
PubMed Central 

Google Scholar 
Stewart, M., Carleton, W. C. & Groucutt, H. S. Climate change, not human population growth, correlates with Late Quaternary megafauna declines in North America. Nat. Commun. 12, 965 (2021).CAS 
PubMed 
PubMed Central 
ADS 

Google Scholar 
Nogués-Bravo, D., Rodríguez, J., Hortal, J., Batra, P. & Araújo, M. B. Climate change, humans, and the extinction of the woolly mammoth. PLoS Biol. 6, e79 (2008).PubMed 
PubMed Central 

Google Scholar 
Koch, P. L. & Barnosky, A. D. Late quaternary extinctions: State of the debate. Annu. Rev. Ecol. Evol. Syst. 37, 215–250 (2006).
Google Scholar 
Cardillo, M. Multiple causes of high extinction risk in large mammal species. Science 309, 1239–1241 (2005).CAS 
PubMed 
ADS 

Google Scholar 
Meiri, S. & Liang, T. Rensch’s rule—Definitions and statistics. Glob. Ecol. Biogeogr. 30, 573–577 (2021).
Google Scholar 
Lyons, S. K. et al. The changing role of mammal life histories in Late Quaternary extinction vulnerability on continents and islands. Biol. Lett. 12, 20160342 (2016).PubMed 
PubMed Central 

Google Scholar 
Alroy, J. A multispecies overkill simulation of the end-pleistocene megafaunal mass extinction. Science 292, 1893–1896 (2001).CAS 
PubMed 
ADS 

Google Scholar 
Smaers, J. B. et al. The evolution of mammalian brain size. Sci. Adv. 7, 1–12 (2021).
Google Scholar 
Jerison, H. J. Evolution of the Brain and Intelligence (Academic Press, 1973). https://doi.org/10.2307/4512058.Book 

Google Scholar 
Sol, D., Bacher, S., Reader, S. M. & Lefebvre, L. Brain size predicts the success of mammal species introduced into novel environments. Am. Nat. 172, S63–S71 (2008).PubMed 

Google Scholar 
Møller, A. P. & Erritzøe, J. Brain size in birds is related to traffic accidents. R. Soc. Open Sci. 4, 161040 (2017).PubMed 
PubMed Central 
ADS 

Google Scholar 
Sayol, F., Sol, D. & Pigot, A. L. Brain size and life history interact to predict urban tolerance in birds. Front. Ecol. Evol. 8, 58 (2020).
Google Scholar 
Budd, G. E. & Jensen, S. The origin of the animals and a ‘Savannah’ hypothesis for early bilaterian evolution. Biol. Rev. 92, 446–473 (2017).PubMed 

Google Scholar 
Benoit, J. et al. Brain evolution in Proboscidea (Mammalia, Afrotheria) across the Cenozoic. Sci. Rep. 9, 9323 (2019).PubMed 
PubMed Central 
ADS 

Google Scholar 
Møller, A. P. & Erritzøe, J. Brain size and the risk of getting shot. Biol. Lett. 12, 20160647 (2016).PubMed 
PubMed Central 

Google Scholar 
Di Febbraro, M. et al. Does the jack of all trades fare best? Survival and niche width in Late Pleistocene megafauna. J. Biogeogr. 44, 2828–2838 (2017).
Google Scholar 
Morris, S. D., Kearney, M. R., Johnson, C. N. & Brook, B. W. Too hot for the devil? Did climate change cause the mid-Holocene extinction of the Tasmanian devil Sacrophilus harrisii from mainland Australia? Ecography 2022, (2022).Fillios, M., Crowther, M. S. & Letnic, M. The impact of the dingo on the thylacine in Holocene Australia. World Archaeol. 44, 118–134 (2012).
Google Scholar 
González-Lagos, C., Sol, D. & Reader, S. M. Large-brained mammals live longer. J. Evol. Biol. 23, 1064–1074 (2010).PubMed 

Google Scholar 
Barton, R. A. & Capellini, I. Maternal investment, life histories, and the costs of brain growth in mammals. Proc. Natl. Acad. Sci. U.S.A. 108, 6169–6174 (2011).CAS 
PubMed 
PubMed Central 
ADS 

Google Scholar 
Abelson, E. S. Brain size is correlated with endangerment status in mammals. Proc. R. Soc. B Biol. Sci. 283, 20152772 (2016).
Google Scholar 
Gonzalez-Voyer, A., González-Suárez, M., Vilà, C. & Revilla, E. Larger brain size indirectly increases vulnerability to extinction in mammals. Evolution (N.Y.) 70, 1364–1375 (2016).
Google Scholar 
Ives, A. R. & Helmus, M. R. Generalized linear mixed models for phylogenetic analyses of community structure. Ecol. Monogr. 81, 511–525 (2011).
Google Scholar 
Castiglione, S. et al. A new method for testing evolutionary rate variation and shifts in phenotypic evolution. Methods Ecol. Evol. 9, 974–983 (2018).
Google Scholar 
Billet, G. Phylogeny of the Notoungulata (Mammalia) based on cranial and dental characters. J. Syst. Palaeontol. 9, 481–497 (2011).
Google Scholar 
Shultz, S., Bradbury, R. B., Evans, K. L., Gregory, R. D. & Blackburn, T. M. Brain size and resource specialization predict long-term population trends in British birds. Proc. R. Soc. B Biol. Sci. 272, 2305–2311 (2005).
Google Scholar 
Ducatez, S., Sol, D., Sayol, F. & Lefebvre, L. Behavioural plasticity is associated with reduced extinction risk in birds. Nat. Ecol. Evol. 4, 788–793 (2020).PubMed 

Google Scholar 
Abelson, E. S. Big brains reduce extinction risk in Carnivora. Oecologia 191, 721–729 (2019).PubMed 
ADS 

Google Scholar 
Lundgren, E. J. et al. Introduced herbivores restore Late Pleistocene ecological functions. Proceedings of the National Academy of Sciences 117, 7871–7878 (2020).Article 
PubMed 
PubMed Central 

Google Scholar 
Shultz, S. & Dunbar, R. Encephalization is not a universal macroevolutionary phenomenon in mammals but is associated with sociality. Proceedings of the National Academy of Sciences 107, 21582–21586 (2010).Article 
PubMed 
PubMed Central 

Google Scholar 
Gould, S. J. & Vrba, E. S. Exaptation—A missing term in the science of form. Paleobiology 8, 4–15 (1982).
Google Scholar 
Wroe, S. et al. Climate change frames debate over the extinction of megafauna in Sahul (Pleistocene Australia-New Guinea). Proc. Natl. Acad. Sci. U.S.A. 110, 8777–8781 (2013).CAS 
PubMed 
PubMed Central 
ADS 

Google Scholar 
Barnosky, A. D., Koch, P. L., Feranec, R. S., Wing, S. L. & Shabel, A. B. Assessing the Causes of Late Pleistocene Extinctions on the Continents. Science 306, 70–75 (2004).Article 
PubMed 

Google Scholar 
Profico, A., Buzi, C., Melchionna, M., Veneziano, A. & Raia, P. Endomaker, a new algorithm for fully automatic extraction of cranial endocasts and the calculation of their volumes. Am. J. Phys. Anthropol. 172, 511–515 (2020).PubMed 

Google Scholar 
Damuth, J. & Macfadden, B. J. Body Size in Mammalian Paleobiology: Estimation and Biological Implications (Cambridge University Press, 1990).
Google Scholar 
Zagwijn, W. H. The beginning of the Ice Age in Europe and its major subdivisions. Quatern. Sci. Rev. 11, 583–591 (1992).ADS 

Google Scholar 
Hearty, P. J., Hollin, J. T., Neumann, A. C., O’Leary, M. J. & McCulloch, M. Global sea-level fluctuations during the Last Interglaciation (MIS 5e). Quatern. Sci. Rev. 26, 2090–2112 (2007).ADS 

Google Scholar 
Ashwell, K. W. S., Hardman, C. D. & Musser, A. M. Brain and behaviour of living and extinct echidnas. Zoology 117, 349–361 (2014).PubMed 

Google Scholar 
Castiglione, S. et al. The influence of domestication, insularity and sociality on the tempo and mode of brain size evolution in mammals. Biol. J. Linn. Soc. 132, 221–231 (2021).
Google Scholar 
Wilkins, A. S., Wrangham, R. W. & Tecumseh Fitch, W. The ‘domestication syndrome’ in mammals: A unified explanation based on neural crest cell behavior and genetics. Genetics 197, 795–808 (2014).PubMed 
PubMed Central 

Google Scholar 
Sayol, F., Steinbauer, M. J., Blackburn, T. M., Antonelli, A. & Faurby, S. Anthropogenic extinctions conceal widespread evolution of flightlessness in birds. Sci. Adv. 6, eabb6095 (2020).PubMed 
PubMed Central 
ADS 

Google Scholar 
Fromm, A., Meiri, S. & McGuire, J. Big, flightless, insular and dead: Characterising the extinct birds of the Quaternary. J. Biogeogr. 48(9), 2350–2359. https://doi.org/10.1111/jbi.14206 (2021).Article 

Google Scholar 
Meiri, S., Dayan, T. & Simberloff, D. The generality of the island rule reexamined. J. Biogeogr. 33, 1571–1577 (2006).
Google Scholar 
Larramendi, A. & Palombo, M. R. Body Size, Structure, Biology and Encephalization Quotient of Palaeoloxodon ex gr. P. falconeri from Spinagallo Cave (Hyblean plateau, Sicily). Hystrix, the Italian Journal of Mammalogy 26, 102–109 (2015).Article 

Google Scholar 
Slavenko, A., Tallowin, O. J. S., Itescu, Y., Raia, P. & Meiri, S. Late Quaternary reptile extinctions: Size matters, insularity dominates. Glob. Ecol. Biogeogr. 25, 1308–1320 (2016).
Google Scholar 
Tracy, C. R. & George, T. L. On the determinants of extinction. Am. Nat. 139, 102–122 (1992).
Google Scholar 
Manne, L. L., Brooks, T. M. & Pimm, S. L. Relative risk of extinction of passerine birds on continents and islands. Nature 399, 258–261 (1999).CAS 
ADS 

Google Scholar 
Turvey, S. T. In the shadow of the megafauna: prehistoric mammal and bird extinctions across the Holocene. in Holocene Extinctions 17–40 (Oxford University Press, 2009). https://doi.org/10.1093/acprof:oso/9780199535095.003.0002Ebinger, P. A cytoarchitectonic volumetric comparison of brains in wild and domestic sheep. Zeitschrift für Anat. und Entwicklungsgeschichte 144, 267–302 (1974).CAS 

Google Scholar 
Röhrs, M. & Ebinger, P. Welche quantitativen beziehungen bestehen bei säugetieren zwischen schädelkapazität und hirnvolumen? Mammalian Biology 66, 102–110 (2001).Köhler, M. & Moyà-Solà, S. Reduction of brain and sense organs in the fossil insular bovid Myotragus. Brain Behav. Evol. 63, 125–140 (2004).PubMed 

Google Scholar 
de Bello, F. et al. On the need for phylogenetic ‘corrections’ in functional trait-based approaches. Folia Geobot. 50, 349–357 (2015).
Google Scholar 
Bates, D., Sarkar, D., Bates, M. D. & Matrix, L. The lme4 Package. October (2007).Kuznetsova, A., Brockhoff, P. B. & Christensen, R. H. B. lmerTest Package: Tests in linear mixed effects models. J. Stat. Softw. 82, 1–26 (2017).
Google Scholar 
Raia, P. & Meiri, S. The tempo and mode of evolution: Body sizes of island mammals. Evolution 65, 1927–1934 (2011).

Google Scholar 
Montgomery, S. H. et al. The evolutionary history of cetacean brain and body size. Evolution 67, 3339–3353 (2013).
PubMed 

Google Scholar 
Li, D., Dinnage, R., Nell, L. A., Helmus, M. R. & Ives, A. R. phyr: An r package for phylogenetic species-distribution modelling in ecological communities. Methods Ecol. Evol. 11, 1455–1463 (2020).
Google Scholar 
Melchionna, M. et al. Macroevolutionary trends of brain mass in Primates. Biological Journal of the Linnean Society 129, 14–25 (2020).Article 

Google Scholar 
Serio, C. et al. Macroevolution of toothed whales exceptional relative brain size. Evol. Biol. 46, 332–342 (2019).
Google Scholar 
Wickham, H. et al. Welcome to the Tidyverse. Journal of Open Source Software 4, 1686 (2019).Barton, K. Package ‘MuMIn’ Title Multi-Model Inference. CRAN-R (2018). More

Forest Ecology and Forest Management Group, Wageningen University, Wageningen, the NetherlandsPieter A. Zuidema & Ute Sass-KlaassenSchool of Natural Resources and the Environment, University of Arizona, Tucson, AZ, USAFlurin BabstLaboratory of Tree-Ring Research, University of Arizona, Tucson, AZ, USAFlurin Babst, Valerie Trouet, Zakia Hassan Khamisi, Paul R. Sheppard & Ramzi TouchanDepartment of Plant Biology, Institute of Biology, University of Campinas (UNICAMP), Campinas, BrazilPeter Groenendijk & José Roberto Vieira AragãoWorld Agroforestry Centre (ICRAF), Addis Ababa, EthiopiaAbrham AbiyuDepartment of Microbiology and Parasitology, Universidad Nacional Autónoma de México, Mexico City, MexicoRodolfo Acuña-SotoLaboratory of Protection and Forest Management, Department of Forest Engineering, Universidade Regional de Blumenau, Santa Catarina, BrazilEduardo Adenesky-FilhoDepartment of Biology, Wilfrid Laurier University, Waterloo, Ontario, CanadaRaquel Alfaro-SánchezDepartment of Forest Sciences, Luiz de Queiroz College of Agriculture, University of Sao Paulo, Piracicaba, BrazilGabriel Assis-Pereira, Claudia Fontana & Mario Tomazello-FilhoTree-Ring Laboratory, Forest Science Department, Federal University of Lavras, Lavras, BrazilGabriel Assis-Pereira & Ana Carolina BarbosaCAS Key Laboratory of Tropical Forest Ecology, Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences, Mengla, ChinaXue Bai, Ze-Xin Fan, Shankar Panthi & Zhe-Kun ZhouDepartment of Environmental, Biological and Pharmaceutical Sciences and Technologies, University of Campania “L. Vanvitelli”, Caserta, ItalyGiovanna BattipagliaService of Wood Biology, Royal Museum for Central Africa, Tervuren, BelgiumHans Beeckman, Camille Couralet & Benjamin ToirambeBrazilian Agricultural Research Corporation (Embrapa), Embrapa Forestry, Colombo, BrazilPaulo Cesar BotossoU.S. Department of Agriculture, Forest Service, NWCG Member Agency, Washington, DC, USATim BradleyInstitute of Geography, Friedrich-Alexander-University Erlangen-Nuremberg, Erlangen, GermanyAchim Bräuning, Mahmuda Islam, Mulugeta Mokria & Mizanur RahmanSchool of Geography, University of Leeds, Leeds, UKRoel Brienen & Emanuel GloorLamont-Doherty Earth Observatory, Columbia University, Palisades, NY, USABrendan M. Buckley & Rosanne D’ArrigoInstituto Pirenaico de Ecología (IPE-CSIC), Zaragoza, SpainJ. Julio CamareroCentre for Functional Ecology, Department of Life Sciences, Faculty of Sciences and Technology, University of Coimbra, Coimbra, PortugalAna Carvalho & Cristina NabaisDepartment of Botany, Institute of Biosciences, University of São Paulo, São Paulo, BrazilGregório Ceccantini, Bruno Barçante Ladvocat Cintra & Giuliano Maselli LocosselliInstituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias (INIFAP), Centro Nacional de Investigación Disciplinaría en Relación Agua-Suelo-Planta-Atmósfera (CENID-RASPA), Gómez Palacio, MéxicoLibrado R. Centeno-Erguera, Julián Cerano-Paredes & Jose Villanueva-DiazInstituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias (INIFAP), Campo Experimental Centro – Altos de Jalisco, Tepatitlán de Morelos, MéxicoÁlvaro Agustín Chávez-DuránDepartment of Geosciences, University of Arkansas, Fayetteville, AR, USAMalcolm K. Cleaveland & Daniela Granato-SouzaDepartment of Forest Sciences, Universidad Nacional de Colombia – Sede Medellín, Medellín, ColombiaJorge Ignacio del ValleMaster School for Carpentry and Cabinetmaking, Ebern, GermanyOliver DünischDepartment of Ecology and Evolutionary Biology, University of Arizona, Tucson, AZ, USABrian J. EnquistSanta Fe Institute, Santa Fe, NM, USABrian J. EnquistDepartment of Biological Sciences, University of Joinville Region ‐ UNIVILLE, Joinville, BrazilKarin Esemann-QuadrosPostgraduate Program in Forestry, Regional University of Blumenau – FURB, Blumenau, BrazilKarin Esemann-QuadrosCollege of Life Science, Climate Science Center and Department of Earth Science, Addis Ababa University, Addis Ababa, EthiopiaZewdu EshetuDepartamento de Dendrocronología e Historia Ambiental, IANIGLA, CCT-CONICET-Mendoza, Mendoza, ArgentinaM. Eugenia Ferrero, Lidio Lopez, Fidel Alejandro Roig & Ricardo VillalbaLaboratorio de Dendrocronología, Universidad Continental, Huancayo, PerúM. Eugenia Ferrero, Janet G. Inga & Edilson Jimmy Requena-RojasDepartment of Crop Sciences, Tropical Plant Production and Agricultural Systems Modelling, Göttingen University, Göttingen, GermanyEsther FichtlerInstitute of Pacific Islands Forestry, USDA Forest Service Pacific Southwest Research Station, Hilo, HI, USAKainana S. Francisco & Mulugeta MokriaWorld Agroforestry Centre (ICRAF), Nairobi, KenyaAster GebrekirstosFlanders Heritage Agency, Brussels, BelgiumKristof HanecaDepartment of Geography and Geological Sciences, University of Idaho, Moscow, ID, USAGrant Logan HarleyGerman Archaeological Institute DAI, Berlin, GermanyIngo HeinrichGeography Department, Humboldt University Berlin, Berlin, GermanyIngo HeinrichGFZ German Research Centre for Geosciences, Potsdam, GermanyIngo Heinrich & Gerd HelleDepartment of Forestry and Environmental Science, Shahjalal University of Science and Technology, Sylhet, BangladeshMahmuda Islam & Mizanur RahmanFaculty of Forestry and Wood Sciences, Czech University of Life Sciences Prague, Prague, Czech RepublicYu-mei JiangUS Fish and Wildlife Service, Albuquerque, NM, USAMark KaibDepartment of Ecology and Biogeography, Faculty of Biological and Veterinary Sciences, Nicolaus Copernicus University, Toruń, PolandMarcin KoprowskiCentre for Climate Change Research, Nicolaus Copernicus University, Toruń, PolandMarcin KoprowskiWater Systems and Global Change Group, Wageningen University and Research, Wageningen, the NetherlandsBart KruijtInstituto Nacional de Innovación Agraria, Programa Nacional de Investigación Forestal, Huancayo, PerúEva LaymeEnvironmental Systems Analysis Group, Wageningen University and Research, Wageningen, the NetherlandsRik LeemansDepartment of Natural Resource Management, South Dakota State University, Brookings, USA, SDA. Joshua LefflerLaboratory of Plant Anatomy and Dendrochronology, Department of Biology, Universidade Federal de Sergipe, Sergipe, BrazilClaudio Sergio Lisi, Mariana Alves Pagotto & Adauto de Souza Ribeiro Department of Geography, Swansea University, Swansea, UKNeil J. Loader & Iain RobertsonDepartamento Forestal, Universidad Autónoma Agraria Antonio Narro, Saltillo, MexicoMaría I. López-HernándezCITAB – Department of Forestry Sciences and Landscape (CIFAP), University of Trás-os-Montes and Alto Douro, Vila Real, PortugalJosé Luís Penetra Cerveira LousadaEscuela de Ciencias Biológicas, Universidad Pedagógica y Tecnológica de Colombia (UPTC), Tunja, ColombiaHooz A. MendivelsoBrazilian Agricultural Research Corporation (Embrapa), Embrapa Amazônia Ocidental, Manaus, BrazilValdinez Ribeiro MontóiaIHE Delft, Delft, the NetherlandsEddy MoorsVU University Amsterdam, Amsterdam, the NetherlandsEddy MoorsDepartment of Biomaterials Science and Technology, School of Natural Resources, The Copperbelt University, Kitwe, ZambiaJustine NgomaLaboratory of Ecology and Dendrology of the Federal Institute of Sergipe, São Cristovão, BrazilFrancisco de Carvalho Nogueira JúniorLaboratory of Plant Ecology, Universidade do Vale do Rio dos Sinos (UNISINOS), São Leopoldo, BrazilJuliano Morales Oliveira & Gabriela Morais OlmedoBIOAPLIC, Departamento de Botánica, Universidade de Santiago de Compostela, EPSE, Lugo, SpainGonzalo Pérez-De-LisLaboratorio de Dendrocronología, Carrera de Ingeniería Forestal, Universidad Nacional de Loja, Loja, EcuadorDarwin Pucha-CofrepFaculty of Environment and Resource studies, Mahidol University, Nakhon Pathom, ThailandNathsuda PumijumnongFacultad de Ciencias Agrarias, Universidad del Cauca, Popayán, ColombiaJorge Andres RamirezHémera Centro de Observación de la Tierra, Escuela de Ingeniería Forestal, Facultad de Ciencias, Universidad Mayor, Santiago, ChileFidel Alejandro Roig & Alejandro Venegas-GonzálezInstituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias (INIFAP), Centro de Investigación Regional Pacífico Centro – Campo Experimental, Centro Altos de Jalisco, MéxicoErnesto Alonso Rubio-CamachoNational Institute for Amazon Research, Petrópolis, Manaus, BrazilJochen SchöngartDepartment of Earth Sciences, Freie Universität Berlin, Berlin, GermanyFranziska SlottaDepartment of Earth and Environmental Systems, Indiana State University, Terre Haute, IN, USAJames H. SpeerDepartment of Geography, University of Alabama, Tuscaloosa, AL, USAMatthew D. TherrellDepartment of Civil, Environmental and Geodetic Engineering, The Ohio State University, Columbus, OH, USAMax C. A. TorbensonDepartment of Geography, Johannes Gutenberg University, Mainz, GermanyMax C. A. TorbensonDepartment of Plant and Environmental Sciences, School of Natural Resources, The Copperbelt University, Kitwe, ZambiaRoyd VinyaForest and Nature Management, Van Hall Larenstein University of Applied Sciences, Velp, the NetherlandsMart VlamSchool of Teacher Training for Secondary Education Tilburg, Fontys University of Applied Sciences, Tilburg, the NetherlandsTommy WilsP.A.Z., P.G. and V.T. initiated the tropical tree-ring network; P.A.Z., F.B., P.G. and V.T. designed the study; all co-authors except F.B. contributed tree-ring data; F.B. and P.G. analysed the data, with important contributions from P.A.Z.; P.A.Z. and V.T. wrote the manuscript, with important contributions from F.B. and P.G. All co-authors read and approved the manuscript. More

Study systemThe cuckoo catfish (Synodontis multipunctatus) belongs to the African catfish family Mochokidae. The genus Synodontis, with 131 species distributed across African freshwaters57, gave rise to a small radiation in Lake Tanganyika, with 10 described endemic species58. The taxonomy of the group is not well established59 and we use the name S. multipunctatus as this species is confirmed as a brood parasite30 and the name was used in previous studies4,30,32,37,42. Cuckoo catfish primarily parasitise mouthbrooding cichlids from the tribe Tropheini30, but species from other lineages can also be parasitised59.Experimental designAll experiments took place between January and August 2020 at the Institute of Vertebrate Biology, Czech Republic. Prior to experimental use, fish were housed in mixed-sex groups in tanks equipped with shelter and internal filtration. Cuckoo catfish were F1 generation of commercially imported wild-caught parents (10 pairs). Host cichlids were descendant of wild fish imported from Kalambo, Zambia. Experimental tanks (420 L; length 150 cm, depth 70 cm, height 40 cm) were equipped with internal filtration, fine gravel (2–4 mm diameter), half a clay pot as a shelter on each side of the tank, and one artificial plant in the centre of each tank. Water temperature was maintained at 27 °C (±1 °C) and the dark – light regime was set to 11 h:13 h. In total, we stocked 18 tanks with 4 males and 12 females of the mouthbrooding cichlid Astatotilapia burtoni and introduced 3 cuckoo catfish pairs of one of three different experience levels. Naïve catfish (n = 36 individuals) had no prior experience with cichlids. Experienced catfish (n = 36) were housed together with reproductive cichlids for 12 months prior to the experiment and were age-matched to naïve catfish (5 years old). Highly experienced catfish (n = 36) were raised, coexisted and reproduced with cichlids for 7 years (and were on average 7–8% larger than both naïve and experienced catfish; mean ± SE, naïve: 116.2 ± 1.9 mm, experienced: 117.1 ± 1.5 mm, highly experienced: 125.6 ± 1.4 mm; Linear Model (LM): experienced vs. highly experienced, estimate ± S.E = 8.44 ± 2.29, t = 3.68, P = 0.0004, experienced vs. naïve, estimate ± S.E = −0.94 ± 2.29, t = −0.41, P = 0.681, n = 108). Additionally, both naïve and experienced cuckoo catfish were bred using in-vitro fertilisation32 to avoid cichlid imprinting (i.e., priming with cichlid cues), while highly experienced catfish were bred under natural conditions within the buccal cavities of their hosts. Each experimental tank contained catfish with the same experience level. Due to space limitations, we split the experiment into two consecutive phases with 3 replicate tanks for each treatment within both phases (in total 9 experimental tanks per phase). Between the two experimental phases, host cichlids were placed together and haphazardly assigned to new experimental tanks. During the second phase, we removed some cichlids from the tanks because of injuries caused by their intraspecific aggression (3 males and 3 females in total), and those hosts were not replaced. Over an experimental phase, cuckoo catfish and cichlids freely interacted for 15–16 weeks. During this period, each tank was checked for mouthbrooding cichlids twice each week (Tuesday and Friday). We caught the mouthbrooding females, gently washed the eggs out of their mouths using a jet of water from a Pasteur pipette, measured their body size to the nearest mm, and released them back to their experimental tank. For each female, we counted the number of cichlid eggs and cuckoo catfish eggs (if present). At the end of each experimental phase, we measured body size of all cuckoo catfish to the nearest mm. There was no significant difference between the number of cichlid spawnings between naïve and experienced catfish treatments (Generalised Linear Models with negative binomial error distribution, estimate ± S.E.: −0.093 ± 0.145, z = −0.644, P = 0.519), nor between naïve and highly experienced catfish (estimate ± S.E.: −0.269 ± 0.148, z = −1.810, P = 0.070).Behavioural recordingOver the experimental period, we successfully recorded 18 videos of spawning events (Lamax x3.1 ATLAS cameras; naïve catfish treatment, n = 9; experienced catfish treatment, n = 6; highly experienced catfish treatment, n = 3). One camera was placed near the spawning site approximately 20 cm away from spawning activity and a second camera was placed outside the experimental tank to obtain an overall view. Nine spawnings were recorded from the start (n = 7 naïve catfish experiments and 2 experienced catfish experiments) and nine spawnings were recorded from the timepoint when we recognised ongoing spawning activity (n = 2 naïve, 4 experienced, and 3 highly experienced catfish experiments). From the video footage taken for each spawning, we scored all overt aggression that host cichlids directed towards cuckoo catfish, counted the number of intruding catfish during each distinct cichlid spawning behaviour (i.e., male and female cichlid interact in a repeated succession of quivering and T-positions), measured the delay of intruding catfish to each distinct spawning behaviour (i.e., the time from the start of spawning behaviour until the first catfish directly approaches the spawning cichlids), and recorded the presence or absence of catfish during each spawning behaviour. Additionally, we recorded whether cichlids used the available shelters for spawning as this might have impeded catfish recognition of the spawning activity. When spawning was recorded from the start, scoring started 100 s before we detected the first egg laid (cichlid or cuckoo catfish). When spawning was already ongoing, the scoring started immediately after the cameras were in place. Mounting of the cameras did not interrupt the normal behaviour of cichlids or catfish. For all video footage, scoring ended 100 s after the last male-female interaction within the spawning site. To estimate the duration of male T-positions during spawnings, we measured the time period from the start of male nuzzling near female genital papilla until the female turned around either to collect eggs or start nuzzling near the male´s genital papilla (n = 115 male T-positions from 12 cichlid spawnings).Statistical analysisWe used R v. 3.5.1 (R Development Core Team, 2018) for all statistical analyses. All statistical tests were two-sided. First, we compared body size among the three cuckoo catfish experience levels using a Linear Model with catfish size (mm) as response variable and ‘treatment’ (naïve, experienced, and highly experienced catfish) as predictor variable. Second, we formally tested whether the number of host spawnings varied between the treatment groups (total numbers: naïve = 191 spawnings, experienced = 174 spawnings, highly experienced = 146 spawnings). To obtain an insight into temporal dynamics of cichlid spawning, we calculated the number of cichlid spawnings for each treatment in each quarter of the duration of the experimental period. We fitted a GLM with a negative binomial error distribution (to account for slightly overdispersed data) with the number of cichlid spawnings as the response variable and our treatment groups as predictors.To test how experience with host spawning (treatment) affected cuckoo catfish ability to place their eggs in the care of the host, we compared (1) the number of parasitised cichlid clutches among the three catfish experience groups (prevalence of parasitism), (2) the mean number of catfish eggs introduced into cichlid clutches among the three treatment levels (mean parasite egg abundance, the mean number of catfish eggs calculated across all cichlid broods, (3) mean parasite clutch size (the number of catfish eggs calculated only across parasitised cichlid broods), and examined (4) temporal dynamics of all three measures of parasite success within each treatment group throughout the duration of the experiment.To test for differences in prevalence of parasitism among different cuckoo catfish experience treatments, we applied a Generalised Linear Mixed-effects Model (GLMM, R package glmmTMB)60 with a binomial error distribution. We fitted the occurrence of ‘catfish parasitism’ (1 = yes, 0 = no) as the binary response variable and ‘treatment effect’ (i.e., ‘catfish experience’), ‘time progress of experiment’ (1–113 days) and ‘host female body size’ (in mm) as predictor variables. We additionally fitted an interaction between treatment (‘catfish experience’) and ‘time progress of experiment’ to the model to test whether parasitism rate changed over time at treatment-specific rates. We included tank identity (‘tank ID’) as a random intercept to account for nonindependence of data obtained from the same tank.Next, we tested whether the mean number of parasite eggs that were accepted by host females during one spawning bout differed between catfish experience treatments. We applied two GLMMs (R package glmmTMB)60 with a negative binomial error distribution (i.e., nbinom1) to account for over-dispersed count data. We applied GLMMs on the mean abundance of catfish eggs (across all host clutches) and on mean clutch size of cuckoo catfish using a subset of clutches that were parasitised. For both GLMMs, we included the ‘number of cuckoo catfish eggs per clutch’ as the response variable and treatment (‘catfish experience’), ‘time progress of experiment’, and their interaction as predictor variables. We additionally fitted ‘host female body size’ as a predictor variable because larger female cichlids are capable of laying more eggs and may appear more attractive hosts to cuckoo catfish. Further, a higher number of host eggs may increase the number of opportunities for cuckoo catfish to deposit their own eggs in the host clutch. ‘Tank ID’ was included as random intercept to account for nonindependence of data.To test whether cuckoo catfish presence affected cichlid spawning activity, we applied a GLMM (R package glmmTMB)60 with Gaussian error distribution (which provided superior model fit compared to Poisson and negative binomial distributions by ‘simulateResiduals’ and ‘testDispersion’ functions in the R package DHARMa)61. We fitted the ‘number of host eggs’ per clutch as the response variable and treatment (‘catfish experience’), ‘host female body size’, ‘time progress of experiment’, and ‘experimental phase’ (1st or 2nd phase) as predictor variables. We also included ‘tank ID’ as random intercept to account for nonindependence of data. The full model further included an interaction between treatment and ‘time progress of experiment’ to accommodate the possibility that host egg numbers may be affected differently across catfish experience treatments over time. As this full model predicted no difference in temporal aspect of host clutch size among treatments (‘catfish experience’: ‘time progress’, experienced: z = 0.92, P = 0.360, highly experienced: z = 1.46, P = 0.143), we subsequently dropped the interaction term from the final model.We used data collected from video footage to investigate whether naïve, experienced and highly experienced cuckoo catfish differed in their response to host spawnings and, additionally, if catfish from the three treatments elicited different host reactions towards them by applying Linear Mixed-effect Models using the R packages lme462 and glmmTMB60. To account for different starting times of recordings, we calculated either the rate of behaviour per minute of observation (i.e., for aggression) or their relative values (i.e., for the number of host courtships that cuckoo catfish missed).First, we tested whether host spawning pairs increased their aggressions towards cuckoo catfish over the experimental period to rule out the presence of host adaptation to cuckoo catfish intrusions, which would interfere with our aim of understanding parasite learning. We fitted a Generalised Linear Mixed-effects Model (GLMM, R package glmmTMB) with a negative binomial error distribution. The number of overt aggressive behaviours that the spawning pair performed towards cuckoo catfish per minute of catfish presence at the spawning site (summed over male and female cichlid) was fitted as the response variable and treatment (‘catfish experience’) as the predictor variable. We further included ‘time progress of experiment’ and ‘experimental phase’ as predictors to account for their possible effect on host aggression. We additionally included ‘tank ID’ as random intercept in the model to account for individual variation in host aggression levels among experimental tanks.To investigate if naïve cuckoo catfish missed more opportunities to parasitise cichlids than experienced and highly experienced catfish, we fitted a GLMM (R package lme4) with a binomial error distribution. We included the proportion of missed spawning behaviours (coded as ‘missed spawnings behaviours’ versus ‘intruded spawning behaviours’, based on count data for each spawning) as the response variable (‘spawnings missed’) and treatment (‘catfish experience’) as a predictor variable. We fitted ‘tank ID’ as a random intercept to the model to account for nonindependence of data within tanks, and we additionally fitted a random intercept based on whether the spawning was covered by a shelter or not (‘sheltered spawn’, yes / no) since spawning in a shelter may have been less apparent to catfish.We tested whether cuckoo catfish experience played a role in the timing of their intrusion to specific spawning behaviours by fitting a GLMM (R package lme4) with a Gamma error distribution to account for a positive skew in the data distribution. We included the mean delay of catfish to the first appearance of cichlid T-position in seconds (‘catfish delay’, see main text and Supplementary Movie 1 for a detailed description of cichlid spawning sequence) as the response variable and ‘catfish experience’ as the predictor variable. We included ‘tank ID’ and ‘sheltered spawn’ as random intercepts.Finally, we fitted a GLMM with a Poisson error distribution to test whether cuckoo catfish learn to synchronise their intrusion behaviour as they gain experience through interactions with their hosts. We included the maximum number of catfish during a specific cichlid spawning behaviour (‘intruder number’, count data) as the response variable and ‘catfish experience’ as the predictor variable. To account for nonindependence of data within experimental tanks and spawnings, we included a random intercept where each spawning was nested within ‘tank ID’ in the model.Ethical complianceResearch adhered to all national and institutional animal care and use guidelines, was administered under permit No. CZ62760203 and was approved by ethical boards of the Institute of Vertebrate Biology and the Czech Academy of Sciences (approval No. 32-2019).Reporting summaryFurther information on research design is available in the Nature Research Reporting Summary linked to this article. More