Philippa Kaur

Vo, Q. T., Oppelt, N., Leinenkugel, P. & Kuenzer, C. Remote sensing in mapping mangrove ecosystems: An object-based approach. Remote Sens. 5, 183–201. https://doi.org/10.3390/rs5010183 (2013).Article 

Google Scholar 
Kertész, Á. & Křeček, J. Landscape degradation in the world and in Hungary. Hung. Geogr. Bull. 68, 201–221. https://doi.org/10.15201/hungeobull.68.3.1 (2019).Article 

Google Scholar 
Vorster, A. G., Evangelista, P. H., Stovall, A. E. L. & Ex, S. Variability and uncertainty in forest biomass estimates from the tree to landscape scale: The role of allometric equations. Carbon Balance Manag. 15, 8. https://doi.org/10.1186/s13021-020-00143-6 (2020).Article 
CAS 
PubMed 
PubMed Central 

Google Scholar 
Blackman, A. Evaluating forest conservation policies in developing countries using remote sensing data: An introduction and practical guide. For. Policy Econ. 34, 1–16. https://doi.org/10.1016/j.forpol.2013.04.006 (2013).Article 

Google Scholar 
Wilfong, B. N., Gorchov, D. L. & Henry, M. C. Detecting an invasive shrub in deciduous forest understories using remote sensing. Weed Sci. 57, 512–520. https://doi.org/10.1614/WS-09-012.1 (2009).Article 
CAS 

Google Scholar 
Dyderski, M. K. & Pawlik, Ł. Spatial distribution of tree species in mountain national parks depends on geomorphology and climate. For. Ecol. Manag. 474, 118366. https://doi.org/10.1016/j.foreco.2020.118366 (2020).Article 

Google Scholar 
Milosevic, D., Dunjić, J. & Stojanović, V. Investigating micrometeorological differences between saline steppe, forest-steppe and forest environments in northern Serbia during a clear and sunny autumn day. Geogr. Pannonica 24(3), 176–186. https://doi.org/10.5937/gp24-25885 (2020).Article 

Google Scholar 
Modzelewska, A., Fassnacht, F. E. & Stereńczak, K. Tree species identification within an extensive forest area with diverse management regimes using airborne hyperspectral data. Int. J. Appl. Earth Obs. Geoinf. 84, 101960. https://doi.org/10.1016/j.jag.2019.101960 (2020).Article 

Google Scholar 
Wulder, M. Optical remote-sensing techniques for the assessment of forest inventory and biophysical parameters. Prog. Phys. Geogr. Earth Environ. 22, 449–476. https://doi.org/10.1177/030913339802200402 (1998).Article 

Google Scholar 
Tang, L., Shao, G. & Dai, L. Roles of digital technology in China’s sustainable forestry development. Int. J. Sustain. Dev. World Ecol. 16, 94–101. https://doi.org/10.1080/13504500902794000 (2009).Article 

Google Scholar 
Richter, R., Reu, B., Wirth, C., Doktor, D. & Vohland, M. The use of airborne hyperspectral data for tree species classification in a species-rich Central European forest area. Int. J. Appl. Earth Obs. Geoinform. 52, 464–474. https://doi.org/10.1016/j.jag.2016.07.018 (2016).Article 

Google Scholar 
Thenkabail, P., Gumma, M., Teluguntla, P. & Ahmed, M. I. Hyperspectral remote sensing of vegetation and agricultural crops. Photogramm. Eng. Remote Sens. 80, 695–723 (2014).
Google Scholar 
Fassnacht, F. E. et al. Review of studies on tree species classification from remotely sensed data. Remote Sens. Environ. 186, 64–87. https://doi.org/10.1016/j.rse.2016.08.013 (2016).Article 

Google Scholar 
Vangi, E. et al. The new hyperspectral satellite PRISMA: Imagery for forest types discrimination. Sensors 21, 1182. https://doi.org/10.3390/s21041182 (2021).Article 
PubMed 
PubMed Central 

Google Scholar 
Burai, P., Beko, L., Lenart, C., Tomor, T. & Kovacs, Z. Individual tree species classification using airborne hyperspectral imagery and lidar data. In 2019 10th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing (WHISPERS) 1–4. https://doi.org/10.1109/WHISPERS.2019.8921016 (2019).Kumar, B., Dikshit, O., Gupta, A. & Singh, M. K. Feature extraction for hyperspectral image classification: A review. Int. J. Remote Sens. 41, 6248–6287. https://doi.org/10.1080/01431161.2020.1736732 (2020).Article 

Google Scholar 
Li, X., Li, Z., Qiu, H., Hou, G. & Fan, P. An overview of hyperspectral image feature extraction, classification methods and the methods based on small samples. Appl. Spectrosc. Rev. https://doi.org/10.1080/05704928.2021.1999252 (2021).Article 

Google Scholar 
Wang, J. & Chang, C.-I. Independent component analysis-based dimensionality reduction with applications in hyperspectral image analysis. IEEE Trans. Geosci. Remote Sens. 44, 1586–1600. https://doi.org/10.1109/TGRS.2005.863297 (2006).Article 

Google Scholar 
Hamada, Y., Stow, D. A., Coulter, L. L., Jafolla, J. C. & Hendricks, L. W. Detecting Tamarisk species (Tamarix spp.) in riparian habitats of Southern California using high spatial resolution hyperspectral imagery. Remote Sens. Environ. 109, 237–248. https://doi.org/10.1016/j.rse.2007.01.003 (2007).Article 

Google Scholar 
Ibarrola-Ulzurrun, E., Marcello, J. & Gonzalo-Martin, C. Assessment of component selection strategies in hyperspectral imagery. Entropy 19, 666. https://doi.org/10.3390/e19120666 (2017).Article 
MathSciNet 

Google Scholar 
Dabiri, Z. & Lang, S. Comparison of independent component analysis, principal component analysis, and minimum noise fraction transformation for tree species classification using APEX hyperspectral imagery. ISPRS Int. J. Geo-Inf. 7, 488. https://doi.org/10.3390/ijgi7120488 (2018).Article 

Google Scholar 
Priyadarshini, K. N., Sivashankari, V., Shekhar, S. & Balasubramani, K. Comparison and evaluation of dimensionality reduction techniques for hyperspectral data analysis. Proceedings 24, 6. https://doi.org/10.3390/IECG2019-06209 (2019).Article 

Google Scholar 
Arslan, O., Akyürek, Ö., Kaya, Ş & Şeker, D. Z. Dimension reduction methods applied to coastline extraction on hyperspectral imagery. Geocarto Int. 35, 376–390. https://doi.org/10.1080/10106049.2018.1520920 (2020).Article 

Google Scholar 
Kadavi, P. R., Lee, W.-J. & Lee, C.-W. Analysis of the pyroclastic flow deposits of mount sinabung and Merapi using landsat imagery and the artificial neural networks approach. Appl. Sci. 7, 935. https://doi.org/10.3390/app7090935 (2017).Article 

Google Scholar 
Schlosser, A. D. et al. Building extraction using orthophotos and dense point cloud derived from visual band aerial imagery based on machine learning and segmentation. Remote Sens. 12, 2397. https://doi.org/10.3390/rs12152397 (2020).Article 

Google Scholar 
Latifi, H., Fassnacht, F. & Koch, B. Forest structure modeling with combined airborne hyperspectral and LiDAR data. Remote Sens. Environ. 121, 10–25. https://doi.org/10.1016/j.rse.2012.01.015 (2012).Article 

Google Scholar 
Clark, M. L., Roberts, D. A. & Clark, D. B. Hyperspectral discrimination of tropical rain forest tree species at leaf to crown scales. Remote Sens. Environ. 96, 375–398. https://doi.org/10.1016/j.rse.2005.03.009 (2005).Article 

Google Scholar 
Melgani, F. & Bruzzone, L. Classification of hyperspectral remote sensing images with support vector machines. IEEE Trans. Geosci. Remote Sens. 42, 1778–1790. https://doi.org/10.1109/ICIECS.2009.5363456 (2004).Article 

Google Scholar 
Belgiu, M. & Drăguţ, L. Random forest in remote sensing: A review of applications and future directions. ISPRS J. Photogramm. Remote Sens. 114, 24–31. https://doi.org/10.1016/j.isprsjprs.2016.01.011 (2016).Article 

Google Scholar 
Manandhar, R., Odeh, I. O. A. & Ancev, T. Improving the accuracy of land use and land cover classification of landsat data using post-classification enhancement. Remote Sens. 1, 330–344. https://doi.org/10.3390/rs1030330 (2009).Article 

Google Scholar 
Thakkar, A. K., Desai, V. R., Patel, A. & Potdar, M. B. Post-classification corrections in improving the classification of Land Use/Land Cover of arid region using RS and GIS: The case of Arjuni watershed, Gujarat, India. Egypt. J. Remote Sens. Space Sci. 20, 79–89. https://doi.org/10.1016/j.ejrs.2016.11.006 (2017).Article 

Google Scholar 
El-Hattab, M. M. Applying post classification change detection technique to monitor an Egyptian coastal zone (Abu Qir Bay), Egypt. J. Remote Sens. Space Sci. 19, 23–36. https://doi.org/10.1016/j.ejrs.2016.02.002 (2016).Article 

Google Scholar 
Bhosale, N., Manza, R., Kale, K., Scholar, R. & Professor, A. Analysis of effect of gaussian, salt and pepper noise removal from noisy remote sensing images. Pceedings of teh Second International Conference on ERCICA 386–390. http://rameshmanza.in/Publication/Narayan_Bhosle/Analysis%20of%20Effect%20of%20Gaussian.pdf (2014).Schöll, K., Kiss, A., Dinka, M. & Berczik, Á. Flood-pulse effects on zooplankton assemblages in a river-floodplain system (Gemenc Floodplain of the Danube, Hungary). Int. Rev. Hydrobiol. 97, 41–54. https://doi.org/10.1002/iroh.201111427 (2012).Article 

Google Scholar 
Ágoston-Szabó, E., Schöll, K., Kiss, A. & Dinka, M. The effects of tree species richness and composition on leaf litter decomposition in a Danube oxbow lake (Gemenc, Hungary). Fundam. Appl. Limnol. https://doi.org/10.1127/fal/2017/0675 (2017).Article 

Google Scholar 
Guti, G. Water bodies in the Gemenc floodplain of the Danube, Hungary: (A theoretical basis for their typology). Opusc Zool. 33, 49–60 (2001).
Google Scholar 
Berczik, Á. & Dinka, M. Bibliography of hydrobiological research on the Gemenc and Béda: Karapancsa floodplains of the River Danube (1498–1436 rkm) including the publications of the Danube Research Institute of the Hungarian Academy of Sciences between 1968 and 2017. Opusc. Zool. 49, 191–197. https://doi.org/10.18348/opzool.2018.2.191 (2018).Article 

Google Scholar 
Ceulemans, R., McDonald, A. J. S. & Pereira, J. S. A comparison among eucalypt, poplar and willow characteristics with particular reference to a coppice, growth-modelling approach. Biomass Bioenergy 11, 215–231. https://doi.org/10.1016/0961-9534(96)00035-9 (1996).Article 

Google Scholar 
Haneca, K., Katarina, Č & Beeckman, H. Oaks, tree-rings and wooden cultural heritage: A review of the main characteristics and applications of oak dendrochronology in Europe. J. Archaeol. Sci. 36, 1–11. https://doi.org/10.1016/j.jas.2008.07.005 (2009).Article 

Google Scholar 
Jones, T. G., Coops, N. C. & Sharma, T. Assessing the utility of airborne hyperspectral and LiDAR data for species distribution mapping in the coastal Pacific Northwest, Canada. Remote Sens. Environ. 114, 2841–2852. https://doi.org/10.1016/j.rse.2010.07.002 (2010).Article 

Google Scholar 
Sothe, C. et al. Tree species classification in a highly diverse subtropical forest integrating UAV-based photogrammetric point cloud and hyperspectral data. Remote Sens. 11, 1338. https://doi.org/10.3390/rs11111338 (2019).Article 

Google Scholar 
Nambiar, E. K. S. & Sands, R. Competition for water and nutrients in forests. Can. J. For. Res. 23, 1955–1968. https://doi.org/10.1139/x93-247 (1993).Article 

Google Scholar 
Mayoral, C., Calama, R., Sánchez-González, M. & Pardos, M. Modelling the influence of light, water and temperature on photosynthesis in young trees of mixed Mediterranean forests. New For. 46, 485–506. https://doi.org/10.1007/s11056-015-9471-y (2015).Article 

Google Scholar 
Stojanović, D. B., Levanič, T., Matović, B. & Orlović, S. Growth decrease and mortality of oak floodplain forests as a response to change of water regime and climate. Eur. J. For. Res. 134, 555–567. https://doi.org/10.1007/s10342-015-0871-5 (2015).Article 

Google Scholar 
Dyderski, M. K. & Jagodziński, A. M. Impact of invasive tree species on natural regeneration species composition, diversity, and density. Forests 11, 456. https://doi.org/10.3390/f11040456 (2020).Article 

Google Scholar 
Jia, S., Ji, Z., Qian, Y. & Shen, L. Unsupervised band selection for hyperspectral imagery classification without manual band removal. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 5, 531–543. https://doi.org/10.1109/JSTARS.2012.2187434 (2012).Article 

Google Scholar 
Karpouzli, E. & Malthus, T. The empirical line method for the atmospheric correction of IKONOS imagery. Int. J. Remote Sens. 24, 1143–1150. https://doi.org/10.1080/0143116021000026779 (2003).Article 

Google Scholar 
Richards, J. A. Remote Sensing Digital Image Analysis (Springer, 2013). https://doi.org/10.1007/978-3-642-30062-2.Book 

Google Scholar 
Sharifi Hashjin, S. & Khazai, S. A new method to detect targets in hyperspectral images based on principal component analysis. Geocarto Int. 37, 2679–2697. https://doi.org/10.1080/10106049.2020.1831625 (2022).Article 

Google Scholar 
Kaiser, H. F. The varimax criterion for analytic rotation in factor analysis. Psychometrika 23, 187–200 (1958).Article 
MATH 

Google Scholar 
Shah, C. A., Arora, M. K. & Varshney, P. K. Unsupervised classification of hyperspectral data: An ICA mixture model based approach. Int. J. Remote Sens. 25, 481–487. https://doi.org/10.1080/01431160310001618040 (2004).Article 

Google Scholar 
Tharwat, A. Independent component analysis: An introduction. Appl. Comput. Inform. 17, 222–249. https://doi.org/10.1016/S1364-6613(00)01813-1 (2020).Article 

Google Scholar 
Villa, A., Chanussot, J., Jutten, C., Benediktsson, J. A. & Moussaoui, S. On the use of ICA for hyperspectral image analysis. In 2009 IEEE International Geoscience and Remote Sensing Symposium vol. 4 IV-97-IV–100. https://doi.org/10.1109/IGARSS.2009.5417363 (2009).Hyvärinen, A. & Oja, E. Independent component analysis: Algorithms and applications. Neural Netw. 13, 411–430. https://doi.org/10.1016/s0893-6080(00)00026-5 (2000).Article 
PubMed 

Google Scholar 
Otukei, J. R. & Blaschke, T. Land cover change assessment using decision trees, support vector machines and maximum likelihood classification algorithms. Int. J. Appl. Earth Obs. Geoinf. 12, S27–S31. https://doi.org/10.1016/j.jag.2009.11.002 (2010).Article 

Google Scholar 
Murty, M. N. & Raghava, R. Kernel-based SVM. In Support Vector Machines and Perceptrons: Learning, Optimization, Classification, and Application to Social Networks (eds Murty, M. N. & Raghava, R.) 57–67 (Springer, 2016). https://doi.org/10.1007/978-3-319-41063-0_5.Chapter 
MATH 

Google Scholar 
Seidl, D., Ružiak, I., Koštialová Jančíková, Z. & Koštial, P. Sensitivity analysis: A tool for tailoring environmentally friendly materials. Expert Syst. Appl. 208, 118039. https://doi.org/10.1016/j.eswa.2022.118039 (2022).Article 

Google Scholar 
Zhao, D., Pang, Y., Liu, L. & Li, Z. Individual tree classification using airborne LiDAR and hyperspectral data in a natural mixed forest of Northeast China. Forests 11, 303. https://doi.org/10.3390/f11030303 (2020).Article 

Google Scholar 
Aksoy, S. & Akcay, H. G. Multi-resolution segmentation and shape analysis for remote sensing image classification. In Proceedings of 2nd International Conference on Recent Advances in Space Technologies, 2005. RAST 2005. 599–604 (2005). https://doi.org/10.1109/RAST.2005.1512638.Dalponte, M., Ørka, H. O., Ene, L. T., Gobakken, T. & Næsset, E. Tree crown delineation and tree species classification in boreal forests using hyperspectral and ALS data. Remote Sens. Environ. 140, 306–317. https://doi.org/10.1016/j.rse.2013.09.006 (2014).Article 

Google Scholar 
Amini, S., Homayouni, S., Safari, A. & Darvishsefat, A. A. Object-based classification of hyperspectral data using Random Forest algorithm. Geo-Spat. Inf. Sci. 21, 127–138. https://doi.org/10.1080/10095020.2017.1399674 (2018).Article 

Google Scholar 
Congalton, R. G. A review of assessing the accuracy of classifications of remotely sensed data. Remote Sens. Environ. 37, 35–46. https://doi.org/10.1016/0034-4257(91)90048-B (1991).Article 

Google Scholar 
Foody, G. M. Status of land cover classification accuracy assessment. Remote Sens. Environ. 80, 185–201. https://doi.org/10.1016/S0034-4257(01)00295-4 (2002).Article 

Google Scholar 
Tharwat, A. Classification assessment methods. Appl. Comput. Inform. 17, 168–192. https://doi.org/10.1016/j.aci.2018.08.003 (2020).Article 

Google Scholar 
Field, F. Discovering Statistics Using IBM SPSS Statistics. SAGE Publications Ltd https://uk.sagepub.com/en-gb/eur/discovering-statistics-using-ibm-spss-statistics/book257672 (2022).R Core Team. R: A language and environment for statistical computing. https://www.gbif.org/tool/81287/r-a-language-and-environment-for-statistical-computing (2022).Galucci, M. Generalized Mixed Models module. R package version 2.0.5. https://gamlj.github.io/gzlmmixed.html More

Study sites and study designThe study was performed in the frame of a dog ecology research project, with details on the study locations published elsewhere15,42,43. For the current study, five study sites located in Indonesia and Guatemala were included. Site selection was carried out by each country’s research team, taking into consideration rural and urban settings, as well as differing expected number of dogs present at each location. The Indonesian study sites were semi-urban Habi and rural Pogon, in the Sikka regency, at the eastern area of Flores Island (Supplementary Fig. 6). In Guatemala, the study sites were Poptún (urban setting), Sabaneta and La Romana (both rural settings), located in the Guatemalan department of Péten, in the northern part of the country (Supplementary Fig. 7). Data were collected during May to June 2018 in Guatemala and from July to September 2018 in Indonesia.In each location, a 1 km2 area was predefined using Google Earth within which the study took place. The 1 km2 area was chosen because of the research goals of another part of the project, investigating the contact network of the dogs15. Within these areas, the teams visited all dog-owning households. In each household, the study was presented to an adult of the family, who was then asked if they owned a dog and if they were willing to participate in the study. After the dog owner’s oral or written consent was granted, a questionnaire was answered, and the dogs collared. The handling of the dogs was performed by a trained veterinarian or a trained veterinary paramedic of the team.The questionnaire data was collected through interviews with the dog owners. Multiple dogs per household could be included as multiple entries in the questionnaire. The detailed questionnaire contains information on the household location, dog demographics (age, sex, reproductive status) and management (dog’s purpose, origin, confinement, vaccination status, feeding and human-mediated transportation within and outside the pre-determined area).All dogs of a household fulfilling the inclusion criteria were equipped with a geo-referenced contact sensor (GCS) developed by Bonsai Systems (https://www.bonsai-systems.com), containing a GPS module and an Ultra-High-Frequency (UHF) sensor for contact data recording43,44. GCS devices report a 5-m maximum accuracy, a run-time of up to 10 years, can store up to 4 million data points and carry a lithium-polymer-battery (LiPo). For this study, only GPS data were analysed. The GCS were set to record each dog’s geographical position at one-minute intervals. Dogs remained collared for 3 to 5 days with the duration of the data collection being limited by the device’s battery capacity, as batteries were not re-charged or changed during the study. Throughout the time of recording, date, hour, GPS coordinates and signal quality (HDOP) raw data were collected by the GPS module and amassed into the workable databases.Exclusion criteria were dogs of less than four months of age (since they were not big enough to carry a collar), sick dogs and pregnant bitches (to avoid any risk of stress-induced miscarriages). Reasons for non-participation of eligible dogs included dog owner’s absence, dog’s absence, inability to catch the dog, and refusal of participation by the dog owner. In addition, dogs foreseen for slaughtering within the following four days were excluded in Indonesia to ensure data collection for at least four to five days. All dogs included in this study were constantly free roaming or at least part-time (day only, night only and for some hours a day). Human and/or animal ethical approval were obtained depending on the country-specific regulations. All the procedures were carried out in accordance with relevant guidelines. Ethical clearance was granted in Guatemala by the UVG’s International Animal Care and Use Committee [Protocol No. I-2018(3)] and the Community Development Councils of the two rural sites, which included Maya Q’eqchi’ communities45. In Indonesia, the study was approved by the Animal Ethics Commission of the Faculty of Veterinary Medicine, Nusa Cendana University (Protocol KEH/FKH/NPEH/2019/009). In addition, dogs that participated in the study were vaccinated against rabies and/or dewormed to acknowledge the owners for their participation in the study.Data cleaningData were stored in an application developed by Bonsai Systems compatible with Apple operating system (iOS iPhone Operating Systems), downloaded as individual csv file for each unit, and further analysed in R (version 3.6.1)46.The GPS data were cleaned based on three automatised criteria. First, the speed was calculated between any two consecutive GPS fixes, and fixes with speed of  > 20 km/h were excluded, given the implausibility of a dog running at such speed over a one-minute timespan47. It is noteworthy that car travel causes speeds over 20 km/h. However, as we were interested in analysing the dog’s behaviour outside of car transports, removing these fixes was in line with our objectives. Second, the Horizontal Dilution of Precision (HDOP), which is a measure of accuracy48 and automatically recorded by the devices for each GPS fix, was used to exclude fixes with low precision. According to Lewis et al.49, GPS fixes with HDOP higher than five were excluded, which deleted 1.3% of data in Habi, 2.2% in Pogon, 3.3% in Poptún, 1.8% in La Romana and 2.1% in Sabaneta. Third, the angles built by three consecutive fixes were calculated for each dog. When studying animals’ trajectories as their measure of movement, acute inner angles are often connected to error GPS fixes50. The fixes having the 2.5% smallest angles were excluded, to target those fixes with highest risks of being errors, while balancing against the loss of GPS fixes due to the cleaning process. With the exclusion of the smallest angles, 2.6% of data were deleted in Habi, 3% in Pogon, 2.9% in Poptún, 2.6% in La Romana and 2.7% in Sabaneta. After the automatised cleaning was concluded, 18 obvious error GPS fixes (unachievable or inexplicable locations by dogs) still prevailed in the Habi dataset and were manually removed.Habitat resource identification and calculation of terrain slopeTo analyse habitat selection of the collared FRDD, resources were delimited by a 100% Minimum Convex Polygon (MCP) including all cleaned GPS fixes per study site, using QGIS51 (Fig. 1).Figure 1GPS fixes plotted over a Google satellite imagery layer with its respective outlined computed Minimum Convex Polygon (MCP) delimitating the habitat available for the study population in: (a) Habi; (b) Pogon; (c) Poptún; (d) La Romana and (e) Sabaneta. Source QGIS (version 3.4 Madeira, http://qgis.org), map data: Google Satellite.Full size imageResources were defined by taking into consideration the following criteria: resources are (i) likely to impact upon movement patterns of dogs, (ii) identifiable by landscape satellite topography, and (iii) chosen considering information on relevant gathering places for FRDD observed by the field teams. Three resources were disclosed in all study sites: buildings, roads and vegetation coverage. All habitat relevant resources were manually identified within the available area (MCP) in QGIS using satellite imagery. All building-like structures were identified using vector polygons and summed under the layer “buildings”. Roads were identified and manually traced using vector lines in all sites, except in Poptún where the roads were automatically traced using an OpenStreetMap road layer of the area (https://www.openstreetmap.org/export). A buffer vector polygon was generated to encompass the full potential width of the roads, with a 5 m width in Habi and Poptún (semi-urban and urban site) and a 2 m width in Pogon, La Romana and Sabaneta (rural sites). In Habi, a “beach” layer was defined by generating a five-meter buffer from the shoreline in both directions using a vector polygon. The layer “sea” was defined as the vector polygon resulting from the difference between the MCP sea outer limit and the beach buffer polygon. Vegetation coverage was distinct between study sites with sparse vegetation and bushes present in all sites except Pogon, and dense forest-like vegetation present in La Romana and Pogon. These two types of vegetation were defined as “low” and “high vegetation”, respectively. In Habi and La Romana, “low” and “high vegetation”, respectively, were manually identified using vector polygons and summarised under the respective layers. Finally, open field in Habi, high vegetation in Pogon and low vegetation in Poptún, La Romana and Sabaneta were the last vector layers to be established since they represented the difference between all other polygon vector layers and the MCP total area. After all resource vector polygons had been created, an encompassing vector layer was generated by merging all resource polygon vectors for final resource classification (Fig. 2). As part of the resource classification in Habi, the airport terminal and runaway as well as waterways enclosed in the MCP area were identified but excluded from the analysis.Figure 2(a) Habi, (b) Pogon, (c) Poptún, (d) La Romana and (e) Sabaneta Habitat classification vector layers. The different habitat resources, identifiable by colour, were merged to create the comprehensive Habitat classification vector. In the Indonesian sites (a, b) and Guatemalan sites (c–e) buildings are coloured red, vegetation low in Habi, Poptún, La Romana and Sabaneta is coloured light green, vegetation high in Pogon and La Romana dark green, roads black, beach yellow, sea dark blue, airport grey, waterways light blue and open field light orange. The airport area (gray) and waterways (light blue) in Habi were not classified as separate habitat layers and were excluded from further analysis. Source QGIS (version 3.4 Madeira, http://qgis.org), map data: Google Satellite.Full size imageAfter the construction of the habitat resource layers, all GPS fixes were assigned to the respective resource they were located, using the QGIS join attributes by location algorithm. Fixes located exactly on the MCP border in Indonesia were not classified automatically and had to be manually classified to the respective resource.In non-flat topographies (all locations expect Habi) we tested the hypothesis of whether the steepness would influence the dogs’ movement patterns. The degrees of slope were calculated using a 30-m raster-cell resolution (STRM 1-Arc Second Global, downloaded from the United States Geological Survey (USGS) Earth Explorer, https://earthexplorer.usgs.gov/). The slope was assigned by the QGIS join attributes by location algorithm to each GPS fix.Statistical analysisTo quantify habitat selection in each study site, we compared resources used by the dogs with the resources available, according to Freitas et al.52. Adapting the methodology applied by O’Neill et al.18, the observed number of GPS fixes for each dog was used to generate an equivalent number of locations that were randomly distributed within the MCP area using the Random points in layer bound vector tool from QGIS. For example, if dog “D300” had 100 recorded GPS fixes, 100 random points were generated within the MCP of the respective study site and assigned to “D300”. Random points were then assigned to the respective resources and slope of that location, as previously done with the observed GPS fixes. Using this approach, the habitat resources used by each dog could be compared to the available resources in the respective study site, using a regression model.Observation independence is a fundamental presupposition of any regression model. However, the spatial nature of the point-referenced data permits perception of spatial dependence. In our dataset, spatial autocorrelation was proven for all study sites using the Moran’s I test. Therefore, we applied a spatial regression model, which takes into consideration spatial autocorrelation while exploring the effects of the study variables. A mixed effects logistic regression model accounting for spatial autocorrelation was created to quantify the effect of variables on used (i.e. observed GPS fix) versus available (i.e. randomly generated GPS fixes) resources, using the fitme function in the spaMM package in R53,54. The model’s binary outcome variable was defined as either observed (1) or random (0) GPS fix, i.e. the dog being present or absent from a position. The explanatory variable was the resource classification with “buildings”, “roads”, “low vegetation”, “beach”, “sea” and “open field” as levels in Habi; “buildings”, “roads” and “high vegetation” in Pogon; “buildings”, “roads”, “low vegetation” in Poptún and Sabaneta; and “buildings”, “roads”, and “high” and “low vegetation” in La Romana. Different habitat resources were used interchangeably as reference level. In all study sites except Habi, the slope was included as an additional explanatory variable. As observations were not evenly distributed in time, with less observations recorded towards the end of the study, a variable ”hour” was added as an additional continuous fixed effect.Each observed GPS fix was assigned to the hour of its record, with the earliest timestamp registered in each study site being assigned the hour zero. The randomly generated points were randomly assigned to an hour within the determined time continuum of the observed GPS fixes. As our focus was investigating habitat selection at a population-level, we assumed there was no within-dog autocorrelation (space/time) and each dog was independent and exhibited no group behaviour38. Still, to partially account for spatial autocorrelation of each dog’s household, the random effects included in models were defined as each dog’s household geographical location recorded during fieldwork by a GPS device. The restricted maximum likelihood (REML) through Laplace approximations, which can be applied to models with non-Gaussian random effects55, and the Matérn correlation function were used to fit the spatial models with the Matérn family dispersion parameter ν, indicator of strength of decay in the spatial effect, was set at 0.554. More

Many eukaryotes display intraspecific genome size (GS) variation due to varying amounts of non-coding DNA1,2,3,4,5. Such GS variation can be mediated by additional genomic elements, which are physically represented either by extra (B-)chromosomes or by large heterozygous insertions into the regular chromosomes. On a DNA sequence level, non-coding DNA can be classified as highly repetitive, e.g. interspersedly repeated transposable elements or tandemly repeated satellite DNA, or as the result of previous duplications of the genome followed by pseudogenization6. The long-term gain and loss of such non-coding DNA sequences is thought to be governed by largely neutral evolutionary processes, and their excessive accumulation in some genomes can be explained by genetic drift7,8, even though selection might also sometimes play a role9,10.Non-coding DNA can affect organisms in different ways. A large number of studies document correlations between genome size and organismic traits such as cell size11,12, body size13,14, or developmental rates15, sometimes even at the within-population level13. Under some circumstances, differential amounts of non-coding DNA might even affect fitness16. Furthermore, DNA can have coding-independent effects that operate at lower levels, such as intragenomic selection. For example, (additional) genomic elements might increase their own fitness by increasing their transmission rates to offspring by meiotic drive, sometimes at the expense of their host’s fitness17,18,19. Meiotic drive in this classical sense occurs during the chromosome segregation during the meiotic divisions, even though later stages during gametogenesis can also be affected20. Recognizing and disentangling such effects is important for a better understanding of the evolution of eukaryotic genomes, in particular, the evolutionary causes of the large intraspecific genome size variation.Here we study meiotic transmission patterns of additional genomic elements in the monogonont rotifer Brachionus aplanchnoidis. Individuals of this species can differ by up to almost two-fold in genome size, which is mediated by several Megabase-sized independently segregating genomic elements (ISEs) consisting mainly of tandemly repeated satellite DNA21. The genomic data are consistent with a mixture of both B-chromosomes and large insertions to normal chromosomes21,22. Individual rotifers and their clonal offspring can be characterized by the number and size of their ISEs and their composition stays constant through hundreds of asexual (mitotic) generations22. Occasionally, monogonont rotifers engage in sexual reproduction (Fig. 1), producing sexual females, whose oocytes undergo classical meiosis with two polar bodies formed23. Unfertilized haploid eggs develop mitotically into males, and sperm production does not involve any meiotic maturation divisions24. By analyzing the genome size distributions of haploid males produced by different mother clones, it has been shown that ISEs segregate in a manner suggesting that they do not pair with each other, nor with any other part of the genome22. For instance, a clone containing three ISEs will produce males (and gametes) that might contain either zero, one, two, or three ISEs, corresponding to four different GS classes of the males in this clone. The frequencies of these different GS classes roughly approximated those expected by random segregation. However, previous studies in B. asplanchnoidis did not resolve different steps during meiotic transmission, so they were not designed to detect meiotic drive or subsequent changes in meiotic transmission, and they also did not test whether there were subtle deviations from completely independent segregation.Figure 1Schematics of rotifer life cycle. Monogonont rotifers are cyclical parthenogens, capable of both ameiotic parthenogenesis and sexual reproduction. The production of sexual females is triggered by quorum sensing chemicals, released by the animals themselves at high population density. In contrast to parthenogenetic females, sexual females produce oocytes by meiosis, and give rise to either haploid males or diploid resting eggs, depending on whether they get fertilized by a male24.Full size imageIn the present study, we test for meiotic transmission biases of ISEs. If meiotic transmission would be completely unbiased, the frequencies of haploid oocytes, or males, with different numbers of ISEs should be identical to those expected by random segregation. For example, a mother with two ISEs should produce males with zero, one, or two ISEs (hence, three male GS classes), which have relative frequencies of 0.25, 0.5, and 0.25, respectively. However, if ISEs avoid segregating into polar bodies due to meiotic drive17,20,25, one would expect to see an increase in the relative frequency of male GS classes with two ISEs, compared to those with no ISE . By contrast, if ISEs are preferentially sequestered into polar bodies due to meiotic drag 7,26, the GS class with two ISEs should be underrepresented. Our experimental approach for detecting meiotic transmission biases relies on measuring (by flow-cytometry) the observed relative frequencies of each male GS class and comparing these to their relative frequencies expected under unbiased transmission (Fig. 2). To allow for clear comparisons, the main output variable in these analyses is the observed/expected ratio (O/E-ratio), i.e., the observed frequency divided by the expected relative frequency for each GS class. If there were no transmission biases, O/E-ratios across all GS classes should equal one. In contrast, O/E-ratios larger than one indicate overrepresentation of a certain GS class, and if O/E ratios increase or decrease with genome size, this indicates drive or drag at a meiotic or postmeiotic stage (Fig. 2d,h).Figure 2Principle of inferring meiotic transmission patterns from the genome size distributions of haploid rotifer males. The first four panels (a–d) show a rotifer clone with one ISE (i.e., two corresponding male GS classes). The last four panels (e–h) show a clone with four ISEs (i.e., five corresponding male GS classes). a, e Example of flow cytometry data. b, f Conceptual model of ISE meiotic segregation. c, g Theoretically predicted GS distributions of males (relative to the female GS) under meiotic drive, meiotic drag, or in the absence of meiotic drive. d, h Theoretically predicted O/E ratios (observed vs. expected frequencies of different male GS classes) under drive, drag, or on absence of drive. O/E values of  > 1 indicate over-representation of a GS class (relative to the frequency expected from unbiased transmission).Full size imageWe implemented these ideas in a mathematical model that contains the two parameters, transmission bias and cosegregation bias. Values for transmission bias may range from − 1 to 1 in our model. For instance, a value of 0.1 denotes a 10% increase in probability that an ISE segregates towards the egg pole (this is equivalent to a transmission rate of 0.55 for this ISE, i.e. mild meiotic drive). Concerning the second parameter, cosegregation bias, a positive value means that pairs of ISEs have an increased probability of being sequestered towards the same pole (irrespective of whether this is the egg pole or polar body pole), while a negative bias favors migration towards opposite poles. Please note that a cosegregation bias value of − 1 (i.e., 100% probability that ISEs migrate towards opposite poles) resembles the default segregation pattern of regular chromosomes. By estimating the transmission bias and cosegregation bias parameter for each rotifer clone, we tried to infer and compare general meiotic transmission patterns across clones, even if they contained different numbers and types of ISEs.Transmission biases may not only arise during meiosis, as described above but also during later stages of male embryonic development. For instance, they might be caused by differences in the survival of embryos, or due to differences in the fitness of hatched males containing different numbers of ISEs. To address these potential sources of variation, we compared the transmission biases in relatively young, synchronized male eggs, older eggs accumulating in growing cultures, and hatched males. Finally, to address the question of whether a high number of ISEs affects male embryonic survival in general, we estimated and compared hatching rates of (haploid) male eggs and (diploid) female eggs in 19 rotifer clones of different genome sizes (which is highly correlated with the number and size of ISEs in the genome22).Our results suggested that the ISEs in B. asplanchnoidis exhibit diverse meiotic segregation patterns: In some rotifer clones, transmission bias was positive, while the ISEs of other clones showed negative transmission bias (indicative of drag). Furthermore, we obtained evidence for a negative cosegregation bias in some clones, i.e., pairs of ISEs showed an increased probability to segregate towards opposite poles. Overall, these transmission patterns seemed to be determined early in the haploid life cycle, probably at or shortly after meiosis, since early and late stages of male embryonic development showed very similar GS distributions. Finally, we found that very large genome size (i.e., a large numbers of ISEs) was associated with reduced male embryonic survival. More

The attachment site of ticks has been studied in the context of both animal and human tick preference. In Oklahoma, a study of horses indicated that A. americanum preferentially bites the inguinal area, while I. scapularis and D. albipictus, the moose-tick, primarily bite the chest and axillary region, with D. albipictus often being found on the back18. A survey of dogs and cats across the US identified a similar distribution of ticks on dogs, with the attachment being most common on the abdomen, axillary and inguinal regions. However, this was species-specific with D. variabilis preferring the head and neck specifically19. Cats were more successfully parasitized by I. scapularis which preferred the head and A. americanum, which preferred the tail and perianal region19. This is similar to a study of tick distribution on wild black bears (Ursus americanus) in Pennsylvania, indicating that the primary tick present was I. scapularis and that the greatest numbers were found in association with the ears and muzzle20. In these cases, the ability for ticks to attach to specific areas is most likely a result of the grooming habits and abilities of the animals in question.Studies of anatomical region preference in humans also reported tick bite-site specificity associated with particular tick species. For example, in Korea, H. longicornis was determined to prefer abdomen and lower extremities (33%) and the abdomen/inguinal area (26.4%)21, which is a behavior similar to that of A. americanum observed here. Although H. longicornis is present in New York1, insufficient numbers were detected to draw definitive conclusions about its biting preference here. Additionally, a study in England (I. ricinus) reported that tick bites were most common in the legs (50%) of adult humans, but in the head and necks of children (43%)22, a differentiation that our survey does not at this time include. A similar phenomenon was observed in Russia, where tick bites were most common on the head and neck of all individuals (39.2%), but were much more common in children (84.9%)23. This study determined that the bite-site of single tick bites that resulted in infection with the Tick-Borne Encephalitis virus (TBEV) were associated with lethal outcomes if the bites were located on the head, neck, arms or axilla, while less lethality was associated with bites to the lower limbs and groin. This is most directly analogous to the transmission of DTV by I. scapularis, suggesting that bite site may have a similar relationship to disease outcomes in the related North American pathogen/vector pair.Under normal circumstances, ticks exist in sylvatic cycles with specific host preferences based on the tick species and life stage, with spillover to humans occasionally occurring for species with generalist feeding habits. Therefore, the feeding behaviors of ticks are variable, and this influences the ways that the ticks interact with humans.Ixodes scapularis is less specific in host-site preferenceThe primary life stages of I. scapularis that bite humans include nymphs and adult females, although males may also be found on humans. The body segment preference of I. scapularis is less specific than for D. variabilis, which prefers the head, and A. americanum, which prefers the thighs and pelvic region. Ixodes scapularis is primarily found on the central trunk, including the groin/pelvic region, the abdomen, the thoracic region, and the head/neck. This varies between the life stages, with more adults found in the thoracic/abdominal region of the body and nymphs being more commonly found on the arms and legs. This is partly due to the substantial size difference between adult and nymph/larval I. scapularis, with larvae being almost imperceptible and nymphs having a total body length of two to four millimeters. This results in nymphs/larvae being much more difficult to see, allowing them to more readily attach to the most visible portions of the human body while adults are restricted mostly to areas covered by clothing and hair.The presence of ticks on the head and neck indicates that I. scapularis tends to climb, although not with the preference for hair observed with D. variabilis. They appear to spend substantial time moving on the host, a period where they can be removed easily without having had a chance to potentially transmit pathogens by biting. On deer, this corresponds to a preference to move toward the neck and ears where the ticks are more difficult to dislodge24,25. On humans, it results in wide distribution across the whole body with less location specificity than other ticks.In addition to body region and life stage identification, I. scapularis ticks were also screened for several pathogens to determine if infection status influences host site preference. Anaplasma phagocytophilum, B. microti, and other pathogens (DTV and B. miyamotoi) did not influence the body segment the ticks chose to feed. However, in ticks infected with B. burgdorferi, a statistically significant change in the distribution of tick bites marked by an increased report of tick bites in the midsection and a decreased tick bites in the arms, legs, and head. While this may suggest a change in tick behavior/fitness in response to infection, it may also relate to the differences in infection rates of adult and nymph/larval ticks. Larvae, having never fed, are not infected with B. burgdorferi, and the rate of infection in nymphs is lower than that of adults1. Nymphs are less likely to be infected and are more likely to attach to the arms and legs, which is a potential source of the observed difference in infection rates. However, it remains unclear why this is not observed for the other pathogens that follow the same trend of increased infection rate in adult versus nymph/larval ticks.Bacterial and protozoal agents transmitted by I. scapularis take several hours for an infectious dose to be transmitted26,27,28. Therefore, prompt detection and removal of ticks is important for preventing tick-borne disease. Furthermore, understanding where the ticks attach allows them to be more easily detected, and also assists in preparing protective clothing for individuals entering tick-endemic areas. Additionally, knowing the biting location of I. scapularis could aid in detecting potential erythema migrans, a skin condition that occurs at the point of B. burgdorferi infected tick exposure in about 80% of cases29, which is highly diagnostic for both Lyme disease and STARI, which is transmitted by A. americanum.
Amblyomma americanum prefers the thighs and groin of subjectsAmblyomma americanum, the lone star tick, is present throughout the southern portion of New York and is particularly dominant on Long Island1. This species is relatively large, fast, and aggressive, feeding on various animals, including deer, medium-sized animals, and birds30. As a generalist feeder, both adult and nymph/larval A. americanum often bite humans in endemic areas. This experiment identified six larvae, 107 nymphs, and 48 adult A. americanum from human sources. The dominance of nymph submissions is likely due to the large size of the tick, making nymphs and adults easier to spot in more visible areas.In terms of body segment location, all life stages of A. americanum were most often found in the thigh/groin/pelvic region. Considering that most humans encounter ticks while walking through vegetation, the ticks most likely first adhere to the legs and move upward before biting. In this case, the ticks bite rapidly instead of ascending in large numbers to the torso or head. This area is also almost invariably covered in relatively tight-fitting clothing. The closeness of the fabric may also assist in inducing the ticks to feed by slowing their ascent and creating contact to induce biting.While it does not transmit the same range of pathogens as I. scapularis, A. americanum is still a medically significant species. This species can transmit Ehrlichia chaffeensis and E. ewingii31,32, which are at present rare in New York, but are likely to increase as more A. americanum becomes established. Amblyomma americanum is also associated with Southern Tick-Borne Rash Associated Illness (STARI)11, a disease of unknown etiology that has previously been observed in New York33 and with galactose-alpha-1,3-galactose (alpha-gal) allergy, a reaction to the tick’s saliva that can result in a long term, potentially serious allergic sensitivity to the consumption of red meat. While the attachment time required to transmit or induce these pathogens is still unclear, prompt detection and removal of the tick is still recommended. Knowing the approach of the tick and where it is likely to be found improves this process.Additionally, it is unclear if the results observed for A. americanum also apply to the related A. maculatum, the vector of Rickettsia parkeri, a cause of spotted fever. These ticks have been observed in the southernmost portions of New York with a high infection rate with R. parkeri34. Since early R. parkeri infection may result in a visible eschar, understanding where the eschar is most likely located can be critical for rapid diagnosis before the onset of severe disease symptoms. Considering the similarities in behavior between the two Amblyomma species, it may have similar preferences to A. americanum. Other escharotic diseases, such as F. tularensis, may also be present and linked to a tick with a highly dissimilar segment preference. The location of the escar itself, therefore, may be at least partially diagnostic for specific pathogens. However, at present, the sample size within this community engaged passive surveillance program is too small to assess its biting behavior in detail.
Dermacentor variabilis exhibits preference for the human headIn this study, D. variabilis was almost exclusively encountered in its adult life stage. This indicates that while the adult ticks are generalist feeders that may bite humans, the nymph and larval stages are not and have much greater host specificity, either feeding exclusively on a specific type of animal or being restricted to the vicinity of animal burrows. The exact identity of the preferred larval and nymphal host of D. variabilis in New York could not be determined from these data, but is presumed to be one or several rodent species, lagomorph, or mesocarnivore with broad distribution across the eastern United States.Additionally, D. variabilis was unique among the three species of ticks studied here. It had a strong bias toward the head and neck of human hosts, as opposed to a higher preference for the midsection and pelvis/groin with I. scapularis and especially A. americanum. This is clear evidence of climbing behavior, tending upward, but is also indicative of a strong preference for dense hair. In contrast to I. scapularis and A. americanum, D. variabilis in its adult stage is less likely to feed on deer35,36, with a preference for canids36, hence its colloquial name as the “American dog tick”. Hair provides the ticks with the same benefits as feeding on canids. It protects them from being immediately detected and removed, obscuring them until they can feed extensively. This can be of potential medical consequence in the case of tick paralysis, a condition of flaccid paralysis associated with the bite of Dermacentor spp. ticks30. In such cases, prompt removal of the tick is critical for treatment. Therefore, understanding its most likely location can be useful for removal of the tick before the onset of the condition, diagnostically to confirm the presence of the tick, or during treatment to ensure its removal. Considering that the tick will most likely be adult, it should be relatively obvious with careful observation.Limitations of this studyThe data described in this manuscript derived from a set of ticks submitted by general public, with site location from a questionnaire completed upon tick submission. While speciation and pathogen testing were performed under laboratory conditions, the public completed the initial survey and is therefore subject to a level of inherent error and ambiguity. In the context of this study, this mainly concerns whether the body location submitted concerns an attachment or a tick that is still crawling over the potential host in preparation for biting. The term “attachment” may be colloquially interpreted as to contain both categories, or a person can potentially be mistaken about the state of the tick. While ticks filled with blood have fed, the situation is more indeterminate for short-duration attachments where the ticks have not yet begun to engorge. This may introduce some level of error from ticks found on a body segment that were not, at the time of collection, attached. However, the data are overall still useful for predicting the most likely location where ticks of specific species can be found on a person. Studies with test subjects and ticks under controlled conditions may assist in elucidating this matter further. Additionally, this data set was compiled without regard to gender and age group. This data was not collected with this version of the questionnaire; therefore, the tick attachment cannot be stratified by any demographic parameters of tick submitters. More

United Nations. World Urbanization Prospects: The 2018 revision. (Department of Economic and Social Affairs, Population Division, United Nations, 2018).Grimm, N. B. et al. Global change and the ecology of cities. Science 319, 756–760 (2008).Article 

Google Scholar 
McPhearson, T. et al. Advancing urban ecology toward a science of cities. Bioscience 66, 198–212 (2016).Article 

Google Scholar 
Dodman, D. et al. Cities, settlements and key infrastructure. In Climate Change 2022: Impacts, Adaptation, and Vulnerability. Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change (eds. Pörtner, H.-O. et al.) 997–1040 (Cambridge University Press, Cambridge, UK and New York, NY, USA, 2022).Díaz, S. et al. Assessing nature’s contributions to people: Recognizing culture, and diverse sources of knowledge, can improve assessments. Science 359, 270–272 (2018).Article 

Google Scholar 
Grabowski, Z. J., McPhearson, T., Matsler, A. M., Groffman, P. & Pickett, S. T. A. What is green infrastructure? A study of definitions in US city planning. Front. Ecol. Environ. 20, 152–160 (2022).Article 

Google Scholar 
Childers, D. L. et al. Urban ecological infrastructure: An inclusive concept for the non-built urban environment. Elementa 7, 1–14 (2019).
Google Scholar 
Gómez-Baggethun, E. et al. Urban ecosystem services. In Urbanization, Biodiversity and Ecosystem Services: Challenges and Opportunities (eds. Elmqvist, T. et al.) 175–251 (Springer, Netherlands, 2013).Díaz, S. & Cabido, M. Vive la différence: Plant functional diversity matters to ecosystem processes. Trends Ecol. Evol. 16, 646–655 (2001).Article 

Google Scholar 
Burkhard, B. & Maes, J. Mapping Ecosystem Services (Pensoft Publishers, Sofia, 2017).Eviner, V. T. & Chapin, F. S. Functional Matrix: A conceptual framework for predicting multiple plant effects on ecosystem processes. Annu. Rev. Ecol. Evol. Syst. 34, 455–485 (2003).Article 

Google Scholar 
Lavorel, S., McIntyre, S., Landsberg, J. & Forbes, T. D. A. Plant functional classifications: From general groups to specific groups based on response to disturbance. Trends Ecol. Evol. 12, 474–478 (1997).Article 

Google Scholar 
Cornelissen, J. H. C. et al. A handbook of protocols for standardised and easy measurement of plant functional traits worldwide. Aust. J. Bot. 51, 335–380 (2003).Article 

Google Scholar 
Suding, K. N. et al. Scaling environmental change through the community-level: A trait-based response-and-effect framework for plants. Glob. Chang. Biol. 14, 1125–1140 (2008).Article 

Google Scholar 
Lavorel, S. & Garnier, E. Predicting changes in community composition and ecosystem functioning from plant traits: Revisiting the Holy Grail. Funct. Ecol. 16, 545–556 (2002).Article 

Google Scholar 
Hevia, V. et al. Trait-based approaches to analyze links between the drivers of change and ecosystem services: Synthesizing existing evidence and future challenges. Ecol. Evol. 7, 831–844 (2017).Article 

Google Scholar 
Cadotte, M. W., Carscadden, K. & Mirotchnick, N. Beyond species: Functional diversity and the maintenance of ecological processes and services. J. Appl. Ecol. 48, 1079–1087 (2011).Article 

Google Scholar 
Lavorel, S. Plant functional effects on ecosystem services. J. Ecol. 101, 4–8 (2013).Article 

Google Scholar 
Andersson, E. et al. What are the traits of a social-ecological system: Towards a framework in support of urban sustainability. npj Urban Sustain. 1, 14 (2021).Article 

Google Scholar 
Pickett, S. T. A. et al. Urban ecological systems: Linking terrestrial ecological, physical, and socioeconomic components of metropolitan areas. Annu. Rev. Ecol. Syst. 32, 127–157 (2001).Article 

Google Scholar 
McPhearson, T., Haase, D., Kabisch, N. & Gren, Å. Advancing understanding of the complex nature of urban systems. Ecol. Indic. 70, 566–573 (2016).Article 

Google Scholar 
Zhou, W., Pickett, S. T. A. & McPhearson, T. Conceptual frameworks facilitate integration for transdisciplinary urban science. npj Urban Sustain. 1, 1 (2021).Article 

Google Scholar 
Andersson, E. et al. Scale and context dependence of ecosystem service providing units. Ecosyst. Serv. 12, 157–164 (2015).Article 

Google Scholar 
Pinho, P. et al. Research agenda on biodiversity and ecosystem functions and services in European cities. Basic Appl. Ecol. 53, 124–133 (2021).Article 

Google Scholar 
Bullock, J. M. et al. Human-mediated dispersal and the rewiring of spatial networks. Trends Ecol. Evol. 33, 958–970 (2018).Article 

Google Scholar 
Avolio, M. L., Swan, C., Pataki, D. E. & Jenerette, G. D. Incorporating human behaviors into theories of urban community assembly and species coexistence. Oikos 130, 1849–1864 (2021).Article 

Google Scholar 
Aronson, M. F. J. et al. Hierarchical filters determine community assembly of urban species pools. Ecology 97, 2952–2963 (2016).Article 

Google Scholar 
Woodward, F. I. & Diament, A. D. Functional approaches to predicting the ecological effects of global change. Funct. Ecol. 5, 212 (1991).Article 

Google Scholar 
Diaz, S., Cabido, M. & Casanoves, F. Plant functional traits and environmental filters at a regional scale. J. Veg. Sci. 9, 113–122 (1998).Article 

Google Scholar 
Boet, O., Arnan, X. & Retana, J. The role of environmental vs. biotic filtering in the structure of European ant communities: A matter of trait type and spatial scale. PLoS ONE 15, e0228625 (2020).Article 

Google Scholar 
Grimm, N. B., Grove, J. M., Pickett, S. T. A. & Redman, C. L. Integrated approaches to long-term studies of urban ecological systems. Bioscience 50, 571–584 (2000).Article 

Google Scholar 
Vandewalle, M. et al. Functional traits as indicators of biodiversity response to land use changes across ecosystems and organisms. Biodivers. Conserv. 19, 2921–2947 (2010).Article 

Google Scholar 
Williams, N. S. G. et al. A conceptual framework for predicting the effects of urban environments on floras. J. Ecol. 97, 4–9 (2009).Article 

Google Scholar 
Cavender-Bares, J. et al. Horticultural availability and homeowner preferences drive plant diversity and composition in urban yards. Ecol. Appl. 30, 1–16 (2020).Article 

Google Scholar 
Pearse, W. D. et al. Homogenization of plant diversity, composition, and structure in North American urban yards. Ecosphere 9, e02105 (2018).Article 

Google Scholar 
Cubino, J. P. et al. Drivers of plant species richness and phylogenetic composition in urban yards at the continental scale. Landsc. Ecol. 34, 63–77 (2019).Article 

Google Scholar 
Oke, T. R. The energetic basis of the urban heat island. Q. J. R. Meteorol. Soc. 108, 1–24 (1982).
Google Scholar 
Sukopp, H. Human-caused impact on preserved vegetation. Landsc. Urban Plan. 68, 347–355 (2004).Article 

Google Scholar 
Díaz, S. et al. Incorporating plant functional diversity effects in ecosystem service assessments. Proc. Natl. Acad. Sci. USA 104, 20684–20689 (2007).Article 

Google Scholar 
Williams, N. S. G., Hahs, A. K. & Vesk, P. A. Urbanisation, plant traits and the composition of urban floras. Perspect. Plant Ecol. Evol. Syst. 17, 78–86 (2015).Article 

Google Scholar 
Teskey, R. et al. Responses of tree species to heat waves and extreme heat events. Plant Cell Environ. 38, 1699–1712 (2015).Article 

Google Scholar 
Jochner, S. & Menzel, A. Urban phenological studies—past, present, future. Environ. Pollut. 203, 250–261 (2015).Article 

Google Scholar 
Cleland, E. E., Chuine, I., Menzel, A., Mooney, H. A. & Schwartz, M. D. Shifting plant phenology in response to global change. Trends Ecol. Evol. 22, 357–365 (2007).Article 

Google Scholar 
de Bello, F. et al. Towards an assessment of multiple ecosystem processes and services via functional traits. Biodivers. Conserv. 19, 2873–2893 (2010).Article 

Google Scholar 
Santangelo, J. S. et al. Global urban environmental change drives adaptation in white clover. Science 375, 1275–1281 (2022).Article 

Google Scholar 
Blois, J. L., Zarnetske, P. L., Fitzpatrick, M. C. & Finnegan, S. Climate change and the past, present, and future of biotic interactions. Science 341, 499–504 (2013).Article 

Google Scholar 
Martin, C. A., Warren, P. S. & Kinzig, A. P. Neighborhood socioeconomic status is a useful predictor of perennial landscape vegetation in residential neighborhoods and embedded small parks of Phoenix, AZ. Landsc. Urban Plan. 69, 355–368 (2004).Article 

Google Scholar 
Kinzig, A. P., Warren, P., Martin, C., Hope, D. & Katti, M. The effects of human socioeconomic status and cultural characteristics on urban patterns of biodiversity. Ecol. Soc. 10, 23 (2005).Article 

Google Scholar 
Stephenson, J. The cultural values model: An integrated approach to values in landscapes. Landsc. Urban Plan. 84, 127–139 (2008).Article 

Google Scholar 
Andersson, E., Barthel, S. & Ahrné, K. Measuring social–ecological dynamics behind the generation of ecosystem services. Ecol. Appl. 17, 1267–1278 (2007).Article 

Google Scholar 
Fraser, E. D. G. & Kenney, W. A. Cultural background and landscape history as factors affecting perceptions of the urban forest. J. Arboric. 26, 106–113 (2000).
Google Scholar 
Hope, D. et al. Socioeconomics drive urban plant diversity. Proc. Natl. Acad. Sci. USA 100, 8788–8792 (2003).Article 

Google Scholar 
Avolio, M. L. et al. Understanding preferences for tree attributes: The relative effects of socio-economic and local environmental factors. Urban Ecosyst. 18, 73–86 (2015).Article 

Google Scholar 
Körmöndi, B., Tempfli, J., Kocsis, J. B., Adams, J. & Szkordilisz, F. E. The secret ingredient—The role of governance in green infrastructure development: Through the examples of European cities. IOP Conf. Ser. Earth Environ. Sci. 323, (2019).Conway, T. M. & Vander Vecht, J. Growing a diverse urban forest: species selection decisions by practitioners planting and supplying trees. Landsc. Urban Plan. 138, 1–10 (2015).Article 

Google Scholar 
Lack, W. H. The Book of Palms (Taschen-Bibliotheca Universalis, 2015).Grilo, F. et al. Using green to cool the grey: Modelling the cooling effect of green spaces with a high spatial resolution. Sci. Total Environ. 724, 138182 (2020).Article 

Google Scholar 
Prasifka, J. R. et al. Using nectar-related traits to enhance crop–pollinator interactions. Front. Plant Sci. 9, 1–8 (2018).Article 

Google Scholar 
Veerkamp, C. J. et al. A review of studies assessing ecosystem services provided by urban green and blue infrastructure. Ecosyst. Serv. 52, 101367 (2021).Article 

Google Scholar 
Theodorou, P. et al. Urban areas as hotspots for bees and pollination but not a panacea for all insects. Nat. Commun. 11, 576 (2020).Article 

Google Scholar 
Farmer, J. Trees in Paradise: A California History (WW Norton & Company, 2013).Goodness, J., Andersson, E., Anderson, P. M. L. & Elmqvist, T. Exploring the links between functional traits and cultural ecosystem services to enhance urban ecosystem management. Ecol. Indic. 70, 597–605 (2016).Article 

Google Scholar 
Masterson, V. A. et al. The contribution of sense of place to social-ecological systems research: A review and research agenda. Ecol. Soc. 22, 49 (2017).Article 

Google Scholar 
Masterson, V. A., Enqvist, J. P., Stedman, R. C. & Tengö, M. Sense of place in social-ecological systems: From theory to empirics. Sustain. Sci. 14, 555–564 (2019).Article 

Google Scholar 
Mukherjee, A. & Agrawal, M. Use of GLM approach to assess the responses of tropical trees to urban air pollution in relation to leaf functional traits and tree characteristics. Ecotoxicol. Environ. Saf. 152, 42–54 (2018).Article 

Google Scholar 
Singh, S. K., Rao, D. N., Agrawal, M., Pandey, J. & Naryan, D. Air pollution tolerance index of plants. J. Environ. Manage. 32, 45–55 (1991).Article 

Google Scholar 
Mukherjee, A. & Agrawal, M. Pollution response score of tree species in relation to ambient air quality in an urban area. Bull. Environ. Contam. Toxicol. 96, 197–202 (2016).Article 

Google Scholar 
Barwise, Y. & Kumar, P. Designing vegetation barriers for urban air pollution abatement: A practical review for appropriate plant species selection. npj Clim. Atmos. Sci. 3, 12 (2020).Article 

Google Scholar 
Grote, R. et al. Functional traits of urban trees: Air pollution mitigation potential. Front. Ecol. Environ. 14, 543–550 (2016).Article 

Google Scholar 
Tomson, M. et al. Green infrastructure for air quality improvement in street canyons. Environ. Int. 146, 106288 (2021).Article 

Google Scholar  More

Pecl, G. et al. Biodiversity redistribution under climate change: impacts on ecosystems and human well-being. Science 355, eaai9214 (2017).Article 
PubMed 

Google Scholar 
Sunday, J. M., Bates, A. E. & Dulvy, N. K. Thermal tolerance and the global redistribution of animals. Nat. Clim. Change 2, 686–690 (2012).Article 

Google Scholar 
Pinsky, M. L., Eikeset, A. M., McCauley, D. J., Payne, J. L. & Sunday, J. M. Greater vulnerability to warming of marine versus terrestrial ectotherms. Nature 569, 108–111 (2019).Article 
PubMed 

Google Scholar 
Lenoir, J. et al. Species better track climate warming in the oceans than on land. Nat. Ecol. Evol. 4, 1044–1059 (2020).Article 
PubMed 

Google Scholar 
Poloczanska, E. S. et al. Global imprint of climate change on marine life. Nat. Clim. Change 3, 919–925 (2013).Article 

Google Scholar 
Sanford, E., Sones, J. L., García-Reyes, M., Goddard, J. H. & Largier, J. L. Widespread shifts in the coastal biota of northern California during the 2014–2016 marine heatwaves. Sci. Rep. 9, 4216 (2019).Article 
PubMed 
PubMed Central 

Google Scholar 
Molinos, J. G., Burrows, M. & Poloczanska, E. Ocean currents modify the coupling between climate change and biogeographical shifts. Sci. Rep. 7, 1–9 (2017).
Google Scholar 
Sunday, J. M. et al. Species traits and climate velocity explain geographic range shifts in an ocean‐warming hotspot. Ecol. Lett. 18, 944–953 (2015).Article 
PubMed 

Google Scholar 
Figueira, W. F., Curley, B. & Booth, D. J. Can temperature-dependent predation rates regulate range expansion potential of tropical vagrant fishes? Mar. Biol. 166, 73 (2019).Article 

Google Scholar 
Champion, C. & Coleman, M. A. Seascape topography slows predicted range shifts in fish under climate change. Limnol. Oceanogr. Lett. 6, 143–153 (2021).Article 

Google Scholar 
Roberts, S. M., Boustany, A. M. & Halpin, P. N. Substrate-dependent fish have shifted less in distribution under climate change. Commun. Biol. 3, 1–7 (2020).Article 

Google Scholar 
Engelhard, G. H., Righton, D. A. & Pinnegar, J. K. Climate change and fishing: a century of shifting distribution in North Sea cod. Glob. Change Biol. 20, 2473–2483 (2014).Article 

Google Scholar 
Twiname, S. et al. A cross‐scale framework to support a mechanistic understanding and modelling of marine climate‐driven species redistribution, from individuals to communities. Ecography 43, 1764–1778 (2020).Article 

Google Scholar 
Bates, A. E. et al. Defining and observing stages of climate-mediated range shifts in marine systems. Glob. Environ. Change 26, 27–38 (2014).Article 

Google Scholar 
Fogarty, H. E., Burrows, M. T., Pecl, G. T., Robinson, L. M. & Poloczanska, E. S. Are fish outside their usual ranges early indicators of climate‐driven range shifts? Glob. Change Biol. 23, 2047–2057 (2017).Article 

Google Scholar 
Jiguet, F. & Barbet‐Massin, M. Climate change and rates of vagrancy of Siberian bird species to Europe. Ibis 155, 194–198 (2013).Article 

Google Scholar 
Burrows, M. T. et al. Geographical limits to species-range shifts are suggested by climate velocity. Nature 507, 492–495 (2014).Article 
PubMed 

Google Scholar 
Peck, M. A. et al. Projecting changes in the distribution and productivity of living marine resources: a critical review of the suite of modeling approaches used in the large European project VECTORS. Estuar., Coast. Shelf Sci. 201, 40–55 (2016).Article 

Google Scholar 
Brito-Morales, I. et al. Climate velocity can inform conservation in a warming world. Trends Ecol. Evol. 33, 441–457 (2018).Article 
PubMed 

Google Scholar 
Pinsky, M. L., Worm, B., Fogarty, M. J., Sarmiento, J. L. & Levin, S. A. Marine taxa track local climate velocities. Science 341, 1239–1242 (2013).Article 
PubMed 

Google Scholar 
Champion, C., Hobday, A. J., Zhang, X., Pecl, G. T. & Tracey, S. R. Changing windows of opportunity: past and future climate-driven shifts in temporal persistence of kingfish (Seriola lalandi) oceanographic habitat within south-eastern Australian bioregions. Mar. Freshw. Res. 70, 33–42 (2019).Article 

Google Scholar 
Pinsky, M. L., Selden, R. L. & Kitchel, Z. J. Climate-driven shifts in marine species ranges: scaling from organisms to communities. Annu. Rev. Mar. Sci. 12, 153–179 (2020).Article 

Google Scholar 
Lonhart, S. I., Jeppesen, R., Beas-Luna, R., Crooks, J. A. & Lorda, J. Shifts in the distribution and abundance of coastal marine species along the eastern Pacific Ocean during marine heatwaves from 2013 to 2018. Mar. Biodivers. Rec. 12, 1–15 (2019).Article 

Google Scholar 
Wernberg, T. et al. An extreme climatic event alters marine ecosystem structure in a global biodiversity hotspot. Nat. Clim. Change 3, 78–82 (2013).Article 

Google Scholar 
Lenanton, R., Dowling, C., Smith, K., Fairclough, D. & Jackson, G. Potential influence of a marine heatwave on range extensions of tropical fishes in the eastern Indian Ocean—Invaluable contributions from amateur observers. Regional Stud. Mar. Sci. 13, 19–31 (2017).Article 

Google Scholar 
Leriorato, J. C. & Nakamura, Y. Unpredictable extreme cold events: a threat to range-shifting tropical reef fishes in temperate waters. Mar. Biol. 166, 1–10 (2019).Article 

Google Scholar 
Hobday, A. J. & Pecl, G. T. Identification of global marine hotspots: sentinels for change and vanguards for adaptation action. Rev. Fish. Biol. Fish. 24, 415–425 (2014).Article 

Google Scholar 
Pecl, G. T. et al. Redmap Australia: challenges and successes with a large-scale citizen science-based approach to ecological monitoring and community engagement on climate change. Front. Mar. Sci. 6, 349 (2019).Article 

Google Scholar 
Jacox, M. G., Alexander, M. A., Bograd, S. J. & Scott, J. D. Thermal displacement by marine heatwaves. Nature 584, 82–86 (2020).Article 
PubMed 

Google Scholar 
Brown, C. J. et al. Ecological and methodological drivers of species’ distribution and phenology responses to climate change. Glob. Change Biol. 22, 1548–1560 (2016).Article 

Google Scholar 
Fuchs, H. L. et al. Wrong-way migrations of benthic species driven by ocean warming and larval transport. Nat. Clim. Change 10, 1052–1056 (2020).Article 

Google Scholar 
Rooney, N., McCann, K. S. & Moore, J. C. A landscape theory for food web architecture. Ecol. Lett. 11, 867–881 (2008).Article 
PubMed 

Google Scholar 
Feary, D. A. et al. Latitudinal shifts in coral reef fishes: why some species do and others do not shift. Fish. Fish. 15, 593–615 (2014).Article 

Google Scholar 
Beissinger, S. R. & Riddell, E. A. Why are species’ traits weak predictors of range shifts? Ann. Rev. Ecol. Evol. Syst. 52, 47–66 (2021).Pearce, A. F. & Feng, M. The rise and fall of the “marine heat wave” off Western Australia during the summer of 2010/2011. J. Mar. Syst. 111, 139–156 (2013).Article 

Google Scholar 
Oliver, E. C. et al. The unprecedented 2015/16 Tasman Sea marine heatwave. Nat. Commun. 8, 16101 (2017).Article 
PubMed 
PubMed Central 

Google Scholar 
Gervais, C. R., Champion, C. & Pecl, G. T. Species on the move around the Australian coastline: a continental scale review of climate‐driven species redistribution in marine systems. Glob. Change Biol. 27, 3200–3217 (2021).Article 

Google Scholar 
Nursey-Bray, M., Palmer, R. & Pecl, G. Spot, log, map: assessing a marine virtual citizen science program against Reed’s best practice for stakeholder participation in environmental management. Ocean Coast. Manag. 151, 1–9 (2018).Article 

Google Scholar 
Pecl, G. T. et al. Ocean warming hotspots provide early warning laboratories for climate change impacts. Rev. Fish. Biol. Fish. 24, 409–413 (2014).Article 

Google Scholar 
Stuart-Smith, J. et al. Southernmost records of two Seriola species in an Australian ocean-warming hotspot. Mar. Biodivers. 48, 1579–1582 (2018).Article 

Google Scholar 
Provoost, P. & Bosch, S. robis: R Client to access data from the OBIS API. Ocean Biogeographic Information System. Intergovernmental Oceanographic Commission of UNESCO. R package version 2.1.8, https://cran.r-project.org/package=robis (2019).Froese, R. & Pauly, D. (eds). FishBase. World Wide Web electronic publication. www.fishbase.org. (2022). Accessed 14 July 2019.ABRS. Australian Faunal Directory. Australian Biological Resources Study, Canberra. https://biodiversity.org.au/afd/home. (2020). Accessed 15 July 2019.Robinson, L. M. et al. Rapid assessment of an ocean warming hotspot reveals “high” confidence in potential species’ range extensions. Glob. Environ. Change 31, 28–37 (2015).Article 

Google Scholar 
Hijmans, R. J. raster: geographic data analysis and modeling. R package version 3.4-5. https://CRAN.R-project.org/package=raster (2020).van Etten, J. R package gdistance: distances and routes on geographical grids. J. Stat. Softw. 76, 1–21 (2017).
Google Scholar 
Hobday, A. J. et al. A hierarchical approach to defining marine heatwaves. Prog. Oceanogr. 141, 227–238 (2016).Article 

Google Scholar 
Molinos, J. G., Schoeman, D. S., Brown, C. J. & Burrows, M. T. VoCC: an r package for calculating the velocity of climate change and related climatic metrics. Methods Ecol. Evol. 10, 2195–2202 (2019).Article 

Google Scholar 
Schlegel, R. W. & Smit, A. J. heatwaveR: a central algorithm for the detection of heatwaves and cold-spells. J. Open Source Softw. 3, 821 (2018).Article 

Google Scholar 
Venables, W. N. & Ripley, B. D. Modern applied statistics with S. 4th edn, (Springer, 2002).Lüdecke, D. sjPlot: data visualization for statistics in social science. R package version 2.8.6. https://CRAN.R-project.org/package=sjPlot (2020). More

Rogers, A. et al. Antarctic futures: An assessment of climate-driven changes in ecosystem structure, function, and service provisioning in the southern ocean. Annual Review of Marine Science 12, 87–120, https://doi.org/10.1146/annurev-marine-010419-011028 (2020).Article 
PubMed 

Google Scholar 
Tin, T. et al. Impacts of local human activities on the antarctic environment. Antarctic Science 21, 3–33, https://doi.org/10.1017/S0954102009001722 (2009).Article 

Google Scholar 
Pineda-Metz, S. E. A., Gerdes, D. & Richter, C. Benthic fauna declined on a whitening antarctic continental shelf. Nature Communications 11, 2226, https://doi.org/10.1038/s41467-020-16093-z (2020).Article 
PubMed 
PubMed Central 

Google Scholar 
Convey, P. Antarctic terrestrial biodiversity in a changing world. Polar Biology 34, 1629, https://doi.org/10.1007/s00300-011-1068-0 (2011).Article 

Google Scholar 
Kang, Y. H. et al. Composition and structure of the marine benthic community in terra nova bay, antarctica: Responses of the benthic assemblage to disturbances. PLOS ONE 14, 1–16, https://doi.org/10.1371/journal.pone.0225551 (2019).Article 

Google Scholar 
Piazza, P. et al. Underwater photogrammetry in antarctica: long-term observations in benthic ecosystems and legacy data rescue. Polar Biology 42, 1061–1079, https://doi.org/10.1007/s00300-019-02480-w (2019).Article 

Google Scholar 
SOOS. Southern Ocean Observing System – Report on the 2017 Ross Sea Working Group Meeting. http://www.soos.aq. [Online; accessed 2022/11/15] (2017).SCAR. Scientific Committee on Antarctic Research. https://www.scar.org. [Online; accessed 2022/11/15] (2021).ANTOS. Antarctic near-shore and terrestrial observing system. https://www.scar.org/science/antos/home. [Online; accessed 2022/11/15] (2015).Dayton, P. K. et al. Benthic responses to an antarctic regime shift: food particle size and recruitment biology. Ecological Applications 29, e01823, https://doi.org/10.1002/eap.1823 (2019).Article 
PubMed 
PubMed Central 

Google Scholar 
Watters, G. M., Hinke, J. T. & Reiss, C. S. Long-term observations from antarctica demonstrate that mismatched scales of fisheries management and predator-prey interaction lead to erroneous conclusions about precaution. Scientific Reports 10, 2314, https://doi.org/10.1038/s41598-020-59223-9 (2020).Article 
PubMed 
PubMed Central 

Google Scholar 
Bolinesi, F. et al. Spatial-related community structure and dynamics in phytoplankton of the ross sea, antarctica. Frontiers in Marine Science 7, https://doi.org/10.3389/fmars.2020.574963 (2020).Stenni, B. et al. Three-year monitoring of stable isotopes of precipitation at concordia station, east antarctica. The Cryosphere 10, 2415–2428, https://doi.org/10.5194/tc-10-2415-2016 (2016).Article 

Google Scholar 
Ramesh, K. & Soni, V. Perspectives of antarctic weather monitoring and research efforts. Polar Science 18, 183–188, https://doi.org/10.1016/j.polar.2018.04.005 (2018). Recent Advances in Climate Science of Polar Region (to commemorate the contributions of Late Dr. S.Z. Qasim, a pioneering doyen of the Indian Polar programme).Article 

Google Scholar 
Shepherd, A. et al. Mass balance of the antarctic ice sheet from 1992 to 2017. Nature 558, 219–222, https://doi.org/10.1038/s41586-018-0179-y (2018).Article 

Google Scholar 
Budge, J. S. & Long, D. G. A comprehensive database for antarctic iceberg tracking using scatterometer data. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 11, 434–442, https://doi.org/10.1109/JSTARS.2017.2784186 (2018).Article 

Google Scholar 
Rignot, E. et al. Four decades of antarctic ice sheet mass balance from 1979–2017. Proceedings of the National Academy of Sciences of the United States of America 116, 1095–1103, https://doi.org/10.1073/pnas.1812883116 (2019).Article 
PubMed 
PubMed Central 

Google Scholar 
Barbat, M. M., Rackow, T., Wesche, C., Hellmer, H. H. & Mata, M. M. Automated iceberg tracking with a machine learning approach applied to sar imagery: A weddell sea case study. ISPRS Journal of Photogrammetry and Remote Sensing 172, 189–206, https://doi.org/10.1016/j.isprsjprs.2020.12.006 (2021).Article 

Google Scholar 
Aguzzi, J. et al. New high-tech flexible networks for the monitoring of deep-sea ecosystems. Environmental Science & Technology 53, 6616–6631, https://doi.org/10.1021/acs.est.9b00409 (2019).Article 

Google Scholar 
Piazza, P., Gattone, S., Guzzi, A. & Schiaparelli, S. Towards a robust baseline for long-term monitoring of antarctic coastal benthos. Hydrobiologia 847, 1753–1771, https://doi.org/10.1007/s10750-020-04177-2 (2020).Article 

Google Scholar 
Rountree, R. et al. Towards an optimal design for ecosystem-level ocean observatories. Oceanography and Marine Biology 58, 79–105, https://doi.org/10.1201/9780429351495-2 (2020).Article 

Google Scholar 
Katsanevakis, S. et al. Monitoring marine populations and communities: Methods dealing with imperfect detectability. Aquatic Biology 16, 31–52, https://doi.org/10.3354/ab00426 (2012).Article 

Google Scholar 
Zampoukas, N. et al. Technical guidance on monitoring for the marine strategy framework directive. Tech. Rep., European Commission, Report EUR 26499 (2014).Bicknell, A. W., Godley, B. J., Sheehan, E. V., Votier, S. C. & Witt, M. J. Camera technology for monitoring marine biodiversity and human impact. Frontiers in Ecology and the Environment 14, 424–432, https://doi.org/10.1002/fee.1322 (2016).Article 

Google Scholar 
European Marine Board. Working Group on Big Data in Marine Science. https://www.marineboard.eu/publications/big-data-marine-science. [Online; accessed 2022/11/15] (2020).Zurowietz, M. & Nattkemper, T. W. Current trends and future directions of large scale image and video annotation: Observations from four years of biigle 2.0. Frontiers in Marine Science 8, https://doi.org/10.3389/fmars.2021.760036 (2021).Kim, S. L., Thurber, A., Hammerstrom, K. & Conlan, K. Seastar response to organic enrichment in an oligotrophic polar habitat. Journal of Experimental Marine Biology and Ecology 346, 66–75, https://doi.org/10.1016/j.jembe.2007.03.004 (2007).Article 

Google Scholar 
Peirano, A., Bordone, A., Marini, S., Piazza, P. & Schiaparelli, S. A simple time-lapse apparatus for monitoring macrozoobenthos activity in antarctica. Antarctic Science 28, 473–474, https://doi.org/10.1017/S0954102016000377 (2016).Article 

Google Scholar 
Peirano, A., Marini, S., Bordone, A. & Schiaparelli, S. ICE-LAPSE: Analysis of antarctic benthos dynamics by using non-destructive monitoring devices and permanent stations, pnra 2013/az1.16, funded by the italian national antarctic program (2015-2016).Marini, S. et al. Long-term automated visual monitoring of antarctic benthic fauna. Methods in Ecology and Evolution 13, 1746–1764, https://doi.org/10.1111/2041-210X.13898 (2022).Article 

Google Scholar 
Marini, S. et al. EP2863257 (A1) – Underwater images acquisition and processing system. https://data.epo.org/gpi/EP2863257B1. [Online; accessed 2022/11/15] (2013).Corgnati, L. et al. Looking inside the ocean: Toward an autonomous imaging system for monitoring gelatinous zooplankton. Sensors 16, 2124, https://doi.org/10.3390/s16122124 (2016).Article 
PubMed 
PubMed Central 

Google Scholar 
Marini, S. et al. Automated estimate of fish abundance through the autonomous imaging device guard1. Measurement 126, 72–75, https://doi.org/10.1016/j.measurement.2018.05.035 (2018).Article 

Google Scholar 
Pensieri, S. et al. Environmental acoustic noise observations in tethys bay (terra nova bay, ross sea, antarctica). In 2014 Oceans – St. John’s, 1–6, https://doi.org/10.1109/OCEANS.2014.7003196 (2014).Jung, J. et al. Multibeam bathymetry and distribution of clay minerals on surface sediments of a small bay in terra nova bay, antarctica. Minerals 11, https://doi.org/10.3390/min11010072 (2021).Balog, I. et al. Estimation of direct normal irradiance at antarctica for concentrated solar technology. Applied System Innovation 2, https://doi.org/10.3390/asi2030021 (2019).Caputi, S. S. et al. Seasonal food web dynamics in the antarctic benthos of tethys bay (ross sea): Implications for biodiversity persistence under different seasonal sea-ice coverage. Frontiers in Marine Science 7, 1046, https://doi.org/10.3389/fmars.2020.594454 (2020).Article 

Google Scholar 
van Leeuwe, M. A. et al. Annual patterns in phytoplankton phenology in antarctic coastal waters explained by environmental drivers. Limnology and Oceanography 65, 1651–1668, https://doi.org/10.1002/lno.11477 (2020).Article 

Google Scholar 
OEngineering. OEngineering s.r.l. – GUARD-1, Underwater Autonomous Smart Camera. https://www.oengineering.eu//GUARD-1/. [Online; accessed 2022/11/15] (2021).Magic Lantern. https://magiclantern.fm. [Online; accessed 2022/11/15] (2021).Marini, S. et al. Guard1: An autonomous system for gelatinous zooplankton image-based recognition. In OCEANS 2015 – Genova, 1–7, https://doi.org/10.1109/OCEANS-Genova.2015.7271704 (2015).CR2. The Canon RAW (CRW) File Format. https://exiftool.org/canon_raw.html. [Online; accessed 2022/11/15] (2022).Marini, S. et al. ICE-LAPSE image dataset. Zenodo https://doi.org/10.5281/zenodo.6418163 (2022).LabelImg. A graphical image annotation tool. https://github.com/tzutalin/labelImg. [Online; accessed 2022/11/15] (2021).Schoening, T. et al. Making marine image data fair. Scientific Data 9, 414, https://doi.org/10.1038/s41597-022-01491-3 (2022).Article 
PubMed 
PubMed Central 

Google Scholar 
Cattaneo-Vietti, R., Chiantore, M., Schiaparelli, S. & Albertelli, G. Shallow- and deep-water mollusc distribution at terra nova bay (ross sea, antarctica). Polar Biology 23, 173–182, https://doi.org/10.1007/s003000050024 (2000).Article 

Google Scholar 
Cattaneo-Vietti, R. et al. Spatial and Vertical Distribution of Benthic Littoral Communities in Terra Nova Bay, 503–514 (Springer Berlin Heidelberg, Berlin, Heidelberg, 2000).Cummings, V. J. et al. Linking ross sea coastal benthic communities to environmental conditions: Documenting baselines in a spatially variable and changing world. Frontiers in Marine Science 5, 232, https://doi.org/10.3389/fmars.2018.00232 (2018).Article 

Google Scholar 
Redmon, J., Divvala, S., Girshick, R. & Farhadi, A. You only look once: Unified, real-time object detection. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 779–788, https://doi.org/10.1109/CVPR.2016.91 (2016).YOLO V5. https://github.com/ultralytics/yolov5. [Online; accessed 2022/11/15] (2022). More

Surface morphology analysisSEM images were shown in Fig. 1. It showed that the contour of three soils were fairly clear before adsorption. But it became fuzzier and the degree of cementation was increased when phenanthrene was adsorbed on the soils. According to the surface morphology, the silty sand (A) had furrows on the surface before adsorption compared with the fairly smooth without any furrows after adsorption (B). The silts (C) were flaky and the lamellar accumulation decreased (D). The loess (E) had a smooth surface with some flaky and rod like structure, after adsorption (F), the surface of loess increased in clay-like structure.Figure 1SEM micrographs of the three soil samples. (A) Silty sand; (B) Adsorbing 5 h of Silty sand; (C) Silts; (D) Adsorbing 5 h of Silts; (E) Loess; (F) Adsorbing 5 h of Loess.Full size imageAdsorption and desorption experimentsAdsorption and desorption kineticsAdsorption kinetics is one of the most important characteristics governing solute uptake rate and represents adsorption efficiency33. The sorption and desorption kinetics of phenanthrene in three soils were shown in Fig. 2. The results showed that the adsorption processes among all soils were similar. The kinetics of phenanthrene in soils was completed in two steps: a “fast” adsorption and a “slow” adsorption. The adsorption amount increased during 0-18h. It was a rapid reaction from 0 to 200 minutes. From 200 to 600 minutes, the adsorption amount increased slightly into balance. This phenomenon was due to the adsorption of phenanthrene occurred on the surface of soil organic matter. With the increase of time, soil surface adsorption sites were gradually saturated, causing the decrease of adsorption rate until reaching the equilibrium. Phenanthrene was a hydrophobic substance. It was easy to reach the soil surface and adhere to the grain surface. The results were consistent with the study of had also found that the balance time was approximately 18h and the adsorption amount increased with the adsorption reaction time34. Under the same conditions, loess had the highest adsorption capacity, which was mainly due to the highest organic content 18. The maximum phenanthrene sorption capacities ranked as follows: loess > silty sand > silts. As shown in Fig. 2, phenanthrene desorption in soils was relatively quick and its desorption equilibrium time was 3h. To reach an adequate desorption balance while remaining consistent with the adsorption reaction time, the balance time of the adsorption–desorption experiment was set at 18h. Generally, PAHs below 4 cycles could reach the adsorption equilibrium for about 16~24h.Figure 2(a)Adsorption equilibration curves of phenanthrene sorption in soils. (b) Desorption equilibration curves of phenanthrene sorption in soils.Full size imagePseudo-second-order and Elovich models were used to study the phenanthrene adsorption mechanism (Table 3). Phenanthrene sorption kinetics were satisfactorily described by a pseudo-second-order model with coefficients of determination (R2) ranging from 0.99875 to 0.99847, compared with R2 values of 0.26508–0.73901 for the Elovich model. This well-fitting pseudo-second-order model indicated that the rate-limiting step was chemical adsorption, including electronic forces through sharing or exchange of electrons35,36. Moreover, it suggested that sorption was governed by the availability of sorption sites on the soil surfaces instead of by the phenanthrene concentration in solution.Table 3 Constants and coeffients of determination of Pseudo-second-order kinetics and Elovich models of sorption.Full size tableAdsorption and desorption isothermsThe isotherm was used for quantitative analysis of phenanthrene transport from liquid to solid phase and for understanding the nature of interactions between phenanthrene and the soil matrix. The sorption and desorption isotherms of phenanthrene in soils were shown in Fig. 3. The data showed that phenanthrene adsorption and desorption capacities of three soils varied markedly due to their different physicochemical properties. With the increase of phenanthrene concentration, the adsorbed amount increased. At the same temperature, the adsorption capacity of silty sand was minimum while loess was maximum. This is mainly related to the soil physicochemical properties. At the same initial concentration, the temperature increase from 20 °C to 40 °C showed that the adsorption and desorption capacity decreased with temperature increase. On the one hand, the rise of temperature can increase the phenanthrene solubility in the liquid phase. On the other hand, it could reduce various forces between the soil surface and phenanthrene37.Figure 3(a)20 °C adsorption isotherms for phenanthrene in soils. (b)30 °C adsorption isotherms for phenanthrene in soils. (c)40 °C adsorption isotherms for phenanthrene in soils. (d) 20 °C desorption isotherms for phenanthrene in soils. (e) 30 °C desorption isotherms for phenanthrene in soils. (f) 40 °C desorption isotherms for phenanthrene in soils.Full size imageThe Freundlich isotherm was used mainly for adsorption surfaces with nonuniform energy distribution, and the Langmuir isotherm was used for monolayer adsorption on perfectly smooth and homogeneous surfaces38. The experimental data were fitted with the Langmuir and Freundlich adsorption models, and the isotherm parameters logKF, 1/n, KL, qmax and the coefficient of determination (R2) of phenanthrene in soils were listed in Table 4.Table 4 Isotherm parameters for Phenanthrene sorption in soils.Full size tableAs shown in Table 4, according to the coefficients of determination (R2), all soils were better fitted with the Freundlich model, which assumes that phenanthrene sorption and desorption occurs on a heterogeneous surface with the possibility of sorption being multi-layered39. This phenomenon has also been observed in humic acid and nanometer clay mineral40. It showed that the soil adsorption of organic matter was not only surface adsorption but also the process of soil organic matter distribution41,42,43 reached the equilibrium isotherm fitted well with the Freundlich equation when studying the adsorption behavior of aromatic compounds by solids.Adsorption and desorption thermodynamicsTo clarify the adsorption mechanisms, the thermodynamic parameters mentioned earlier were calculated and presented in Table 5. Generally, the value of Gibbs free energy changeΔG0 indicated the spontaneity of a chemical reaction. Therefore, it could evaluate whether sorption was relate to spontaneous interaction44. Negative values of ΔG0 indicated that the feasibility and spontaneous nature. The research was under the temperature range about 293–313 K. For adsorption process, all soils ΔG0 was  0 and desorption ΔH  1, P  temperature  > phenanthrene concentration  > pH. In the interaction, the phenanthrene concentration and organic matter have a significant effect on the silt adsorption rate. The coefficient of determination of the silt complex correlation is R2 = 0.9464, indicating that the response model has a good fit, and the experimental error is within the acceptable range. Adjusting the complex correlation coefficient R2 = 0.8982 indicates that the regression relationship can explain 89.82% of the change in the dependent variable. Therefore, this The model can be used to analyze and predict the effect of different factors on the adsorption rate of phenanthrene.3D response surface analysisIn response surface optimization, the three-dimensional response surface graph reflects the influence of the interaction of the other two variables on the response value, and the slope of the response surface reflects the significance of the interaction of the two variables on the response value. The more significant the interaction effect is on the response value, when the slope is gentle, the effect is not significant. If the contour map is elliptical, it indicates that the interaction between the two variables is significant, and if the contour map is circular, it is not significant46. In addition, the slope and density of the contour line also reflect the influence of the variable on the response value. The steeper the contour line and the greater the density, the greater the influence of the variable on the response value47.

(1) Loess Fig. 5 is a three-dimensional response surface diagram of the interaction between initial phenanthrene concentration and pH to phenanthrene adsorption on loess. It can be seen from the figure that the slope of the response surface graph is steep, and the contour line is an approximate circle, indicating that the interaction between phenanthrene concentration and pH is not significant for the response value. With the increase of pH, the adsorption rate of phenanthrene on loess showed a slow decline at first to the lowest point at 6, and then gradually increased. When the soil pH was close to 6, with the increase of the initial phenanthrene concentration, the adsorption rate of loess also showed a trend of first decreasing and then increasing. According to the F value, F = 0.337, P = 0.5532  > 0.05, it can be concluded that soil pH and initial phenanthrene concentration of the solution have no significant interaction on the adsorption rate of loess.

Figure 6 shows the effects of initial phenanthrene concentration and organic matter on phenanthrene adsorption on Loess under the condition that pH value and temperature are at the central point. It can be seen from the figure that the initial phenanthrene concentration and soil organic matter contour are steep, indicating that their interaction is significant. The range of phenanthrene adsorption rate is 70 ~ 95, and the change of surface is steep. From the Loess error analysis, it can be seen that if f value is 6.05 and P value is 0.0275  More