Remote sensing at the interface between ecology and climate sciences

Anthropogenic climate change is causing a severe impact on the survival of organisms (Brondizio et al., 2019; Shukla 2019). Climate one major drivers species distributions, and velocity at which current changing, owing to human activities, already induces redistribution life Earth an unprecedented speed, especially in oceans where marine shifting towards poles six times faster than terrestrial poleward land (Lenoir 2020). Besides, biodiversity may enhance warming through positive feedback loops (e.g., shrubification Arctic altering surface albedo), thus indirectly affecting well-being (Pecl 2017). For this reason, both essential variables (ECVs) (EBVs) have been developed as proxies for early detection (Bojinski 2014; Pereira 2013; Schmeller 2018). Worldwide, organizations tried building research policy networks seriously face problem, or, least, build robust projections that might allow monitoring future changes. One main examples Group Observation (GEO, https://www.earthobservations.org/index.php), namely global partnership governments develop Global System Systems (GEOSS, https://www.earthobservations.org/geoss.php). A specific branch GEO – BON (https://geobon.org) devoted development suitable EBVs improving effectiveness conservation over entire planet. The 22 proposed are trying cover several aspects diversity, clumped into classes: (i) genetic composition co-ancestry, allelic diversity); (ii) populations distribution, population abundance); (iii) traits phenology, morphology); (iv) community taxonomic functional interactions); (v) ecosystem function net primary productivity, nutrient retention); (vi) structure habitat structure, extent fragmentation) (Pereira see also https://geobon.org/ebvs/what-are-ebvs/). Variables useful fields interface ecology sciences include, among others: occurrence plant leaf area nitrogen content) based multispectral hyperspectral data; canopy height biomass light ranging (LiDAR) data (see, e.g., Ecosystem Dynamics Investigation mission [GEDI, https://gedi.umd.edu/]); fragmentation land-cover (LCC) time series Landsat images (Skidmore 2015). common language between remote sensing community, hand, ecological communities, other, was definitively defined by proposal ecosystem-based can be transposed from field observations remotely-sensed perspectives. key concept reinforcement link patterns observed direct relationship with potential processes shaping them. instance, spatial heterogeneity biophysical conditions represent good proxy variation space related processes, niche variability, (Hernandez-Stefanoni 2012). From point view, requirements explicitly expressed communities ecologists scientists specialists include availability different grains well meaningful wavelengths further process perform global-scale analyses (Randin 2020; In era big data, such accomplishments reached simple manner, overall when directly asked provide technical advances about sensors https://sentinel.esa.int/web/sentinel/events). knowledge gap still open possibility monitor (see first EBVs’ class described above) using products. Some studies demonstrated feasibility geographic studying variability Vernesi indicators effective method bridge field-based diversity. However, diversity appears too complex faced proxies, since it dependent many factors like demographic history, genetics, environmental management, gene flow, barriers, so forth (Bruford critical ensuring proper management ecosystems past recent Gathering meteorological worldwide high temporal resolution daily or hourly data) fine (i.e., microclimate) not affordable logistical cost-effective perspective. there promising attempts SoilTemp initiative (Lembrechts 2020), gather detailed micrometeorological scale. Hence, powerful tool providing important covariates interpolate scattered spatialize microclimatic satellite forecasting scenarios spatio-temporal & Lenoir, Maclean, As example, NASA Ozone Monitoring Instrument (OMI), used air quality Galapagos Islands, baseline levels contaminants most vulnerable hotspots (Cazorla Herrera, Additionally, long temperature generated deriving (LST) layers resolution, coarse 5 km MODIS, https://terra.nasa.gov/about/terra-instruments/modis, Figure 1) 4 (GOES network, https://www.nasa.gov/content/goes) 10 m (Sentinel, https://sentinel.esa.int/web/sentinel/missions) 30 (Landsat, https://landsat.gsfc.nasa.gov/) intermediate 1 (Envisat, https://earth.esa.int/web/guest/missions/esa-operational-eo-missions/envisat). Similarly, these derived resolutions, monthly basis Tomlinson 2011 review). Examples showed predicting trends climatic medium Metz 2014) very Lenoir 2017; Zellweger 2019) resolutions. Such rule out number follows: dispersal (Bellier 2012); biological invasions (Gallien 2010); loss (Bartel Sexton, 2009). This true considering facing generally less resistant additional detrimental effects (Travis, 2003). perspective, aerial photography dating back 50s help classifying landscape (Rocchini Ricotta, 2007) linking multi-temporal land-use changes estimated imagery (Gillespie 2008). Furthermore, products now able track above, repetition rate, up four per day few kilometres (Wan, 2008), moderate imaging spectroradiometer (MODIS) temperature/emissivity product. trend analysis available. Satellite has core part World Meteorological Organization (WMO, https://public.wmo.int/en) estimates cloud motion vectors, atmospheric snow ice starting 80s (Leese, 1987), under flag Weather Watch Programme (https://www.wmo.int/). combined complete intra- inter-annual weather find potentially impacts biodiversity. Copernicus program (https://www.copernicus.eu/) coordinated managed European Commission implemented Member States, Space Agency (ESA), Exploitation Satellites (EUMETSAT), Center Medium Forecast (ECMWF) scale properties index, fraction green vegetation cover, burnt areas) energy top reflectance, albedo) climatic-related LST, cryosphere, cover), including Proba-V MODIS sensors. available basis. LST product geostationary constellation (https://land.copernicus.eu/global/products/lst). Moreover, promoted launch Sentinel satellites, revisit time, days, (10 m, reasonably free sets until 2028 dramatically allows study local environment, all researchers globe their reproducible manner. those countries funds low but habitats preserved extremely whole planet African South-American tropical forests) threats deforestation fire spread should constantly monitored. airborne helps predict grain LiDAR laser scanning (Zellweger While seeing spectral peaks microclimate change, structural strictly dynamics derive model forest Frame Wood New Forest southern England (Latif Blackburn, applications technologies inform distribution models microrefugia persist locally anthropogenic successfully spots sudden turbulence (Hon Chan, Wildmann 2019), wind radial (Frehlich 1997), wake vortices characteristics (Holzapfel 2003), hub ocean winds (Hasager 2013). Modelling spot-events they longer alter functioning short period leading devastating threatening organism challenging resilience, single habitats. Remote cutting-edge technology support interdisciplinarity sciences. open-source packages software modules recently relate disciplines. Several freely R statistical (R Core Team, like, rasterdiv package (Marcantonio download situ (Czernecki ClimInd compute indices (Reig-Gracia hsdar lidR manage analyse respectively (Lehnert Roussel RStoolbox (Leutner Numerical prediction far being uncertainty, modelling. physics grounded equation (Abbe, 1901), steps physical ensemble known. On other mapping errors crucial get estimate bias beside exist implement maps error growing (Thompson, 1957) chaos theory (Lorenz, 2006). basic idea any forecast boundary (including, emission scenarios) sensitive propagation (Pijanowski 2011) non-stationarity (Foody, 2004; Gillespie Care must then taken modelling applied order avoid final due initialization forecasts. use cross-disciplinary its infancy frontier past, considered forecasts functions drivers, cycling flow. realized challenges before tools throughput standard. concern, issues, (resolution, sensu; Dungan 2002). would add noise instead information model, while will create smoothed surfaces, useless (Nagendra Rocchini, volumes computational needs once dealing observations, across wide extents (Li initiatives promote computing collaborative cyberinfrastructure development. projects ADAPT (https://www.nccs.nasa.gov/systems/adapt), Goddard Private Cloud (https://www.nas.nasa.gov/SC18/demos/demo34.html), SMCE (https://www.nccs.nasa.gov/systems/SMCE) virtualize, high-performance techniques, gathering leverage data-intensive calculus. Finally, reduction uncertainty parameterization another issue (Bauer 2015) empirical framework changes, guaranteeing geospatial rely (at potentially) unbiased data. Given above drawbacks challenges, efficient linked variables, enable practitioners work together. forthcoming “Climate Science Ecological Forecasting Symposium” interdisciplinary conference ecology, jointly hosted Royal Society British Society. Together, attendees explore opportunities greater interaction disciplines establish necessary better plan our More Society's website: https://www.rmets.org/event/climate-science-ecological-forecasting https://www.rmets.org/event/climate-science-ecological-forecasting. DR partially supported H2020 Project SHOWCASE (Grant agreement No 862480) COST Action CA17134 ‘Optical synergies spatiotemporal scalable ecophysiological (SENSECO)’. Duccio Rocchini: Conceptualization (equal); writing original draft review editing (equal). Jonathan Lenoir: SHOWCASE, Grant/Award Number: 862480; Action,