he Global ARMOR3D L4 Reprocessed dataset is obtained by combining satellite (Sea Level Anomalies, Geostrophic Surface Currents, Sea Surface Temperature) and in-situ (Temperature and Salinity profiles) observations through statistical methods. References : - ARMOR3D: Guinehut S., A.-L. Dhomps, G. Larnicol and P.-Y. Le Traon, 2012: High resolution 3D temperature and salinity fields derived from in situ and satellite observations. Ocean Sci., 8(5):845–857. - ARMOR3D: Guinehut S., P.-Y. Le Traon, G. Larnicol and S. Philipps, 2004: Combining Argo and remote-sensing data to estimate the ocean three-dimensional temperature fields - A first approach based on simulated observations. J. Mar. Sys., 46 (1-4), 85-98. - ARMOR3D: Mulet, S., M.-H. Rio, A. Mignot, S. Guinehut and R. Morrow, 2012: A new estimate of the global 3D geostrophic ocean circulation based on satellite data and in-situ measurements. Deep Sea Research Part II : Topical Studies in Oceanography, 77–80(0):70–81.

'''DEFINITION''' The Strong Wave Incidence index is proposed to quantify the variability of strong wave conditions in the Iberia-Biscay-Ireland regional seas. The anomaly of exceeding a threshold of Significant Wave Height is used to characterize the wave behavior. A sensitivity test of the threshold has been performed evaluating the differences using several ones (percentiles 75, 80, 85, 90, and 95). From this indicator, it has been chosen the 90th percentile as the most representative, coinciding with the state-of-the-art. Two Copernicus Marine products are used to compute the Strong Wave Incidence index: * IBI-WAV-MYP: '''IBI_MULTIYEAR_WAV_005_006''' * IBI-WAV-NRT: '''IBI_ANALYSISFORECAST_WAV_005_005''' The Strong Wave Incidence index (SWI) is defined as the difference between the climatic frequency of exceedance (Fclim) and the observational frequency of exceedance (Fobs) of the threshold defined by the 90th percentile (ThP90) of Significant Wave Height (SWH) computed on a monthly basis from hourly data of IBI-WAV-MYP product: SWI = Fobs(SWH > ThP90) – Fclim(SWH > ThP90) Since the Strong Wave Incidence index is defined as a difference of a climatic mean and an observed value, it can be considered an anomaly. Such index represents the percentage that the stormy conditions have occurred above/below the climatic average. Thus, positive/negative values indicate the percentage of hourly data that exceed the threshold above/below the climatic average, respectively. '''CONTEXT''' Ocean waves have a high relevance over the coastal ecosystems and human activities. Extreme wave events can entail severe impacts over human infrastructures and coastal dynamics. However, the incidence of severe (90th percentile) wave events also have valuable relevance affecting the development of human activities and coastal environments. The Strong Wave Incidence index based on the Copernicus Marine regional analysis and reanalysis product provides information on the frequency of severe wave events. The IBI-MFC covers the Europe’s Atlantic coast in a region bounded by the 26ºN and 56ºN parallels, and the 19ºW and 5ºE meridians. The western European coast is located at the end of the long fetch of the subpolar North Atlantic (Mørk et al., 2010), one of the world’s greatest wave generating regions (Folley, 2017). Several studies have analyzed changes of the ocean wave variability in the North Atlantic Ocean (Bacon and Carter, 1991; Kushnir et al., 1997; WASA Group, 1998; Bauer, 2001; Wang and Swail, 2004; Dupuis et al., 2006; Wolf and Woolf, 2006; Dodet et al., 2010; Young et al., 2011; Young and Ribal, 2019). The observed variability is composed of fluctuations ranging from the weather scale to the seasonal scale, together with long-term fluctuations on interannual to decadal scales associated with large-scale climate oscillations. Since the ocean surface state is mainly driven by wind stresses, part of this variability in Iberia-Biscay-Ireland region is connected to the North Atlantic Oscillation (NAO) index (Bacon and Carter, 1991; Hurrell, 1995; Bouws et al., 1996, Bauer, 2001; Woolf et al., 2002; Tsimplis et al., 2005; Gleeson et al., 2017). However, later studies have quantified the relationships between the wave climate and other atmospheric climate modes such as the East Atlantic pattern, the Arctic Oscillation pattern, the East Atlantic Western Russian pattern and the Scandinavian pattern (Izaguirre et al., 2011, Martínez-Asensio et al., 2016). The Strong Wave Incidence index provides information on incidence of stormy events in four monitoring regions in the IBI domain. The selected monitoring regions (Figure 1.A) are aimed to provide a summarized view of the diverse climatic conditions in the IBI regional domain: Wav1 region monitors the influence of stormy conditions in the West coast of Iberian Peninsula, Wav2 region is devoted to monitor the variability of stormy conditions in the Bay of Biscay, Wav3 region is focused in the northern half of IBI domain, this region is strongly affected by the storms transported by the subpolar front, and Wav4 is focused in the influence of marine storms in the North-East African Coast, the Gulf of Cadiz and Canary Islands. More details and a full scientific evaluation can be found in the CMEMS Ocean State report (Pascual et al., 2020). '''CMEMS KEY FINDINGS''' The trend analysis of the SWI index for the period 1980–2024 shows statistically significant trends (at the 99% confidence level) in wave incidence, with an increase of at least 0.05 percentage points per year in regions WAV1, WAV3, and WAV4. The analysis of the historical period, based on reanalysis data, highlights the major wave events recorded in each monitoring region. In region WAV1 (panel B), the maximum wave event occurred in February 2014, resulting in a 28% increase in strong wave conditions. In region WAV2 (panel C), two notable wave events were identified in November 2009 and February 2014, with increases of 16–18% in strong wave conditions. Similarly, in region WAV3 (panel D), a major event occurred in February 2014, marking one of the most intense events in the region with a 20% increase in storm wave conditions. Additionally, a comparable storm affected the region two months earlier, in December 2013. In region WAV4 (panel E), the most extreme event took place in January 1996, producing a 25% increase in strong wave conditions. Although each monitoring region is generally affected by independent wave events, the analysis reveals several historical events with above-average wave activity that propagated across multiple regions: November–December 2010 (WAV3 and WAV2), February 2014 (WAV1, WAV2, and WAV3), and February–March 2018 (WAV1 and WAV4). The analysis of the near-real-time (NRT) period (from January 2024 onward) identifies a significant event in February 2024 that impacted regions WAV1 and WAV4, resulting in increases of 20% and 15% in strong wave conditions, respectively. For region WAV4, this event represents the second most intense event recorded in the region. '''DOI (product):''' https://doi.org/10.48670/moi-00251

'''This product has been archived''' "''DEFINITION''' Marine primary production corresponds to the amount of inorganic carbon which is converted into organic matter during the photosynthesis, and which feeds upper trophic layers. The daily primary production is estimated from satellite observations with the Antoine and Morel algorithm (1996). This algorithm modelized the potential growth in function of the light and temperature conditions, and with the chlorophyll concentration as a biomass index. The monthly area average is computed from monthly primary production weighted by the pixels size. The trend is computed from the deseasonalised time series (1998-2022), following the Vantrepotte and Mélin (2009) method. The trend estimate is not shown because the length of the time series does not allow to completely differentiate the climate trend to the natural variability of the primary production. More details are provided in the Ocean State Reports 4 (Cossarini et al. ,2020). '''CONTEXT''' Marine primary production is at the basis of the marine food web and produce about 50% of the oxygen we breath every year (Behrenfeld et al., 2001). Study primary production is of paramount importance as ocean health and fisheries are directly linked to the primary production (Pauly and Christensen, 1995, Fee et al., 2019). Changes in primary production can have consequences on biogeochemical cycles, and specially on the carbon cycle, and impact the biological carbon pump intensity, and therefore climate (Chavez et al., 2011). Despite its importance for climate and socio-economics resources, primary production measurements are scarce and do not allow a deep investigation of the primary production evolution over decades. Satellites observations and modelling can fill this gap. However, depending of their parametrisation, models can predict an increase or a decrease in primary production by the end of the century (Laufkötter et al., 2015). Primary production from satellite observations presents therefore the advantage to dispose an archive of more than two decades of global data. This archive can be assimilated in models, in addition to direct environmental analysis, to minimise models uncertainties (Gregg and Rousseaux, 2019). In the Ocean State Reports 4, primary production estimate from satellite and from modelling are compared at the scale of the Mediterranean Sea. This demonstrates the ability of such a comparison to deeply investigate physical and biogeochemical processes associated to the primary production evolution (Cossarini et al., 2020) '''CMEMS KEY FINDINGS''' Global primary production does not show specific trend and remain relatively constant over the archive 1998-2022. The temporal variability of the primary production appears to be mainly driven by the seasonal variation. However, some specific inter-annual event may induce noticeable increase or decrease in primary production, as for example in the second part of 2011. '''DOI (product):''' https://doi.org/10.48670/moi-00225

'''DEFINITION''' The trend map is derived from version 5 of the global climate-quality chlorophyll time series produced by the ESA Ocean Colour Climate Change Initiative (ESA OC-CCI, Sathyendranath et al. 2019; Jackson 2020) and distributed by CMEMS. The trend detection method is based on the Census-I algorithm as described by Vantrepotte et al. (2009), where the time series is decomposed as a fixed seasonal cycle plus a linear trend component plus a residual component. The linear trend is expressed in % year -1, and its level of significance (p) calculated using a t-test. Only significant trends (p < 0.05) are included. '''CONTEXT''' Phytoplankton are key actors in the carbon cycle and, as such, recognised as an Essential Climate Variable (ECV). Chlorophyll concentration is the most widely used measure of the concentration of phytoplankton present in the ocean. Drivers for chlorophyll variability range from small-scale seasonal cycles to long-term climate oscillations and, most importantly, anthropogenic climate change. Due to such diverse factors, the detection of climate signals requires a long-term time series of consistent, well-calibrated, climate-quality data record. Furthermore, chlorophyll analysis also demands the use of robust statistical temporal decomposition techniques, in order to separate the long-term signal from the seasonal component of the time series. '''CMEMS KEY FINDINGS''' The average global trend for the 1997-2021 period was 0.51% per year, with a maximum value of 25% per year and a minimum value of -6.1% per year. Positive trends are pronounced in the high latitudes of both northern and southern hemispheres. The significant increases in chlorophyll reported in 2016-2017 (Sathyendranath et al., 2018b) for the Atlantic and Pacific oceans at high latitudes appear to be plateauing after the 2021 extension. The negative trends shown in equatorial waters in 2020 appear to be remain consistent in 2021. '''DOI (product):''' https://doi.org/10.48670/moi-00230

'''This product has been archived'''                For operationnal and online products, please visit https://marine.copernicus.eu '''DEFINITION''' The ibi_omi_tempsal_sst_area_averaged_anomalies product for 2021 includes Sea Surface Temperature (SST) anomalies, given as monthly mean time series starting on 1993 and averaged over the Iberia-Biscay-Irish Seas. The IBI SST OMI is built from the CMEMS Reprocessed European North West Shelf Iberai-Biscay-Irish Seas (SST_MED_SST_L4_REP_OBSERVATIONS_010_026, see e.g. the OMI QUID, http://marine.copernicus.eu/documents/QUID/CMEMS-OMI-QUID-ATL-SST.pdf), which provided the SSTs used to compute the evolution of SST anomalies over the European North West Shelf Seas. This reprocessed product consists of daily (nighttime) interpolated 0.05° grid resolution SST maps over the European North West Shelf Iberai-Biscay-Irish Seas built from the ESA Climate Change Initiative (CCI) (Merchant et al., 2019) and Copernicus Climate Change Service (C3S) initiatives. Anomalies are computed against the 1993-2014 reference period. '''CONTEXT''' Sea surface temperature (SST) is a key climate variable since it deeply contributes in regulating climate and its variability (Deser et al., 2010). SST is then essential to monitor and characterise the state of the global climate system (GCOS 2010). Long-term SST variability, from interannual to (multi-)decadal timescales, provides insight into the slow variations/changes in SST, i.e. the temperature trend (e.g., Pezzulli et al., 2005). In addition, on shorter timescales, SST anomalies become an essential indicator for extreme events, as e.g. marine heatwaves (Hobday et al., 2018). '''CMEMS KEY FINDINGS''' The overall trend in the SST anomalies in this region is 0.011 ±0.001 °C/year over the period 1993-2021. '''DOI (product):''' https://doi.org/10.48670/moi-00256

'''This product has been archived'''                For operationnal and online products, please visit https://marine.copernicus.eu '''DEFINITION''' The BALTIC_OMI_TEMPSAL_sst_trend product includes the cumulative/net trend in sea surface temperature anomalies for the Baltic Sea from 1993-2021. The cumulative trend is the rate of change (°C/year) scaled by the number of years (29 years). The SST Level 4 analysis products that provide the input to the trend calculations are taken from the reprocessed product SST_BAL_SST_L4_REP_OBSERVATIONS_010_016 with a recent update to include 2021. The product has a spatial resolution of 0.02 degrees in latitude and longitude. The OMI time series runs from Jan 1, 1993 to December 31, 2021 and is constructed by calculating monthly averages from the daily level 4 SST analysis fields of the SST_BAL_SST_L4_REP_OBSERVATIONS_010_016 from 1993 to 2021. See the Copernicus Marine Service Ocean State Reports for more information on the OMI product (section 1.1 in Von Schuckmann et al., 2016; section 3 in Von Schuckmann et al., 2018). The times series of monthly anomalies have been used to calculate the trend in SST using Sen’s method with confidence intervals from the Mann-Kendall test (section 3 in Von Schuckmann et al., 2018). '''CONTEXT''' SST is an essential climate variable that is an important input for initialising numerical weather prediction models and fundamental for understanding air-sea interactions and monitoring climate change. The Baltic Sea is a region that requires special attention regarding the use of satellite SST records and the assessment of climatic variability (Høyer and She 2007; Høyer and Karagali 2016). The Baltic Sea is a semi-enclosed basin with natural variability and it is influenced by large-scale atmospheric processes and by the vicinity of land. In addition, the Baltic Sea is one of the largest brackish seas in the world. When analysing regional-scale climate variability, all these effects have to be considered, which requires dedicated regional and validated SST products. Satellite observations have previously been used to analyse the climatic SST signals in the North Sea and Baltic Sea (BACC II Author Team 2015; Lehmann et al. 2011). Recently, Høyer and Karagali (2016) demonstrated that the Baltic Sea had warmed 1-2oC from 1982 to 2012 considering all months of the year and 3-5oC when only July- September months were considered. This was corroborated in the Ocean State Reports (section 1.1 in Von Schuckmann et al., 2016; section 3 in Von Schuckmann et al., 2018). '''CMEMS KEY FINDINGS''' SST trends were calculated for the Baltic Sea area and the whole region including the North Sea, over the period January 1993 to December 2021. The average trend for the Baltic Sea domain (east of 9°E longitude) is 0.049 °C/year, which represents an average warming of 1.42 °C for the 1993-2021 period considered here. When the North Sea domain is included, the trend decreases to 0.03°C/year corresponding to an average warming of 0.87°C for the 1993-2021 period. Trends are highest for the Baltic Sea region and North Atlantic, especially offshore from Norway, compared to other regions. '''DOI (product):''' https://doi.org/10.48670/moi-00206

'''DEFINITION''' The omi_climate_sst_ibi_trend product includes the Sea Surface Temperature (SST) trend for the Iberia-Biscay-Irish areas over the period 1982-2024, i.e. the rate of change (°C/year). This OMI is derived from the CMEMS REP ATL L4 SST product (SST_ATL_SST_L4_REP_OBSERVATIONS_010_026), see e.g. the OMI QUID, http://marine.copernicus.eu/documents/QUID/CMEMS-OMI-QUID-CLIMATE-SST-IBI_v3.pdf), which provided the SSTs used to compute the SST trend over the Iberia-Biscay-Irish areas. This reprocessed product consists of daily (nighttime) interpolated 0.05° grid resolution SST maps built from re-processed ESA SST CCI, C3S (Embury et al., 2024). Trend analysis has been performed by using the X-11 seasonal adjustment procedure (see e.g. Pezzulli et al., 2005), which has the effect of filtering the input SST time series acting as a low bandpass filter for interannual variations. Mann-Kendall test and Sens’s method (Sen 1968) were applied to assess whether there was a monotonic upward or downward trend and to estimate the slope of the trend and its 95% confidence interval. The reference for this OMI can be found in the first and second issue of the Copernicus Marine Service Ocean State Report (OSR), Section 1.1 (Roquet et al., 2016; Mulet et al., 2018). '''CONTEXT''' Sea surface temperature (SST) is a key climate variable since it deeply contributes in regulating climate and its variability (Deser et al., 2010). SST is then essential to monitor and characterise the state of the global climate system (GCOS 2010). Long-term SST variability, from interannual to (multi-)decadal timescales, provides insight into the slow variations/changes in SST, i.e. the temperature trend (e.g., Pezzulli et al., 2005). In addition, on shorter timescales, SST anomalies become an essential indicator for extreme events, as e.g. marine heatwaves (Hobday et al., 2018). '''CMEMS KEY FINDINGS''' The overall trend in the SST anomalies in this region is 0.012 ±0.001 °C/year over the period 1982-2024. '''DOI (product):''' https://doi.org/10.48670/moi-00257

'''DEFINITION''' The time series are derived from the regional chlorophyll reprocessed (MY) product as distributed by CMEMS (OCEANCOLOUR_MED_BGC_L3_NRT_009_141). This dataset, derived from multi-sensor (SeaStar-SeaWiFS, AQUA-MODIS, NOAA20-VIIRS, NPP-VIIRS, Envisat-MERIS and Sentinel3-OLCI) Rrs spectra produced by CNR using an in-house processing chain, is obtained by means of the Mediterranean Ocean Colour regional algorithms: an updated version of the MedOC4 (Case 1 (off-shore) waters, Volpe et al., 2019, with new coefficients) and AD4 (Case 2 (coastal) waters, Berthon and Zibordi, 2004). The processing chain and the techniques used for algorithms merging are detailed in Colella et al. (2023). Monthly regional mean values are calculated by performing the average of 2D monthly mean (weighted by pixel area) over the region of interest. The deseasonalized time series is obtained by applying the X-11 seasonal adjustment methodology on the original time series as described in Colella et al. (2016), and then the Mann-Kendall test (Mann, 1945; Kendall, 1975) and Sens’s method (Sen, 1968) are subsequently applied to obtain the magnitude of trend. This OMI has been introduced since the 2nd issue of Ocean State Report in 2017. '''CONTEXT''' Phytoplankton and chlorophyll concentration as a proxy for phytoplankton respond rapidly to changes in environmental conditions, such as light, temperature, nutrients and mixing (Colella et al. 2016). The character of the response depends on the nature of the change drivers, and ranges from seasonal cycles to decadal oscillations (Basterretxea et al. 2018). Therefore, it is of critical importance to monitor chlorophyll concentration at multiple temporal and spatial scales, in order to be able to separate potential long-term climate signals from natural variability in the short term. In particular, phytoplankton in the Mediterranean Sea is known to respond to climate variability associated with the North Atlantic Oscillation (NAO) and El Niño Southern Oscillation (ENSO) (Basterretxea et al. 2018, Colella et al. 2016). '''KEY FINDINGS''' In the Mediterranean Sea, the average chlorophyll trend for the 1997–2024 period is slightly negative, at -0.77 ± 0.59% per year, reinforcing the findings of the previous releases. This result contrasts with the analysis by Sathyendranath et al. (2018), which reported increasing chlorophyll concentrations across all European seas. From around 2010–2011 onward, excluding the 2018–2019 period, a noticeable decline in chlorophyll levels is evident in the deseasonalized time series (green line) and in the observed maxima (grey line), particularly from 2015. This sustained decline over the past decade contributes to the overall negative trend observed in the Mediterranean Sea. '''DOI (product):''' https://doi.org/10.48670/moi-00259

'''This product has been archived''' For operationnal and online products, please visit https://marine.copernicus.eu '''Short description:''' The Global Ocean Satellite monitoring and marine ecosystem study group (GOS) of the Italian National Research Council (CNR) in Rome distributes reprocessed surface chlorophyll concentration (Chl) and phytoplankton functional types (PFT). Input Rrs multi-sensor (MODIS-AQUA, NOAA20-VIIRS, NPP-VIIRS, Sentinel3A-OLCI) spectra at the state-of-the-art algorithms for multi-sensor merging. Single sensor Rrs fields are band-shifted, over the SeaWiFS native bands (using the QAAv6 model, Lee et al., 2002) and merged. Reprocessed (multi-year) products are consistent and homogeneous in terms of format, algorithms and processing software. Chl is obtained by means of the Mediterranean regional algorithms: an updated version of the MedOC4 (Volpe et al., 2019) and AD4 (Berthon and Zibordi, 2004). Discrimination between the two water types is performed by comparing the satellite spectrum with the average spectrum from in situ measurements. Reference insitu dataset is MedBiOp (Volpe et al., 2019) where Case II spectra are selected with a k-mean cluster analysis (Melin et al., 2015). Merging of Case I and Case II information is performed estimating the Mahalanobis distance between observed and reference spectra and using it as weight for the final value. The PFT provides estimates of Chl concentration of 9 phytoplankton groups: Micro, Nano, Pico, Diato, Dino, Crypto, Hapto, Green and Prokar. Micro consists of Diato and Dino, Nano includes Crypto and Hapto and Pico is referred to Green and Prokar with the adjustment of Brewin et al. (2010) in the ultra-oligotrophic water for Pico and Nano. These classes are estimated via empirical regional functions, correlating Chl concentration with each in-situ PFT fraction computed by a regional diagnostic pigment analysis (Di Cicco et al. 2017). '''Processing information:''' Multi-sensor product is constituted by MODIS-AQUA, NOAA20-VIIRS, NPP-VIIRS and Sentinel3A-OLCI. For consistency with NASA L2 dataset, BRDF correction was applied to Sentinel3A-OLCI prior to band shifting and multi sensor merging. Single sensor NASA Level-2 data are destriped and then all Level-2 data are remapped at 1 km spatial resolution using cylindrical equirectangular projection. Afterwards, single sensor Rrs fields are band-shifted, over the SeaWiFS native bands (using the QAAv6 model, Lee et al., 2002) and merged with a technique aimed at smoothing the differences among different sensors. This technique is developed by The Global Ocean Satellite monitoring and marine ecosystem study group (GOS) of the Italian National Research Council (CNR, Rome). Then geophysical fields (i.e. chlorophyll and kd490) are estimated via state-of-the-art algorithms for better product quality. The entire data set is consistent and processed in one-shot mode (with an unique software version and identical configurations). '''Description of observation methods/instruments:''' Ocean colour technique exploits the emerging electromagnetic radiation from the sea surface in different wavelengths. The spectral variability of this signal defines the so-called ocean colour which is affected by the presence of phytoplankton. '''Quality / Accuracy / Calibration information:''' A detailed description of the calibration and validation activities performed over this product can be found on the CMEMS web portal. '''Suitability, Expected type of users / uses:''' This product is meant for use for educational purposes and for the managing of the marine safety, marine resources, marine and coastal environment and for climate and seasonal studies. '''Dataset names:''' * dataset-oc-med-chl-multi-l3-chl_1km_daily-rep-v02 * dataset-oc-med-pft-multi-l3-pft_1km_daily-rep-v02 '''Files format:''' *CF-1.4 *INSPIRE compliant '''DOI (product) :''' https://doi.org/10.48670/moi-00112

'''This product has been archived''' '''DEFINITION''' The temporal evolution of thermosteric sea level in an ocean layer is obtained from an integration of temperature driven ocean density variations, which are subtracted from a reference climatology to obtain the fluctuations from an average field. The regional thermosteric sea level values are then averaged from 60°S-60°N aiming to monitor interannual to long term global sea level variations caused by temperature driven ocean volume changes through thermal expansion as expressed in meters (m). '''CONTEXT''' Most of the interannual variability and trends in regional sea level is caused by changes in steric sea level. At mid and low latitudes, the steric sea level signal is essentially due to temperature changes, i.e. the thermosteric effect (Stammer et al., 2013, Meyssignac et al., 2016). Salinity changes play only a local role. Regional trends of thermosteric sea level can be significantly larger compared to their globally averaged versions (Storto et al., 2018). Except for shallow shelf sea and high latitudes (> 60° latitude), regional thermosteric sea level variations are mostly related to ocean circulation changes, in particular in the tropics where the sea level variations and trends are the most intense over the last two decades. '''CMEMS KEY FINDINGS''' Significant (i.e. when the signal exceeds the noise) regional trends for the period 2005-2019 from the Copernicus Marine Service multi-ensemble approach show a thermosteric sea level rise at rates ranging from the global mean average up to more than 8 mm/year. There are specific regions where a negative trend is observed above noise at rates up to about -8 mm/year such as in the subpolar North Atlantic, or the western tropical Pacific. These areas are characterized by strong year-to-year variability (Dubois et al., 2018; Capotondi et al., 2020). Note: The key findings will be updated annually in November, in line with OMI evolutions. '''DOI (product):''' https://doi.org/10.48670/moi-00241