Environmental change in coastal and shelf seas provides a challenge to fisheries and aquaculture worldwide. Coupled physical-biogeochemical models can be used to create projections of environmental conditions under different climate scenarios, but models of this type produce a wealth of information covering multiple environmental variables and a range of timescales. It is challenging to synthesize that information into a form which can be quickly and effectively conveyed to stakeholders. I describe how a range of approaches has been used to analyse projections of environmental conditions in European seas, created using a regional model with a resolution of 0.1 degrees. The projections run from the present-day to 2099 and include temperature, primary production, plankton biomass, pH and nutrient and oxygen concentrations for moderate and high carbon concentration scenarios. A range of formats that convey the scale and extent of projected change in the North East Atlantic, the North Sea and the Mediterranean, are included, aiming for simple, visual communication that goes beyond maps of changing sea surface temperature. Colour-coded tables show at a glance which variables are projected to show significant change at a range of locations. Time series and maps are used together to give complementary information. Contours show geographical shifts in conditions. In each case the content can be adjusted to fit the interests of different stakeholder groups: examples from fisheries and aquaculture are included here.

The onset of plant flowering is one of many factors affected by climate change. This has important consequences for ecosystems, where plants and pollinators are at risk of a temporal mismatch. This also has implications for crops. The timing of their highest-sensitivity window is affected, despite some control exerted by agricultural management. Current efforts to monitor the onset of flowering rest on numerous but sparse ground observations, or on proxies like temperature models or vegetation indices. In crop studies, a time-invariant harvesting calendar is most widely used to infer the flowering period for crops. We use spectral data from satellite imagery, combined with convolutional neural networks, to create a large-scale, high-resolution proxy for the timing of flowering. We exploit a transfer learning method to overcome the sparsity of ground truth data on plant phenology. We first train a temporal convolutional neural network to predict a temperature-based proxy for flowering (the First Bloom Index from the National Phenology Network), using as input the 8-day composite from MODIS at a 500m resolution. This model learns relevant temporal patterns on both short and long time scales within a year, thanks to a dilated convolution structure. We then fine-tune the parameters from the last fully connected layer using ground observations from the National Phenology Network as training data. The root mean squared error of our model is about 8.6 days, though this average hides much spatial heterogeneity: in many places, the model performs well-enough to capture year-to-year variations, with a prediction within a couple days of the target, while it performs extremely poorly in a few places. Our model outperforms the temperature-based proxy since the latter has a root mean squared error of about 9.3 days on the same testing data. More importantly, our model extrapolates much better over space: it has a root mean squared error of about 12 days when predicting outside of our study region, compared to a 35-day error for the temperature-based model. This is a key feature, since our goal is to fill in the spatial gaps left by the sparse ground truth observations. These large-scale predictions of annual flowering times could offer insights into plant-pollinator desynchronization and crop calendar variation under climate change, and could be used for further applications like studying the health impacts of shifts in spring flowering times.

The rapid changes of magnetic fields associated with nighttime magnetic perturbations with amplitudes |ΔB| of hundreds of nT and 5-10 min periods can induce bursts of geomagnetically-induced currents that can harm technological systems. Recent studies of these events in eastern Arctic Canada, based on data from four ground magnetometer arrays and augmented by observations from auroral imagers and high-altitude spacecraft in the nightside magnetosphere, showed them to be highly localized, with largest |dB/dt| values within a ~275 km half-maximum radius that was associated with a region of shear between upward and downward field-aligned currents, and usually but not always associated with substorms. In this study we look in more detail at the field-aligned currents associated with these events using AMPERE data, and compare the context and characteristics of events not associated with substorms (occurring from 60 min to over two days after the most recent substorm onset) to those occurring within 30 min of onset. Preliminary results of this comparison, based on events with |dB/dt|≥ 6 nT/s observed during 2015 and 2017 at Repulse Bay (75.2° CGMLAT), showed that the SYM/H distributions for both categories of events were similar, with 85% between -40 and 10 nT, and the SME values during non-substorm events coincided with the lower half of the range of SME values for events during substorms (200 – 700 nT). Dipolarizations of ≥ 20 nT amplitude at GOES 13 occurred within 45 minutes prior to 73% of the substorm events but only 29% of the non-substorm events. These observations suggest that predictions of GICs cannot focus solely on the occurrence of intense substorms.

We discuss an efficient implementation of the analog ensemble algorithm, the distilled analog ensemble, which is achieved by distilling the post-processing transformation generated by the analog ensemble into a deep neural network. While the analog ensemble has been shown to be able to improve deterministic forecasts and create calibrated probabilistic predictions in many contexts, a common issue with operationalizing a large, global analog ensemble-based system is the amount of data (a corpus of historical forecasts) and post-processing latency required to process that data in the time-critical path of producing a forecast. Deep neural networks are high capacity function approximators, and we demonstrate that we are able to train a network that memorizes the post-processing behavior of the analog ensemble on a particular corpus of forecasts. This technique breaks the scale factor between the size of the historical forecast corpus (larger is better for forecast skill improvements) and the calculation required to post-process the current forecast in real-time operations. We show that the distilled analog ensemble is able to improve European Centre for Medium-Range Weather Forecasts (ECMWF) high-resolution deterministic forecasts of winds in the lower stratosphere using as ground-truth either the ECMWF analysis or observations from Loon high altitude balloons. In this case, rather than requiring terabytes of historical forecast data to apply the conventional analog ensemble, we can perform the post-processing that improves forecast quality on the fly doing computationally efficient forward passes through a pre-trained network that has a data size of only 100’s of kilobytes.

Since 2013 Susquehanna Shale Hills Critical Zone Observatory (SSHCZO) has been monitoring using a Ground Hydrological Observation System (GroundHOG) design consisting of various sensor types in the Shavers Creek watershed in central Pennsylvania. The GroundHOG design was established to study interactions between hydrological systems (surface and groundwater), soils, and ecosystems along catenas. We currently have three GroundHOG sites with differing land uses and geology: one located in a pristine shale watershed, another in a pristine sandstone watershed, and the third in an agricultural setting with mixed lithology. Each catena has three pits set up to compare hill slope position and one additional pit to compare north versus south aspect. Each pit is equipped with both automated and manual sensors that measure soil moisture and soil gas at varying depths. The GroundHOG deployment is accompanied by precipitation gauges, surface water monitoring gauges, and groundwater monitoring wells at all sites. In 2019 we built on the shale site to include weekly electrical resistivity measurements and seismometry near the GroundHOG. Specifically, we designed and installed an innovative chronoamperometric system that can respond to real-time redox reactions that change in response to changes in soil moisture, temperature, and saturation. These experimental sensors are co-located with the GroundHOG sites where soil gas measurements take place and can be coupled to understand the connection of hydrological processes to microbial communities. This presentation will emphasize the design of the GroundHOG and how the new measurements are made and compared in the shale GroundHOG site.

Operational ocean and wave models are used to produce forecasts for navigation but also required for a wide range of services, most of them relying on particle tracking methods. Improving the forecast capabilities and accuracy of a model is then a constant necessity in order to deliver more reliable services. The Marine Institute in Ireland is running several coastal operational models, one of them focuses on the area of Galway Bay in the west coast. At the moment it consists of a stand-alone ocean application, it is the purpose of the work presented here to set-up a coupled application with a wave model. Coupled models are a recent development in ocean modelling, developer teams have included ocean and wave coupling by combining existing models each dedicated to a specific physics. Two theoretical formulations are mostly used for the implementation, both giving the same equations of evolution and interaction terms known as the vortex-force formalism. One approach is using a Lagrangian framework defining an exact averaged operator following the fluid particles, the other approach is Eulerian making use of a multi-scale expansion. In both cases the larger current components are found to be forced by gravity and infra-gravity waves. The Coupled Ocean Atmosphere Wave Sediment Transport (COAWST) modelling system is a widely used code, the vortex-force formalism has been implemented in 2012 coupling the Regional Ocean Modeling System (ROMS) with the Simulation Wave Nearshore wave model (SWAN). The implementation has been validated with academic cases and used in several real case studies in the last decade. The work presented here is making use of COAWST, a coupled model is set-up for Galway Bay running a 1-year hind-cast application for 2017 and preliminary results are shown here. The performance of the coupled model is compared with each stand-alone model, using in-situ data as a reference. In the last releases of COAWST the wave model WAVEWATCHIII has been added and can be used in the coupled system. This new feature is tested and the results are compared against SWAN, both wave codes are solving the same equations but different technical choices have been made resulting in different capabilities.

Solar, auroral, and radiation belt electrons enter the atmosphere at polar regions leading to ionization and affecting its chemistry. For example particle-produced OH and NO molecules affect the ozone content in the middle atmosphere. Climate models usually parametrize this ionization and the related changes in chemistry based on satellite particle measurements. Precise measurements of the particle and energy influx into the upper atmosphere are difficult because they vary substantially in location and time. Widely used particle data are derived from the POES and GOES satellite measurements which provide electron and proton spectra. We present electron energy and flux measurements from the Special Sensor Ultraviolet Spectrographic Imagers (SSUSI) satellite instruments. This formation of satellites observes the auroral zone in the UV from which electron energies and fluxes are inferred. We use these observed electron energies and fluxes to calculate ionization rates and electron densities in the mesosphere and lower thermosphere (≈ 40–200 km). We also present an initial comparison of these rates to other models and compare the electron densities to those measured by the EISCAT radar. Together with photochemical models, trace gas concentrations, for example of NO, can be calculated from these electron densities. These concentrations then provide an independent source for comparing and validating satellite trace gas measurements.

Recent studies have demonstrated a direct relation between climate characteristics and vegetation in catchments. For example, plants appear to develop a root system that allows both optimal growth and resistance against region-specific droughts (Gao et al., 2014; Ho et al., 2005). As climatic conditions also affect the way catchments store and release water (i.e., the transit times), we expect a direct relation between vegetation and transit times. To test this hypothesis, we have established a dataset of water balance and stable water isotope data across more than 50 catchments in various climate zones, which will be further expanded over the course of the project. This dataset allows for determining root zone storage capacities and transit time metrics such as the young water fraction (Kirchner, 2016) across catchment scales and climate zones. We will present how transit time metrics vary as a function of root zone storage capacities and how this can be related to catchment and vegetation characteristics and climatic conditions. The results will help understand how changing vegetation cover due to climate and land use change might affect catchment water storage and release in future. We see a vast potential of isotope studies across diverse catchments. We are thus calling for a community effort to provide streamflow isotope data from previous work in a unified framework as a basis for further global analyses using stable water isotopes. Gao, H.; Hrachowitz, M.; Schymanski, S. J.; Fenicia, F.; Sriwongsitanon, N.; Savenije, H. H. G., Climate controls how ecosystems size the root zone storage capacity at catchment scale. Geophysical Research Letters 2014, 41, 7916–7923, doi:10.1002/2014GL061668. Ho, M. D.; Rosas, J. C.; Brown, K. M.; Lynch, J. P., Root architectural tradeoffs for water and phosphorus acquisition. Functional Plant Biology 2005, 32, (8), 737-748. Kirchner, J. W., Aggregation in environmental systems – Part 1: Seasonal tracer cycles quantify young water fractions, but not mean transit times, in spatially heterogeneous catchments. Hydrology and Earth System Sciences 2016, 20, (1), 279-297.

PM10 has natural and anthropogenic sources, it is an urban air pollutant from desertic areas or emitted by industry and traffic activities, it reduces visibility and threatens human wellbeing mainly in big cities. Casablanca concentrates many industrial units and a large vehicle fleet. The rate of urbanization in the metropolis and the population density are the highest in Morocco. Marrakech is one of the most populated cities in the country where the motorization rate has increased during recent years. The present work is based on PM10 daily measurements between 2013 and 2016. The main objective is to assess the concentrations of PM10 in Casablanca and Marrakech and study their relationship with the atmospheric circulation. First, we assessed PM10 correlations with climate indexes (NAO and MO), then we characterized the contribution of large-scale atmospheric patterns related to PM10 extreme events. The novelty of this research is the creation of a new climate index to characterize the oscillation, in the country’s southern desert, between the Saharan depression and the Azores high. The time series of the new Saharan Oscillation Index (SaOI) were calculated. This study has demonstrated the relationship between MO and PM10 averages and has shown that particulate pollution in the study area is partly induced by continental northeasterly to southwesterly flow. This flow is triggered by the Saharan trough and managed by the high-pressure area in the north. The assessed correlations related to the SaOI confirm the relationship between this index, PM10 averages, and MO and NAO indexes mainly in winter. The Saharan Oscillation is the new relevant key to understand worldwide pollution by fine particles.

Internal climate variability plays an important role in the abundance and distribution of phytoplankton in the global ocean. Previous studies using large ensembles of Earth system models (ESMs) have demonstrated their utility in the study of marine phytoplankton variability. These ESM large ensembles simulate the evolution of multiple alternate realities, each with a different phasing of internal climate variability. However, ESMs may not accurately represent real world variability as recorded via satellite and in situ observations of ocean chlorophyll over the past few decades. Observational records of surface ocean chlorophyll equate to a single ensemble member in the large ensemble framework, and this can cloud the interpretation of long-term trends: are they externally forced, caused by the phasing of internal variability, or both? Here, we use a novel statistical emulation technique to place the observational record of surface ocean chlorophyll into the large ensemble framework. Much like a large initial condition ensemble generated with an ESM, the resulting synthetic ensemble represents multiple possible evolutions of ocean chlorophyll concentration, each with a different phasing of internal climate variability. We further demonstrate the validity of our statistical approach by recreating a ESM ensemble of chlorophyll using only a single ESM ensemble member. We use the synthetic ensemble to explore the interpretation of long-term trends in the presence of internal variability. Our results suggest the potential to explore this approach for other ocean biogeochemical variables.

The Global Precipitation Measurement core satellite (GPM) has been collecting high quality precipitation data since 2014 over the globe with its Dual-frequency Precipitation Radar (DPR; Ku-band and Ka-band). Specificly over Germany, GPM provides data with typically two daily overpasses. Thus providing a unique opportunity to have a satellite based standard for estimation of precipitation in order to compare and evaluate ground-based radar network counterpart products. The German national weather service (DWD, Deutscher Wetterdienst) provides precipitation observations from its operational radar network RADOLAN as a composite products derived from 17 dual-pol C-band radars. The RADOLAN (RY) regular products are Germany-wide composites of precipitation estimates based on a set of precipitation type dependent Z-R relationships derived for liquid hydrometeors applied to radar reflectivity after clutter- and beam blockage-corrections. In this contribution we focus to compare three years of GPM DPR and RADOLAN precipitation products. This allows us to evaluate at which extend these two Near Surface products are consistent when observed from different geometries and obtained by independent instruements and retrieval methods. We quantify the uncertainties when directly comparing the DPR near surface product with RY. It is shown that a direct comparisons might not take into account a set of uncertainties originated from the scans geometry from DPR and RADOLAN, precipitation types, and sampling volumes. Therefore we suggest an adjusted DPR product, which is extracted from the DPR vertical profiles and adapted to fit the specific RY measurement configuration e.g. scans height and beam width. This allows a much more detailed classification of the hydrometoer phases per measuring volume, which we define as non-uniform phase beam filling (NPBF). The NPBF gives information about the ratio of liquid, solid or mixed hydrometeors in a given volume. Orographic, synoptic, microphysical influences as well as NPBF effects are examined and their uncertainties introduced on a direct comparison of satellite with ground-based producs are put into consideration. The adaptation of the DPR precipitation products to the specific scan geometry of the individual ground radars improves the correlation and reduce the RMSE.

The Heliophysics Application Programmer’s Interface (HAPI) represents a grass-roots effort to develop a common interface that data providers can use to offer time series data for computer-to-computer data access. Multiple data centers with Heliophysics and Planetary data are adopting this emerging standard. Several scientific analysis packages (Autoplot, SPEDAS) seamlessly pull from HAPI servers, and we have created and are enhancing Python libraries that can be used to read HAPI data into useful data structures within personalized analysis code. The HAPI specification and supporting software are available on GitHub. We will discuss the latest developments of the HAPI specification, current server and client implementations, all in the context of the push for increased interoperability within Heliophysics and Planetary data analysis. https://hapi-server.github.io/

Barrovian-grade pelites in the Greater Himalayan Crystallines and Lesser Himalayan Formations exposed in the Himalayan core are separated by the Main Central Thrust (MCT). This fault system accommodated a significant amount of India-Asia convergence and is the focus of several models that explore ideas about the development of the range and collisional belts in general. Units separated by the MCT provide critical information regarding the mechanisms of heat transfer within collisional belts. Garnets collected across the MCT record their growth history through changes in chemistry. These chemical changes can be extracted and modeled using a variety of thermodynamic approaches. Here we describe and apply particular thermobarometric techniques to decipher the metamorphic history of several garnet-bearing rocks collected across the MCT in central Nepal, the Sikkim region, and NW India. Comparisons are made between the results of previously-reported conventional rim P-T conditions and P-T paths extracted using the Gibb’s method to isopleth thermobarometry and high-resolution P-T path modeling using the same data and assemblages. Regardless of calibrations used, the P-T conditions and paths, along with previously-reported timing constraints, are consistent with an imbrication model that suggest the MCT shear zone developed as rock packages within the LHF were progressively transferred. In this model, samples within the LHF travel along the MCT at a 5 km/Ma speed rate from 25 to 18 Ma. The hanging wall speed rate is 10 km/Ma, and topography progressively accumulates until a maximum height of 3.5 km. Once the topography is achieved at 18 Ma, a period of cessation is applied to the MCT between 18 and 15 Ma, and topography is reduced at a rate of 1.5 km/Ma. The model returns to activity within the MCT shear zone with the activation of the MCT footwall slivers from 8 to 2 Ma. P‐T changes recorded by the footwall garnets result from thermal advection combined with alterations in topography. For most MCT footwall samples, the P-T paths match the model predictions remarkably well. The P-T paths for some samples in central Nepal are also consistent high exhumation rates (>12mm/year) within the MCT shear zone since the Pliocene, a scenario predicted by this imbrication model.

The broadband stacks (templates) of velocity seismograms of nearly co-located low-frequency earthquakes (LFEs) detected using a 1-8 Hz passband beneath southern Vancouver Island tend to exhibit a simple dipolar shape with a characteristic duration of ~0.3-0.5 s, which is also found to be nearly independent of the seismic moment. An important question left unanswered is whether the duration is due to the nature of the source, is set by attenuation near the source region, or is just a bias introduced by the narrow passband used to detect LFEs. In tremor catalogs detected using a relatively low-frequency passband, 0.5-1.25 Hz, we have found some tremor windows that contain relatively isolated dipole arrivals similar to LFEs. A few of these have a duration apparently longer than that of the LFE templates. Notably, the same location on the fault also seems capable of generating signals with a shorter duration at other times. Figure 1 shows seismograms, at 3 stations, of one such example in the vicinity of LFE family 001 of Bostock et al. (2012), in which the main arrival has a duration of ~1 s, whereas another signal 3 s earlier with a duration of only ~0.4 s comes from roughly the same location (same move-out between the stations). This significant variability in duration at approximately the same location suggests that the long-duration events owe their duration to source processes and not attenuation, provided that attenuation does not vary on extremely short time and space scales during the episodic tremor and slip episode. The relative isolation in time also makes the longer duration less likely to result from the temporal clustering of multiple typical LFEs. We will undertake a more systematic search of our longer- and shorter-period tremor catalogs to assess this possibility. Addressing this question will shed more light on the factors that control the apparent duration of LFEs. Figure 1 The top panel shows the long-duration tremor signal in a relatively lower-frequency band, 0.5-1.25 Hz, whereas the second panel from the top is the same 32-s segment in a higher-frequency band, 1.25-6.5 Hz. The third panel shows the trace in a broader passband, 0.5-6.5 Hz. The bottom panel shows the stacked LFE templates of the same family filtered through 0.5-6.5 Hz.

Himalaya is one of the most seismically active and earthquake-prone regions of the world. Recently an Earthquake Early Warning System (EEWS) for Northern India has been developed which includes around 165 strong ground motion accelerometers installed in Uttarakhand (India). A timely earthquake early warning of few seconds can save people’s lives and thus the development of an improved Warning Dissemination System (WDS) component of EEWS is much required. For this purpose, an improved WDS has been developed. This system currently uses newly designed low-cost public sirens and smartphone apps to alert people in case of an earthquake. The developed sirens have been currently installed in two major cities of Uttarakhand (India) and in all district emergency operations centers for the last 2 years. On 4th August 2021, the newly developed app “Uttarakhand Bhookamp Alert” for WDS has been officially launched by the Uttarakhand State Disaster Management Authority for operational use. Further, a GIS-based maintenance system is also developed for the maintenance of the complete system, which automatically checks the health and keeps records of the sirens along with the smartphone users. This app gives the facility to the users for sharing their current location with disaster rescue teams after the earthquake early warning alert. Detailed scientific principles and working mechanism of newly developed dissemination system along with its applicability to disseminate the real-time information in case of an earthquake will be presented.

Significant investments are made in the collection of water quality data. Yet barriers to effective data sharing and reuse hamper the ability to leverage this information to its full potential in research and water management decisions. Because water monitoring data are collected by a wide range of organizations, through programs of varying scope and focus, and often within jurisdictional or institutional silos, it can be difficult to connect this information together in standardized and accessible formats. DataStream is an online open-access platform that was developed by The Gordon Foundation and its partners to address the challenge of water data accessibility in Canada. DataStream is free to use and allows users to query, visualize, and download water quality data aligned with widely-adopted data and metadata standards (e.g., Water Quality eXchange, ISO19115 and schema.org). The path towards DataStream evolving as a collaborative and open data platform has been guided by the FAIR and CARE data principles. To date, over 140 different groups across Canada are using DataStream to publish water monitoring results including watershed groups, Indigenous organizations, researchers and governments at all levels. We will highlight our lessons learned in developing the platform to align with FAIR data principles and the elements we believe have been key to our success including DataStream’s open data schema, clear data licensing and regional partnership model.

The well-investigated aquifer in southern Tunisia have been selected from the literature to test the potential use of uranium isotopic compositions as a groundwater dating method. This is the Senonian carbonate aquifer of the Nefzaoua basin (https://doi.org/10.1016/j.quaint.2020.01.024). For the this aquifer, an increase in U concentrations along the generalized flow path is observed in proportion to an increase in total dissolved solids, which may indicate the predominance of dissolution processes over alpha-recoil processes under oxidizing conditions for uranium. There is also an increase in U concentrations with a decrease in 14C values. For this aquifer, positive results were obtained on uranium-isotope dating. It was found that the groundwater residence time in the aquifer increases from 440 to 11,300 years from the recharge area along the generalized flow path (Figure). Uranium age correction model under oxidizing conditions in the aquifer is given in “Malov, 2018. Evolution of the groundwater chemistry in the coastal aquifers of the south-eastern White Sea area (NW Russia) using 14C and 234U-238U dating. Science of the Total Environment. https://doi.org/10.1016/j.scitotenv.2017.10.197” It is shown that for uranium-isotope dating of groundwater, three main conditions must be met: i) oxidizing conditions for uranium in the aquifer, ii) an increase in uranium concentrations with a decrease in 14C activities, and iii) the homogeneity of the aquifer in terms of hydraulic conductivity and lithological composition. The author understands the reality of the fact that groundwater dating methods are under development and improvement. Nevertheless, analysis of the evolution of the chemical and isotopic composition of groundwater, analysis of the geological and hydrogeological history of the region, hydraulic estimates of groundwater velocities in aquifers, hydrodynamic and balance justifications for the formation of groundwater, and analysis of the composition of stable isotopes can increase the reliability of dating. In the future, it is necessary to continue studies to assess the retardation factor and recoil loss factor in order to improve the uranium-isotope method for dating groundwater under oxidizing conditions for uranium. This work was supported by the RFBR (project no. 20-05-00045_A)

Confined meandering rivers are bounded between valleys and they are limited in their lateral migration compared to free meandering ones. In this study, we designed a model-based analysis to investigate types of confined meandering rivers. To simulate cases, we used the model developed by (Bogoni et al. 2017) based on a semi-analytical solution for flow and bed morphology ((Zolezzi and Seminara 2001). According to the result of (Lewin and Brindle 1977) the categorization of this type of meandering river should be based on the potential of evolution. We defined a confinement ratio as the ratio between the confined valley width and the meander belt width that the river would theoretically reach without geological confinement. Such unconfined meander belt width has been computed through long-term morphodynamic model simulations. This differs from previous work (Camporeale et al. 2005), semi-empirically suggesting an unconfined meander belt being nearly 3 times the modeled, linearly most unstable, intrinsic meander length. Our results indicate the existence of four different types of confined meandering rivers, which represents the most important point of this work. Compared to previous work, our classification adds a new class and reclassifies one of the previous types. Our proposed classes are; “Moderately confined condition” for confinement ratio in the range 0.1-1, “Strongly confined condition” for confinement ratio in the range 0.03-0.1. For a smaller confinement ratio than 0.03 we obtain a low sinuosity, single thread river, and for a confinement ratio greater than 1, we retrieve a free meandering river. Preliminary testing of our model-based classification with data from real confined meandering rivers shows promising results.