Dr. Riddick Kakati Researcher and full stack developer

River bank erosion due to unstable banks and high flow variability is usually controlled using permeable and impermeable structures, which are not studied much yet and also cannot solely provide desired velocity reduction. These structures, along with a combination of porcupine screens followed by geobag (i.e., Hybrid layout), are investigated using CCHE3D model for emerged, transition, and submerged flow conditions with respect to porcupine height. An optimum hybrid layout showed velocity reductions of 35% in submerged and 70% in emerged conditions, which further increased with multiple porcupine screens.

Protection of river banks is an inseparable part of river training works. Permeable and impermeable structures are most commonly used for riverbank protection. Porcupines impose a mild impact on the river by implementing its effect gradually. However, during high flow conditions, these structures are ineffective and often get washed away. On the other hand, impermeable spurs impose a sudden impact on the river system and drastically reduce the velocity in its zone of influence. Due to this, turbulence is generated near the nose of the structure leading to the formation of scour hole, which results in structural instability. Therefore, an attempt has been made to study the effectiveness of the interventions mentioned above in stabilizing and protecting the rivers. Due to several limitations of the physical models, such as scale effect, steady-state flow, and high cost, which make it difficult to carry out in the case of a braided river system, a three-dimensional hydrodynamic model was used. In this study, the performance of the 3D hydrodynamic model CCHE 3D is evaluated in terms of velocity reduction potential by comparing it with experimental results. It was observed that initially, the velocity was in the range of 0.1 m/s under emergent condition, which reduced by more than 50% in the downstream of single porcupine screen, more than 75% in the downstream of two porcupine screens, and more than 94.36% in case of geobag layout. Flow deflection was also observed, but it was not significant.

Riverbank erosion is widespread in alluvial rivers in India and elsewhere. River training works are frequently used to aid in the prevention of these losses by regulating the river and therefore protecting critical human habitats. These structures often become unstable and incapable of performing adequately during periods of heavy flooding. For the first time, the three-dimensional hydrodynamic open-source Open Field Operation and Manipulation (OpenFOAM) model is used to assess the potential of a novel hybrid river training arrangement to reduce downstream flow velocity and divert downstream flow to the opposite bank. The results indicate that for single-phase approximation, algorithms such as the Semi Implicit Method for Pressure-Linked Equations (SIMPLE) with lower computational requirements can satisfactorily reproduce flow patterns discovered in the laboratory (𝑅² > 0.74). The hybrid configuration outperforms the porcupine and geobag layouts. When compared to geobag, dual-screen porcupine, and single-screen porcupine, its downstream velocity decreases by 1.33%, 11.62%, and 13.34%, respectively. Similarly, flow diversion to the opposing bank increases by 0.49%, 0.65%, and 0.92%. Thus, the porcupine structure reduces the intensity of the incoming flow prior to it reaching the impermeable geobag in a hybrid layout. It dissipates the flow energy to the point where it can no longer scour the bed, thereby eliminating the disadvantage associated with the formation of scour holes.

Bank erosion is a regular occurrence along most rivers. In low-income nations such as India and Bangladesh, economical engineered structures such as porcupines and geobags have been used to counteract such erosions. Nonetheless, at times of ex- treme flooding, these structures often be- come unstable and are subsequently washed away, thereby failing to protect the banks. Vetiver grass, which ties the soil with its roots, is a natural method for preventing bank erosion. However, its flexible struc- ture is unable to significantly reduce veloc- ity. In this study, the OpenFOAM open- source hydrodynamic model was used to assess the efficacy of mangrove root struc- ture in reducing flow velocity. It has been compared to single screen porcupine, dual screen porcupine, and geobag structure in terms of performance in downstream flow velocity reduction. It was observed that sin- gle screen porcupine was the least effective at reducing velocity (0.32 %), followed by dual screen porcupine (3.63 %) and single geobag (5.66 %). On the other hand, the mangrove structure was able to lower down- stream velocity by 14.26%. In terms of its downstream influence, the single screen por- cupine had its influence upto 3.63 cm, fol- lowed by dual screen porcupine with 5.53 cm, and single geobag with 13.03 cm. The mangrove structure influence zone on the other hand was very close to the geobag structure (11.53 cm). With its greater ve- locity reduction capabilities and a consid- erable zone of influence, mangrove planta- tions on riverbanks may therefore function as a cost-effective and ecologically sustain- able soil erosion management strategy.

In this study, chlorophyll indices already developed for selected rivers in Brazil, USA and India are applied to a reach in Brahmaputra River in India, to test their applicability in this large braided river system and also to compare their results. Linear, logarithmic, exponential and quadratic relations of Blue, Green, Red and Near Infrared surface reflectance of Sentinel 2A and Landsat 8 imageries are used in this study to determine the chlorophyll indices. It is found that there is a significant variation of the values obtained by these algorithms. For algorithms using Landsat Imageries, concentrations obtained were in the range 40.727-261.836 mg/L and for algorithms using Sentinel Imageries, concentrations obtained were in the range 1.092-25.612 mg/L in the month of February 2021. This clearly indicates that the algorithms already developed for other regions can only be applied after validation, or new site-specific algorithms need to be developed. Thus, google earth engine can be effectively used for real time ecological monitoring of large water bodies (lakes) and rivers provided the algorithms used are first validated for the region of interest.

For any satellite-based study, ground-based reflectance values are required. Nowadays, these ground-based reflectance products are provided by all the major space agencies, commonly designated as level 2 products. However, the availability of the level 2 products takes some time and once in a while, these products are not available. For the development of near real-time monitoring systems, this poses a major problem, and thus it becomes necessary to correct the raw satellite imagery by using atmospheric correction techniques. The Dark Object Subtraction technique (DOS) is one such commonly used technique used previously in many studies. However, it requires the manual selection of the darkest pixels in the imagery, thus making it unsuitable for automation-based systems. This study aims to automate the process of Dark Object Subtraction Sentinel 2A raw satellite imageries within the Google Earth Engine platform. Mean annual LULC maps generated using automated Dark Object Subtraction could replicate the level 2 product quite accurately. These classified imageries for July 2018−July 2019 produced overall classification accuracies of 74.13 and 67.24% using Random Forest Classifier and Support Vector Machines, respectively, compared to 68.96% obtained for both the classification algorithms using level 2 products. In the period July 2019−July 2020, it was obtained as 81.03 and 77.58%, respectively, compared to 79.31% for the same, and for July 2020- July 2021, it was 72.41 and 68.96% against 68.96 and 67.24%. The automated Dark Object Subtraction technique can thus be employed to develop near real-time automated satellite imagery-based systems within the Google Earth Engine platform.

River bathymetry is a preliminary requirement for studies involving modeling or flood forecasting. Field- based bathymetry mapping has traditionally been conducted for bathymetry generation; however, this approach would be uneconomical and tedious for highly morpho- dynamic rivers like the Brahmaputra in India. Satellite- based remote sensing techniques have been used to overcome these shortcomings, thus enabling rapid and continuous monitoring. Microwave remote sensing combined with satellite altimetry is commonly used to estimate water depths. However, these depths are limited only to the altimeter tracks. Such discontinuous data will lead to uncertainties in the depths of regions between altimeter tracks. For continuous depths, optical remote sensing have also been used in some studies to map bathymetry for coastal regions. However, applications of optical remote sensing to map bathymetry for large braided rivers have not been explored. In this study, using Google Earth Engine cloud computing, Sentinel 2 optical imageries have been used to map river bathymetry using Lyzenga (1985) and Stumpf (2003) algorithms, which have been calibrated using LIDAR data for coastal region of Belize in Central America. The range of values predicted by Stumpf algorithm are in the range of 0.743 m- 6.307 m corresponding to 1 m - 8 m ADCP collected field values in the shallow water Palasbari region of Assam, India. On the other hand, Lyzenga algorithm predicted depths in the range of 15.288 m - 18.442 m, which was very close to the range of values obtained in the field for deep water Pandu region of Assam, India. It was observed that the Stumpf algorithm could predict water depth for wide stretches, whereas the Lyzenga algorithm could be used for narrow stretches in the Brahmaputra River. A combination of Lyzenga and Stumpf algorithms will help in mapping approximate bathymetry for large braided rivers like the River Brahmaputra, where conducting field surveys is very challenging.

In water resources management, predictive services are essential to support sustainable planning and operations over a range of time scales, from the short term (days) to the medium term (seasons) to the long term (years to decades). Current forecasting tools mainly address water availability (i.e., quantity), with limited practical applications for water quality. Within the framework of the project called “Strumenti di monitoraggio e previsionali sullo stato di QUalità delle Acque Superficiali” (SQUAS; founded by CARITRO Foundation, Italy; website: https://sites.google.com/unitn.it/hydrosquas), we aim to fill this gap, which is particularly relevant in view of the ongoing transformation of water resources due to rapidly changing climatic conditions. More specifically, we aim to 1) increase the accessibility of tools for diagnosing and predicting surface water quality for use by local authorities and managers of surface water resources, such as agricultural consortia, hydroelectric plant operators, municipal companies, and public entities, and 2) improve the ability of these entities to plan and manage water resources efficiently and sustainably. Anchored in a multidisciplinary approach, the project integrates physical-based modelling used to forecast key water quality parameters with satellite remote sensing data for monitoring purposes. As for the modelling component, the project will be based on the widely used air2water and air2stream models for water temperature prediction in lakes and rivers. Central to the project is the revision, improvement and extension of these models by including water quality variables (e.g., turbidity, dissolved oxygen) and by integrating them into a state-of-the-art web-based Geographic Information System (GIS) platform. The web-GIS platform will not only allow to forecast future conditions based on the above models but also allow for real-time monitoring of water quality. Its Python fast-api based interface will provide a user-friendly GUI for the user interaction, using any web browser. The speed of computation of the forecasting models will be ensured by efficient Cython-based functions. The intuitive interface of the web-GIS platform will appeal to a wide range of users, from policy makers and water resource managers to academic researchers, facilitating informed decision-making and sustainable management practices. An interactive presentation of the web-GIS tool will be given during the session.

Existing studies on flow through vegetation have focused mainly on understanding the turbulent structure in vegetated channels of single plant type. However, in natural riverine environments, vegetation also occurs as patches with heterogeneous plant forms. The present paper investigates the flow and turbulent characteristics in heterogeneous vegetation patches at a laboratory scale. Experiments were conducted using different forms (grass, leafy and cylindrical) of natural vegetation planted, alternatively and also as a mixed variety of patches in a staggered pattern. The results show that the presence of other vegetation forms in mixed heterogeneous patch increases the velocity reduction up to 10% compared to flexible grass. Moreover, additional drag due to mixed vegetation reduces shear generated turbulence at the canopy top and shifts its peak above the canopy. In the case of heterogeneous patches, spatial heterogeneity in velocity fields and, varying zones of increased and diminished turbulence were observed. Specifically, patch form and its alignment significantly control the velocity reduction and, momentum transfer between the canopy and overflow regions. These findings and furthermore studies on heterogeneous patches may be helpful for riparian management practices in creating ecological and sediment deposition zones.

The evolution of fluvial systems is greatly impacted by mid-channel bars, a typical morphodynamic process in natural rivers. Sometimes, the growth of vegetation over these bars complicates the morphological behaviour by interacting with the flow. It is therefore necessary to have a fundamental interpretation of the flow-turbulence structure around the mid-bar in presence of vegetation cover in order to understand braiding dynamics, still studies in this area are scarce. The present study investigates the process-form-vegetation-interaction through experimental investigation at a flume scale mid-channel bar model with different natural vegetation cover arrangements (paddy, leafy, and rigid stem). The flow-turbulence behaviour has been observed through the bifurcated channel using the three-dimensional Acoustic Doppler Velocimeter (ADV). Results showed that the longitudinal velocity component varies with the different vegetation cover, and it was highest with leafy vegetation (about 32%). Similarly, the Reynolds Stress and Turbulence Intensity were also observed to be higher in case of leafy vegetation. A unique pattern of flow-turbulence parameters was observed near the bar level, the lower canopy level, and the upper canopy level. Moreover, it was found that vegetation structure and its flexible nature influence both longitudinal velocity reduction and momentum transfer at and over the canopy, as well as the thickness of the shear layer region.