Comparison of polynomial chaos and Gaussian process surrogates for uncertainty quantification and correlation estimation of spatially distributed open-channel steady flows

Data assimilation is widely used to improve flood forecasting capability, especially through parameter inference requiring statistical information on the uncertain input parameters (upstream discharge, friction coefficient) as well as on the variability of the water level and its sensitivity with respect to the inputs. For particle filter or ensemble Kalman filter, stochastically estimating probability density function and covariance matrices from a Monte Carlo random sampling requires a large ensemble of model evaluations, limiting their use in real-time application. To tackle this issue, fast surrogate models based on polynomial chaos and Gaussian process can be used to represent the spatially distributed water level in place of solving the shallow water equations. This study investigates the use of these surrogates to estimate probability density functions and covariance matrices at a reduced computational cost and without the loss of accuracy, in the perspective of ensemble-based data assimilation. This study focuses on 1-D steady state flow simulated with MASCARET over the Garonne River (South-West France). Results show that both surrogates feature similar performance to the Monte-Carlo random sampling, but for a much smaller computational budget; a few MASCARET simulations (on the order of 10–100) are sufficient to accurately retrieve covariance matrices and probability density functions all along the river, even where the flow dynamic is more complex due to heterogeneous bathymetry. This paves the way for the design of surrogate strategies suitable for representing unsteady open-channel flows in data assimilation.     

Using stable water isotopes to identify spatio-temporal controls on groundwater recharge in two contrasting East African aquifer systems

Understanding the spatio-temporal variability in groundwater recharge is a prerequisite to sustainable management of aquifers. Spatial analysis of groundwater stable isotopes uncovered predominant controls on groundwater recharge in the Nairobi aquifer system (NAS) and the South Coast aquifer (SC), two exemplar East African aquifers relied upon by 7 million people. We analysed 368 samples for stable isotopes and basic physico-chemical parameters. The NAS groundwater isotopes are controlled by precipitation orographic effects and enriched recharge from impounded lakes/wetlands; the SC isotopes are correlated with water-table depth influencing evapotranspiration. Global Network of Isotopes in Precipitation (GNIP) data revealed groundwater recharge during months of heavy rains in the NAS, whilst the SC experiences spatio-temporally diffuse recharge. Inferred “isoscapes” show: in the NAS, (1) direct, rapid recharge favoured by faults, well-drained soils and ample rainfall in uplands, (2) delayed recharge from impounded lakes and wetlands in mid-lands, and (3) focused, event-based recharge in floodplains; and in the SC, diffuse recharge complicated by significant water-table evapotranspiration processes.     

Pre-Permian uplift and diffuse extensional deformation in the High Zagros Belt (Iran): integration in the geodynamic evolution of the Arabian plate

The High Zagros Belt includes exposures of Lower Palaeozoic rocks in the core of several thrusted anticlines developed during the Cenozoic Zagros orogeny. The structural style and tectonic evolution of this area during the Palaeozoic remain poorly understood due to the complexity of the subsequent deformation. We present the preliminary results of a field study focusing on the structural geology of Palaeozoic rocks in this area. We confirm the existence of an angular unconformity below the Lower Permian Faraghan formation. In the geological literature, this unconformity is reported as the “Hercynian Unconformity” suggesting a relationship with the Hercynian (Variscan) orogeny, which affected Western Europe and westernmost Africa during the Carboniferous. Surprisingly the only observable structures sealed by this unconformity are N to NE trending normal faults and tilted blocks without any evidence of compressional deformation. This pre-Permian extensional deformation, which is general at the scale of the HZB, raises questions about the geodynamic significance of the “Hercynian unconformity” in the study area and, more generally, in the Arabian plate.     

Formation and preservation of pedogenic carbonates in South India, links with paleo-monsoon and pedological conditions: Clues from Sr isotopes, U-Th series and REEs

The influence of the pedogenic and climatic contexts on the formation and preservation of pedogenic carbonates in a climosequence in the Western Ghats (Karnataka Plateau, South West India) has been studied. Along the climosequence, the current mean annual rainfall (MAR) varies within a 80 km transect from 6000 mm at the edge of the Plateau to 500 mm inland. Pedogenic carbonates occur in the MAR range of 500-1200 mm. In the semi-arid zone (MAR: 500-900 mm), carbonates occur (i) as thick hardpan calcretes on pediment slopes and (ii) as nodular horizons in polygenic black soils (i.e. vertisols). In the sub-humid zone (MAR: 900-1500 mm), pedogenic carbonates are disseminated in the black soil matrices either as loose, irregular and friable nodules of millimetric size or as indurated botryoidal nodules of centimetric to pluricentimetric size. They also occur at the top layers of the saprolite either as disseminated pluricentimetric indurated nodules or carbonate-cemented lumps of centimetric to decimetric size.Chemical and isotopic (⁸⁷Sr/⁸⁶Sr) compositions of the carbonate fraction were determined after leaching with 0.25 N HCl. The corresponding residual fractions containing both primary minerals and authigenic clays were digested separately and analyzed. The trend defined by the ⁸⁷Sr/⁸⁶Sr signatures of both labile carbonate fractions and corresponding residual fractions indicates that a part of the labile carbonate fraction is genetically linked to the local soil composition. Considering the residual fraction of each sample as the most likely lithogenic source of Ca in carbonates, it is estimated that from 24% to 82% (55% on average) of Ca is derived from local bedrock weathering, leading to a consumption of an equivalent proportion of atmospheric CO₂. These values indicate that climatic conditions were humid enough to allow silicate weathering: MAR at the time of carbonate formation likely ranged from 400 to 700 mm, which is 2- to 3-fold less than the current MAR at these locations.The Sr, U and Mg contents and the (²³⁴U/²³⁸U) activity ratio in the labile carbonate fraction help to understand the conditions of carbonate formation. The relatively high concentrations of Sr, U and Mg in black soil carbonates may indicate fast growth and accumulation compared to carbonates in saprolite, possibly due to a better confinement of the pore waters which is supported by their high (²³⁴U/²³⁸U) signatures, and/or to higher content of dissolved carbonates in the pore waters. The occurrence of Ce, Mn and Fe oxides in the cracks of carbonate reflects the existence of relatively humid periods after carbonate formation. The carbonate ages determined by the U-Th method range from 1.33 ± 0.84 kyr to 7.5 ± 2.7 kyr and to a cluster of five ages around 20 kyr, i.e. the Last Glacial Maximum period. The young occurrences are only located in the black soils, which therefore constitute sensitive environments for trapping and retaining atmospheric CO₂ even on short time scales. The maximum age of carbonates depends on their location in the climatic gradient: from about 20 kyr for centimetric nodules at Mule Hole (MAR = 1100 mm/yr) to 200 kyr for the calcrete at Gundlupet (MAR = 700 mm/yr, Durand et al., 2007). The intensity of rainfall during wet periods would indeed control the lifetime of pedogenic carbonates and thus the duration of inorganic carbon storage in soils.     

Estimating aquifer thickness using multiple pumping tests

A method to estimate aquifer thickness and hydraulic conductivity has been developed, consisting of multiple pumping tests. The method requires short-duration pumping cycles on an unconfined aquifer with significant seasonal water-table fluctuations. The interpretation of several pumping tests at a site in India under various initial conditions provides information on the change in hydrodynamic parameters in relation to the initial water-table level. The transmissivity linearly decreases compared with the initial water level, suggesting a homogeneous distribution of hydraulic conductivity with depth. The hydraulic conductivity is estimated from the slope of this linear relationship. The extrapolation of the relationship between transmissivity and water level provides an estimate of the aquifer thickness that is in good agreement with geophysical investigations. The hydraulically active part of the aquifer is located in both the shallow weathered and the underlying densely fractured zones of the crystalline basement. However, no significant relationship is found between the aquifer storage coefficient and initial water level. This new method contributes to filling the methodological gap between single pumping tests and hydraulic tomography, in providing information on the variation of the global transmissivity according to depth. It can be applied to any unconfined aquifer experiencing large seasonal water-table fluctuations and short pumping cycles.     

A landscape mosaics approach for characterizing swidden systems from a REDD+ perspective

Swidden agriculture is often deemed responsible for deforestation and forest degradation in tropical regions, yet swidden landscapes are commonly not visible on land cover/use maps, making it difficult to prove this assertion. For a future REDD+ scheme, the correct identification of deforestation and forest degradation and linking these processes to land use is crucial. However, it is a key challenge to distinguish degradation and deforestation from temporal vegetation dynamics inherent to swiddening. In this article we present an approach for spatial delineation of swidden systems based on landscape mosaics. Furthermore we introduce a classification for change processes based on the change matrix of these landscape mosaics. Our approach is illustrated by a case study in Viengkham district in northern Laos. Over a 30-year time period the swidden landscapes have increased in extent and they have degraded, shifting from long crop-fallow cycles to short cycles. From 2007 to 2009 degradation within the swidden system accounted for half of all the landscape mosaics change processes. Pioneering shifting cultivation did not prevail. The landscape mosaics approach could be used in a swidden compatible monitoring, reporting and verification (MRV) system of a future REDD+ framework.     

Optimizing a piezometric network in the estimation of the groundwater budget: a case study from a crystalline-rock watershed in southern India

An estimate of the groundwater budget at the catchment scale is extremely important for the sustainable management of available water resources. Water resources are generally subjected to over-exploitation for agricultural and domestic purposes in agrarian economies like India. The double water-table fluctuation method is a reliable method for calculating the water budget in semi-arid crystalline rock areas. Extensive measurements of water levels from a dense network before and after the monsoon rainfall were made in a 53 km² watershed in southern India and various components of the water balance were then calculated. Later, water level data underwent geostatistical analyses to determine the priority and/or redundancy of each measurement point using a cross-validation method. An optimal network evolved from these analyses. The network was then used in re-calculation of the water-balance components. It was established that such an optimized network provides far fewer measurement points without considerably changing the conclusions regarding groundwater budget. This exercise is helpful in reducing the time and expenditure involved in exhaustive piezometric surveys and also in determining the water budget for large watersheds (watersheds greater than 50 km²).     

Coamps<Superscript>®</Superscript>-Les: Model Evaluation and Analysis of Second-and Third-Moment Vertical Velocity Budgets

The Naval Research Laboratory Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPS^®) has been extended to perform as a large-eddy simulation (LES) model. It has been validated with a series of boundary-layer experiments spanning a range of cloud nighttime, and includes a nighttime stratocumulus case, a trade wind cumulus layer, shallow cumulus convection over land, and a mixed regime consisting of cumulus clouds under broken stratocumulus. COAMPS-LES results are in good agreement with other models for all the cases simulated. Exact numerical budgets for the vertical velocity second\((\overline{w^{'2}})\) and third moment\((\overline{w^{'3}})\) have been derived for the stratocumulus and trade wind cumulus cases. For the\(\overline{w^{'3}}\) budget in the stratocumulus, the buoyancy contribution from the updraughts and downdraughts largely cancel each other due to their similar magnitudes but opposite signs. In contrast, for the cumulus layer, the negative buoyancy contribution from the environmental downdraughts is negligible and the positive contribution from the updraughts completely dominates due to the conditional instability in the environment. As a result,\(\overline{w^{'3}}\) is significantly larger in the cumulus than in the stratocumulus layer.