Ocean iron fertilization: Why further research is needed

Despite large uncertainties in the fertilization efficiency, natural iron fertilization studies and some of the purposeful iron enrichment studies have demonstrated that Southern Ocean iron fertilization can lead to a significant export of carbon from the sea surface to the ocean interior. From an economic perspective the potential of ocean iron fertilization (OIF) is far from negligible in relation to other abatement options. Comparing the range of cost estimates to the range of estimates for forestation projects they are in the same order of magnitude, but OIF could provide more carbon credits even if high discount rates are used to account for potential leakage and non-permanence. However, the uncertainty about undesired adverse effects of purposeful iron fertilization on marine ecosystems and biogeochemistry has led to attempts to ban commercial and, to some extent, scientific experiments aimed at a better understanding of the processes involved, effectively precluding further consideration of this mitigation option. As regards the perspective of public international law, the pertinent agreements dealing with the protection of the marine environment indicate that OIF is to be considered as lawful if and to the extent to which it represents legitimate scientific research. In this respect, the precautionary principle can be used to balance the risks arising out of scientific OIF activities for the marine environment with the potential advantages relevant to the objectives of the climate change regime. As scientific OIF experiments involve only comparatively small negative impacts within a limited marine area, further scientific research must be permitted to explore the carbon sequestration potential of OIF in order to either reject this concept or integrate it into the flexible mechanisms contained in the Kyoto Protocol.     

Undersampling in action and at scale: application to the COVID-19 pandemic

Stochastic Environmental Research and Risk Assessment - It is the purpose of this short communication to analyze the possible caveats in the statistical interpretation of collected data,...     

Machine learning based identification of dominant controls on runoff dynamics

Hydrological models used for flood prediction in ungauged catchments are commonly fitted to regionally transferred data. The key issue of this procedure is to identify hydrologically similar catchments. Therefore, the dominant controls for the process of interest have to be known. In this study, we applied a new machine learning based approach to identify the catchment characteristics that can be used to identify the active processes controlling runoff dynamics. A random forest (RF) regressor has been trained to estimate the drainage velocity parameters of a geomorphologic instantaneous unit hydrograph (GIUH) in ungauged catchments, based on regionally available data. We analyzed the learning procedure of the algorithm and identified preferred donor catchments for each ungauged catchment. Based on the obtained machine learning results from catchment grouping, a classification scheme for drainage network characteristics has been derived. This classification scheme has been applied in a flood forecasting case study. The results demonstrate that the RF could be trained properly with the selected donor catchments to successfully estimate the required GIUH parameters. Moreover, our results showed that drainage network characteristics can be used to identify the influence of geomorphological dispersion on the dynamics of catchment response.     

Offset dependence of overburden time-shifts from ultrasonic data

Depletion or injection into a reservoir implies stress changes and strains in the reservoir and its surroundings. This may lead to measurable time-shifts for seismic waves propagating in the subsurface. To better understand the offset dependence of time-shifts in the overburden, we have systematically quantified the time-shifts of three different overburden shales in controlled laboratory tests. These experiments may be viewed as an analogue to the time-shifts recorded from seismic field surveys. For a range of different stress paths, defined as the ratio between the horizontal and the vertical stress changes, the changes of the P-wave velocities in different directions were measured such that the offset dependence of time-shifts for different stress paths could be studied. The time-shifts are stress path dependent, which is particularly pronounced at large offsets. For all stress paths, the time-shifts exhibit a linearly decreasing trend with increasing offset, that is, a negative offset gradient. At zero offset, for which the ray path is normal to the bedding, the time-shifts are similar for all investigated stress paths. The isotropic stress path is associated with the smallest offset gradient of the time-shifts. In contrast, the constant-mean-stress path shows the largest gradient with a flip in the polarity of the time-shifts for the largest offsets. The separate contributions from the strain and velocity changes to the time-shifts were also quantified. The time-shifts for the isotropic stress path are dominated by the contribution from velocity changes at all offsets. In contrast, the strain contributes significantly to the time-shifts at small offsets for the constant-mean-stress path. This shows that the offset dependence in pre-stack seismic data may be a key to understand the changes of subsurface stresses, pore pressure and strain upon depletion or injection. To utilize this knowledge from laboratory experiments, calibrated rock physics models and correlations are needed to constrain the seismic time-shifts and to obtain an adequately updated geological model reflecting the true anisotropic nature of the subsurface. This may have important implications for improved recovery and safety, particularly in mature fields.     

Design considerations for using Distributed Acoustic Sensing for cross-well seismics: A case study for CO2 storage

Downhole monitoring with fibre-optic Distributed Acoustic Sensing (DAS) systems offers unprecedented spatial resolution. At the same time, costs are reduced since repeated wireline surveys can be replaced by the permanent installation of comparatively cheap fibre cables. However, the single component nature of fibre data requires novel approaches when designing a monitoring project such as cross-well seismics. At the example of the shallow CO₂ injection test site in Svelvik, Norway, we model the evolution of velocity changes during CO₂ injection based on rock physics theory. Different cross-well seismic design scenarios are then considered to evaluate the best design and the limits of this method to detect containment breach. We present a series of evaluation tools to compare the effect of different well spacings for cross-well seismic tomography. In addition to travel-times, we also consider characteristic amplitude changes along the fibre unique to DAS strain measurements, which might add a constraint to the inversion. We also compare the effect of using helical fibres instead of classical straight fibres. We thus present a toolbox to evaluate and compare different monitoring design options for fibre optic downhole installations for cross-well monitoring.     

Interpretation of tracer tests performed in fractured rock of the Lange Bramke basin, Germany

 Two multitracer tests performed in one of the major cross-fault zones of the Lange Bramke basin (Harz Mountains, Germany) confirm the dominant role of the fault zone in groundwater flow and solute transport. Tracers having different coefficients of molecular diffusion (deuterium, bromide, uranine, and eosine) yielded breakthrough curves that can only be explained by a model that couples the advective–dispersive transport in the fractures with the molecular diffusion exchange in the matrix. For the scale of the tests (maximum distance of 225 m), an approximation was used in which the influence of adjacent fractures is neglected. That model yielded nearly the same rock and transport parameters for each tracer, which means that the single-fracture approximation is acceptable and that matrix diffusion plays an important role. The hydraulic conductivity of the fault zone obtained from the tracer tests is about 1.5×10^–2 m/s, whereas the regional hydraulic conductivity of the fractured rock mass is about 3×10^–7 m/s, as estimated from the tritium age and the matrix porosity of about 2%. These values show that the hydraulic conductivity along the fault is several orders of magnitude larger than that of the remaining fractured part of the aquifer, which confirms the dominant role of the fault zones as collectors of water and conductors of fast flow. Received, April 1997 Revised, January 1998, August 1998 Accepted, August 1998     

Mobility and H2O loss from fluid inclusions in natural quartz crystals

Petrographic observations on quartz crystals from the Mole Granite (Australia) and other localities shed new light on the mechanisms of post-entrapment modification of fluid inclusions. These modifications include migration away from pseudosecondary trails, changes in fluid salinity and density, shape distortion and the formation of “sweat-haloes” around strongly deformed inclusions. Increases in fluid salinity, which usually are associated with inclusion migration, indicate water-losses of up to 50%. However, LA-ICP-MS-analysis of unmobilized and mobilized inclusions of the same trail reveals basically unchanged ratios of major – and trace element cations, with the exception of Li, which seems to be incorporated into the crystal lattice during migration. Despite the fact that all these modifications are closely related to deformation processes, they occur not only in mechanically deformed quartz, but also in free-standing crystals. In the latter samples, stress has been generated internally as a result of brazil-twinned growth and compositional zonation. These observations and their interpretation leads to a list of practical criteria that should help in differentiating between reliable and suspect fluid inclusions in other samples. Received: 17 November 1998 / Accepted: 16 April 1999     

A Density Correction for Lagrangian Particle Dispersion Models

Current Lagrangian particle dispersion models, used to simulate the dispersion of passive tracers in the turbulent planetary boundary layer (PBL), assume that the density is constant within the PBL. In deep PBLs, where the density at the boundary-layer top may be lower by more than 20% than at the surface, this assumption leads to errors in the tracer concentrations on the order of 10%. In the presence of a vertical wind shear, this also leads to inaccurate calculations of the horizontal tracer transport. To remove this deficiency, a Langevin equation is presented that contains a density correction term. The effect of the density correction is studied using data from a large-scale tracer experiment. It is found that for this experiment, the main effect of the density correction is an increase in the surface tracer concentrations, whereas the horizontal tracer transport patterns remain largely unaffected.     

The Transport Time pdf for a Simple Turbulent Flow

In a turbulent fluid, the time a particle needs to travel from a point source to the observation point, can be considered as a random variable. It is shown that the probability density function (pdf) for this random variable is determined by the Lagrangian particle position pdf. The characteristics of the transport time pdf are discussed for the simple case of a turbulent fluid moving with a constant mean velocity.