Response to comment by Keith Beven on “Equifinality of formal (DREAM) and informal (GLUE) Bayesian approaches in hydrologic modeling?”

In recent years, a strong debate has emerged in the hydrologic literature regarding what constitutes an appropriate framework for uncertainty estimation. Particularly, there is strong disagreement whether an uncertainty framework should have its roots within a proper statistical (Bayesian) context, or whether such a framework should be based on a different philosophy and implement informal measures and weaker inference to summarize parameter and predictive distributions. In this paper, we compare a formal Bayesian approach using Markov Chain Monte Carlo (MCMC) with generalized likelihood uncertainty estimation (GLUE) for assessing uncertainty in conceptual watershed modeling. Our formal Bayesian approach is implemented using the recently developed differential evolution adaptive metropolis (DREAM) MCMC scheme with a likelihood function that explicitly considers model structural, input and parameter uncertainty. Our results demonstrate that DREAM and GLUE can generate very similar estimates of total streamflow uncertainty. This suggests that formal and informal Bayesian approaches have more common ground than the hydrologic literature and ongoing debate might suggest. The main advantage of formal approaches is, however, that they attempt to disentangle the effect of forcing, parameter and model structural error on total predictive uncertainty. This is key to improving hydrologic theory and to better understand and predict the flow of water through catchments.     

Aquatic surface microlayer contamination in chesapeake bay

The aquatic surface microlayer (SMIC), 50 μm thick, serves as a concentration point for metal and organic contaminants that have low water solubility or are associated with floatable particles. Also, the eggs and larvae of many fish and shellfish species float on, or come in contact with, the water surface throughout their early development. The objectives of this study were (1) to determine the present degree of aquatic surface microlayer pollution at selected sites in Chesapeake Bay, and (2) to provide a preliminary evaluation of sources contributing to any observed contamination.Twelve stations located in urban bays, major rivers, and the north central bay were sampled three times, each at 5-day intervals during May 1986. Samples of 1.4–4.1 each were collected from the upper 30–60-μm water surface (surface microlayer, SMIC) using a Teflon-coated rotating drum microlayer sampler. One sample of subsurface water was collected in the central bay.At all stations, concentrations of metals, alkanes, and aromatic hydrocarbons in the SMIC were high compared with one bulk-water sample and with typical concentrations in water of Chesapeake Bay and elsewhere. SMIC contamination varied greatly among the three sampling times, but high mean contaminant levels (total polycyclic aromatic hydrocarbons, 1.9–6.2 μg 1⁻¹; Pb, 4.9–24 μg 1⁻¹; Cu, 4–16 μg 1⁻¹; and Zn, 34–59 μg 1⁻¹) were found at the upper Potomac and northern bay sites. Three separate areas were identified on the basis of relative concentrations of different aromatic hydrocarbons in SMIC samples - the northern bay, the Potomac River, and the cleaner southern and eastern portions of the sampling area.Suspected sources of surface contamination include gasoline and diesel fuel combustion, coal combustion, and petroleum product releases. Concentrations of metals and hydrocarbons, at approximately half the stations sampled, are sufficient to pose a threat to the reproductive stages of some fish and shellfish. Sampling and analysis of the surface microlayer provides a sensitive tool for source identification and monitoring of potentially harmful aquatic pollution.     

Lightning‐flash locations related to the precipitation pattern of the storm

Abstract

Triangulation from paired sferics (100 ± 50 kHz) received at stations 57 km apart gave 42 map locations of lightning flashes belonging to a thunderstorm well‐defined on weather radar. This storm was one of many in a long line along the leading edge of a band of light stratiform precipitation (snow above 3 km). The band was 9 km high and 60 km wide, and the storms were embedded in it. The convective region of the storm studied had a north‐south extent of 20 km, the east‐west extent averaging two‐thirds as much. Cumulus heights progressed from 3 km on the west to 7 ± 1 km on the east. Two‐thirds of the lightning flashes were in the convective region. The remaining third were in 160 km² of the light stratiform precipitation immediately west of it.     

Chironomid-based paleosalinity records in southern British Columbia, Canada: a comparison of transfer functions

Chironomid remains from Big Lake, British Columbia were analysed and paleosalinities were estimated using a pre-existing transfer function and several developed using new regression methods. A two component partial-least-squares model (PLS-2) had the highest coefficient of determination (R² _(Jackknifed) = 0.75) and lowest root-mean-squared error-of-prediction (RMSEP). As compared to the pre-existing model, it was also less sensitive to the influence of rare taxa. Nevertheless, the marginally larger R² _(Jackknifed) and lower RMSEP do not clearly identify a single best model. The models were applied to Big, Mahoney and Kilpoola lakes, revealing the sensitivity of paleosalinity inferences to model selection. A synopsis of chironomid-based paleosalinities in British Columbia and their correspondence with other paleoclimatic data are presented and discussed.     

High-resolution seismic stratigraphy of the late Neogene of the central sector of the Colombian Pacific continental shelf: A seismic expression of an active continental margin

The sedimentary prism of the central Pacific continental shelf of Colombia was affected by regional folding and faulting, and probably later mud diapirism, from the Late Miocene to the Holocene. Interpretation of high-resolution seismic lines (2 s/dt) revealed that the prism consists of 13 high-resolution seismic units, that can be separated into 5 seismic groups.Deposition of the prism and the associated stacking pattern, are probably the response to variable uplift and subsidence in a fore-arc basin that underwent important tectonic events by the end of the Miocene. Throughout the Pliocene, the continental shelf sedimentation was affected by the growing of a dome structure probable due to mud diapirism. This fact caused peripheral faults both normal and reverse that controlled the distribution of some of the seismic units. During the Late Pleistocene (Wisconsin stage?) a eustatic sea level fall caused the shoreline to advance about 50 km westward of its present position. Because of this eustatic sea level change, a strong fluvial dissection took place and is interpreted as the probable extension of the San Juan River to the south of the present day river mouth. Within this framework it is believed that the Malaga and Buenaventura Bays were the passageways of branches of the old drainage system of the San Juan River. The inner branch circulated through the present Buenaventura Bay and runs southward leaving the mark of an apparent valley identified in the seismic information in the eastern sector of the study area. This old fluvial valley and its filling material located in the present day inner continental shelf front of Buenaventura are postulated as important targets to find placer minerals such as gold and platinum.     

利用毛管压力曲线分形分维方法研究流动单元

利用取心井铸体薄片获得的图像资料和毛管压力曲线,通过图像分形几何学方法以分维数的形式定量地表征出了复杂的微观孔隙喉道结构特征,发现能够很好地划分和评价孔隙岩石中油、气、水的渗流差异,可以用于储层微观流动单元表征。文中阐述了岩石微观孔隙喉道结构分形的理论基础、计算方法和应用于表征流动单元的依据。建立了中国西部砾岩低渗透油藏微观孔隙喉道分维数与孔隙度、渗透率之间计算图版,据此在油藏中利用常规测井资料获得的孔隙度、渗透率参数计算微观孔隙喉道分维数,开展全油藏流动单元划分与评价,取得了良好的效果。研究结果表明,利用毛管压力曲线分形分维方法研究储层微观流动单元是一种很有效的途径。     

Microwave emission from a sunspot

From the gyroresonance brightness temperature spectrum of a sunspot, one can determine the magnetic field strength by using the property that microwave brightness is limited above a frequency given by an integer-multiple of the gyrofrequency. In this paper, we use this idea to find the radial distribution of magnetic field at the coronal base of a sunspot in the active region, NOAA 4741. The gyroresonance brightness temperature spectra of this sunspot are obtained from multi-frequency interferometric observations made at the Owens Valley Radio Observatory at 24 frequencies in the range of 4.0–12.4 GHz with spatial resolution 2.2″–6.8″. The main results of present study are summarized as follows: first, by comparison of the coronal magnetic flux deduced from our microwave observation with the photospheric magnetic flux measured by KPNO magnetograms, we show that theo-mode emission must arise predominantly from the second harmonic of the gyrofrequency, while thex-mode arises from the third harmonic. Second, the radial distribution of magnetic fieldsB(r) at the coronal base of this spot (say, 2000–4000 km above the photosphere) can be adequately fitted by $$B(r) = 1420(1 \pm 0.080)\exp \left[ { - \left( {\frac{r}{{11.05''(1 \pm 0.014)}}} \right)^2 } \right]G,$$ wherer is the radial distance from the spot center at coronal base. Third, it is found that coronal magnetic fields originate mostly from the photospheric umbral region. Fourth, although the derived vertical variation of magnetic fields can be approximated roughly by a dipole model with dipole moment 1.6 × 10³⁰ erg G^?1 buried at 11000 km below the photosphere, the radial field distribution at coronal heights is found to be more confined than predicted by the dipole model.