Fifty years of runoff response to conversion of old-growth forest to planted forest in the H. J. Andrews Forest,Oregon, USA

Long-term watershed experiments provide the opportunity to understand forest hydrology responses to past logging, road construction, forest regrowth, and their interactions with climate and geomorphic processes such as road-related landslides. We examined a 50-year record from paired-watershed experiments in the H. J. Andrews Experimental Forest, Oregon, USA in which 125 to 450-year-old conifer forests were harvested in the 1960s and 1970s and converted to planted conifer forests. We evaluated how quickflow and delayed flow for 1222 events in treated and reference watersheds changed by season after clearcutting and road construction, including 50 years of growth of planted forest, major floods, and multi-decade reductions in snowpack. Quickflow runoff early in the water year (fall) increased by up to +99% in the first decade, declining to below pre-harvest levels (−1% to −15%) by the third to fifth decade after clearcutting. Fall delayed flow responded more dramatically than quickflow and fell below pre-treatment levels in all watersheds by the fifth decade, consistent with increased transpiration in the planted forests. Quickflow increased less (+12% to 70%) during the winter and spring but remained higher than pre-treatment levels throughout the fourth or fifth decade, potentially impacted by post-harvest burning, roads, and landslides. Quickflow remained high throughout the 50-year period of study, and much higher than delayed flow in the last two decades in a watershed in which road-related changes in flow routing and debris flows after the flood of record increased network connectivity. A long-term decline in regional snowpack was not clearly associated with responses of treated vs. reference watersheds. Hydrologic processes altered by harvest of old-growth conifer forest more than 50 years ago (transpiration, interception, snowmelt, and flow routing) continued to modify streamflow, with no clear evidence of hydrologic recovery. These findings underscore the importance of continued long-term watershed experiments.     

Neural network based daily precipitation generator (NNGEN-P)

Daily weather generators are used in many applications and risk analyses. The present paper explores the potential of neural network architectures to design daily weather generator models. Focusing this first paper on precipitation, we design a collection of neural networks (multi-layer perceptrons in the present case), which are trained so as to approximate the empirical cumulative distribution (CDF) function for the occurrence of wet and dry spells and for the precipitation amounts. This approach contributes to correct some of the biases of the usual two-step weather generator models. As compared to a rainfall occurrence Markov model, NNGEN-P represents fairly well the mean and standard deviation of the number of wet days per month, and it significantly improves the simulation of the longest dry and wet periods. Then, we compared NNGEN-P to three parametric distribution functions usually applied to fit rainfall cumulative distribution functions (Gamma, Weibull and double-exponential). A data set of 19 Argentine stations was used. Also, data corresponding to stations in the United States, in Europe and in the Tropics were included to confirm the results. One of the advantages of NNGEN-P is that it is non-parametric. Unlike other parametric function, which adapt to certain types of climate regimes, NNGEN-P is fully adaptive to the observed cumulative distribution functions, which, on some occasions, may present complex shapes. On-going works will soon produce an extended version of NNGEN to temperature and radiation.     

On zero-thickness interface elements for diffusion problems

The present study focuses on the hydraulic behaviour of joints, and, specially, on its numerical implementation in terms of the FEM analysis using a discrete fracture flow approach. Fluid flow through discontinuities has traditionally been modelled using special elements of zero-thickness, which we can classify into single, double and triple-nodded. Single node elements are the simplest and consist of ‘line’ or ‘pipe’ elements which are superimposed onto the standard continuum mesh and that can only model the longitudinal conductivity through the discontinuity. On the other hand, some authors have included the influence of a transversal conductivity, and the subsequent localized potential drop, by using triple node interface elements. In those, the two nodes of the adjacent continuum elements represent the potentials in the pore system on each side of the interface, and a third node in the middle represents the average potential of the fluid in the channel represented by the discontinuity. Finally, double node interface elements have also been proposed, which have the advantage of making it possible to use the same FE mesh for both mechanical and flow analysis. In some cases the influence of a transversal conductivity is not considered and, therefore, although geometrically double-nodded, these elements belong to the single node type and when time comes to solve the system the two nodes must have the same potential, which can only be obtained by the ‘trick’ of prescribing the equivalence of these two d.o.f. before solving the global system of equations. This limitation may, however, be avoided by assuming that the potential in the channel is the average of the two sides of the interface. Based in this simple assumption, an alternative flow interface model has been recently developed and implemented, which preserves both longitudinal and transversal conductivities. An application example is developed and solved with the three types of interfaces described. The results offer useful information regarding the range of applicability and limitations of the new double-nodded interface element proposed. Copyright © 2004 John Wiley & Sons, Ltd.     

Characterizing stream temperature hysteresis in forested headwater streams

Stream temperature (T_s) is a key water quality parameter that controls several biological, ecological, and chemical processes in aquatic systems. In forested headwaters, exchanges of energy across air-water-streambed interfaces may influence T_s regimes, especially during storm events as the sources of runoff change over space and time. Analysis of the hysteretic behaviour of T_s during storm events may provide insights into rainfall-runoff responses, but such relationships have not been thoroughly investigated. As such, our objectives were to (a) quantify the variability of stream temperature hysteresis across seasons in different sub-regions and (b) investigate the relationship between the hysteretic response and catchment characteristics. T_s hysteresis during storm events was assessed based on the hysteresis index (HI), which describes the directionality of hysteresis loops, and the temperature response index (TRI), which indicates whether T_s increased or decreased during a storm event. We analysed T_s data from 10 forested headwater reaches in two sub-regions (McGarvey and West Fork Tectah) in Northern California. We also performed a clustering analysis to examine the relationship amongst HI, TRI, topographic metrics, and meteorological characteristics of the study areas. Overall, the hysteretic behaviour of T_s varied across seasons—the greatest HI occurred during spring and summer. Interestingly, in the McGarvey streams the variability in T_s hysteresis co-varied strongly with topographic metrics (i.e., upslope accumulative area, average channel slope, topographic wetness index). Comparatively, in West Fork Tectah the variability of T_s hysteresis co-varied most strongly with meteorological metrics (i.e., antecedent rainfall events, solar radiation, and air temperature). Variables such as the gradient between stream and air temperatures, slope, and wetted width were significant for both sub-regional hysteretic patterns. We posit that the drivers of T_s response during storms are likely dependent on catchment physiographic characteristics. Our study also illustrated the potential utility of stream temperature as a tracer for improving the understanding of hydrologic connectivity and shifts in the dominant runoff contributions to streamflow during storm events.     

General patterns of circulation, sediment fluxes and ecology of the Palamós (La Fonera) submarine canyon, northwestern Mediterranean

Currents, particle fluxes and ecology were studied in the Palamós submarine canyon (also known as the Fonera canyon), located in the northwestern Mediterranean. Seven mooring arrays equipped with current meters and sediment traps were deployed along the main canyon axis, on the canyon walls and on the adjacent slope. Additionally, local and regional hydrographic cruises were carried out. Current data showed that mean near surface and mid-depth currents were oriented along the mean flow direction (NE–SW), although at 400 and 1200 m depth within the canyon current reversals were significant, indicating a more closed circulation inside the canyon. Mean near-bottom currents were constrained by the local bathymetry, especially at the canyon head. The most significant frequency at all levels was the inertial frequency. A second frequency of about three days, attributed to a topographic wave, was observed at all depths, suggesting that this wave was probably not trapped near the bottom. The current field observed during the most complete survey revealed a meandering pattern with cyclonic vorticity just upstream from and within the canyon. The associated vertical velocity ranged between 10 and 20 m/day and was constrained to the upper 300 m. This latter feature, together with other computations, suggests that during this survey the meander was not induced by the canyon but by some kind of instability of the mean flow.In the canyon, suspended sediment concentration, downward particle fluxes, chlorophyll and particulate C and N were significantly higher up-canyon from about 1200 m depth than offshore, defining, along with the different hydrodynamics, two canyon domains: one from the canyon head to about 1200 m depth more affected by the canyon confinement and the other deeper than 1200 m depth more controlled by the mean flow and the shelf-slope front. The higher near-bottom downward total mass fluxes were recorded in the canyon axis at 1200 m depth along with sharp turbidity increases and are related to sediment gravity flows. During the deployment period, the increase in downward particle fluxes occurred by mid-November, when a severe storm took place. On the canyon walls at 1200 m depth, suspended sediment concentrations, downward particle fluxes, chlorophyll and particulate C and N were higher on the southern wall than on the northern wall inversely to the current’s energy. This could be caused by an upward water supply on the southern canyon wall and/or the mean flow interacting with the canyon bathymetry. In the swimmers collected by the sediment traps, the dominant species was an elasipod holothurian, which has not been recorded in other canyons or elsewhere in the Mediterranean, indicating particular speciation.     

Loggerhead turtles Caretta caretta (Linnaeus) preying on the invading gastropod Rapana venosa (Valenciennes) in the Río de la Plata Estuary

Here we report the first observations of loggerhead turtles Caretta caretta preying on a conspicuous molluscan invader, the rapa whelk Rapana venosa. An average number of 136 opercula were found in stomach contents of five turtles, the curved carapace length ranging in from 51 to 112 cm. No other alimentary items were found in the turtles analyzed. We suggest that the rapa whelk may constitute up to 100% of the diet for immature and mature loggerheads in the Río de la Plata estuary (Uruguay), highlighting the plastic nature of the foraging behavior of loggerheads.     

The effect of salinity on clearance rate in the suspension-feeding estuarine gastropod Crepipatella dilatata under natural and controlled conditions

The suspension-feeding gastropod Crepipatella dilatata occurs in estuaries in southern Chile that experience considerable fluctuations in salinity, driven by tidal and atmospheric forces. In the Quempillén estuary salinities as low as 9 psu may occur after severe rainstorms, and persist for several hours. In this study salinity was the major factor influencing the clearance rate of C. dilatata. At salinities below 20 psu, filtration ceased, whereas at high salinities (>22 psu) mean clearance rate was 0.24 l h⁻¹ standard animal⁻¹ (S.D. 0.18) for actively filtering individuals. This was confirmed by laboratory experiments under controlled conditions. Endoscopic observations were consistent with measurements of clearance rate, and showed that at salinity 25–30 psu the rate of transport along the gill filaments of particulate material embedded in mucus was 759 μm s⁻¹ (S.D. 480), but particle transport ceased at and below salinity 20 psu. Complete or partial isolation of the mantle cavity from the environment may be a mechanism to protect soft tissues and/or incubated egg capsules from osmotic stress.     

Relation between flow,surface‐layer armoring and sediment transport in gravel‐bed rivers

This study investigates trends in bed surface and substrate grain sizes in relation to reach‐scale hydraulics using data from more than 100 gravel‐bed stream reaches in Colorado and Utah. Collocated measurements of surface and substrate sediment, bankfull channel geometry and channel slope are used to examine relations between reach‐average shear stress and bed sediment grain size. Slopes at the study sites range from 0·0003 to 0·07; bankfull depths range from 0·2 to 5 m and bankfull widths range from 2 to 200 m. The data show that there is much less variation in the median grain size of the substrate, D_50s, than there is in the median grain size of the surface, D₅₀; the ratio of D₅₀ to D_50s thus decreases from about four in headwater reaches with high shear stress to less than two in downstream reaches with low shear stress. Similar trends are observed in an independent data set obtained from measurements in gravel‐bed streams in Idaho. A conceptual quantitative model is developed on the basis of these observations to track differences in bed load transport through an idealized stream system. The results of the transport model suggest that downstream trends in total bed load flux may vary appreciably, depending on the assumed relation between surface and substrate grain sizes. Copyright © 2007 John Wiley & Sons, Ltd.     

Empirical evaluation of site effects in the metropolitan area of San José, Costa Rica

Site effects for 11 selected locations were determined in the capital city of Costa Rica. We used a strong motion network made of eight K2 and three SSA accelerographs. The network recorded more than 60 earthquakes in the magnitude range from 2 to 5 during a period of nine months. The site effects were determined using the sediment-to-bedrock spectral ratio (SBSR) and the horizontal-to-vertical spectral ratio (HVSR) techniques and a time window 4 s beginning from the S-wave arrival. The result suggests that the amplification in the capital city is to be in the range from 2.0 to 3.0. The fundamental frequencies were found to be high in the southern and eastern part of the study area and low in the northern and western part. A possible topographic effect was also observed for one of the stations located nearby a river canyon. The results from earthquake data were compared with the ones obtained from noise data. The horizontal-to-vertical noise ratio (HVNR) technique was used to estimate the site effects using ambient noise. The fundamental frequencies were found to correlate very well between both sets of data; on the other hand, the amplitude given by the noise was observed to be always lower than the one derived from the earthquake data.     

CHAC: a weather pattern classification system for regional climate downscaling of daily precipitation

A weather pattern clustering method is applied and calibrated to Argentinean daily weather stations in order to predict daily precipitation data. The clustering technique is based on k-means and is applied to a set of 17 atmospheric variables from the ERA-40 reanalysis covering the period 1979–1999. The set of atmospheric variables represent the different components of the atmosphere (dynamical, thermal and moisture). Different sensitivity tests are applied to optimize (1) the number of observations (weather patterns) per cluster, (2) the spatial domain size of the weather pattern around the station and (3) the number of members of the ensembles. All the sensitivity tests are compared using the ROC (Relative Operating Characteristic) Skill Score (RSS) derived from the ROC curve used to assess the performance of a predictive system. First, we found the number of observations per cluster to be optimum for values larger than 39. Second, the spatial domain size (~4° × 4°) was found to be closer to a local scale than to a synoptic scale, certainly due to a dominant role of the moisture components in the optimization of the transfer function. Indeed, when reducing the set of variables to the subset of dynamical variables, the predictive skill of the method is significantly reduced, but at the same time the domain size must be increased. A potential improvement of the method may therefore be to consider different domains for dynamical and non-dynamical variables. Third, the number of members per ensembles of simulations was estimated to be always two to three times larger than the mean number of observations per cluster (meaning that at least all the observed weather patterns are selected by one member). The skill of the statistical method to predict daily precipitation is found to be relatively homogeneous all over the country for different thresholds of precipitation.