Connectivity of Streams and Wetlands to Downstream Waters: An Integrated Systems Framework

Interest in connectivity has increased in the aquatic sciences, partly because of its relevance to the Clean Water Act. This paper has two objectives: (1) provide a framework to understand hydrological, chemical, and biological connectivity, focusing on how headwater streams and wetlands connect to and contribute to rivers; and (2) briefly review methods to quantify hydrological and chemical connectivity. Streams and wetlands affect river structure and function by altering material and biological fluxes to the river; this depends on two factors: (1) functions within streams and wetlands that affect material fluxes; and (2) connectivity (or isolation) from streams and wetlands to rivers that allows (or prevents) material transport between systems. Connectivity can be described in terms of frequency, magnitude, duration, timing, and rate of change. It results from physical characteristics of a system, e.g., climate, soils, geology, topography, and the spatial distribution of aquatic components. Biological connectivity is also affected by traits and behavior of the biota. Connectivity can be altered by human impacts, often in complex ways. Because of variability in these factors, connectivity is not constant but varies over time and space. Connectivity can be quantified with field‐based methods, modeling, and remote sensing. Further studies using these methods are needed to classify and quantify connectivity of aquatic ecosystems and to understand how impacts affect connectivity.     

Relative Importance of Water‐Quality Stressors in Predicting Fish Community Responses in Midwestern Streams

Fish, habitat, and water chemistry data were collected from 98 streams in the midwestern United States, an area dominated by intense cultivation of row crops, in order to identify important water‐quality stressors to fish communities. We focused on 10 stressors including riparian disturbance, riparian vegetative cover, instream fish cover, streambed sedimentation, streamflow variability, total nitrogen, total phosphorus, minimum dissolved oxygen, pesticides, and bed sediment contaminants. Fish community response variables included a measure of observed/expected taxonomic completeness; species‐specific tolerances to nitrogen, phosphorus, dissolved oxygen, and water temperature; the percent of species classified as macrohabitat generalists; and an index of pesticide toxicity to fish. Multivariate analysis indicated that total nitrogen was the most important stressor, signifying that fish communities were responding to total nitrogen despite relatively high levels common to an agricultural setting. Individually, fish taxonomic completeness decreased with increasing streambed sedimentation, whereas fish community tolerance to total phosphorus increased with increasing streambed sedimentation, riparian disturbance, and total nitrogen. These findings underscore the importance of multiple biological response metrics to better understand the effects of water‐quality stressors on fish communities and highlight the complex relations between total phosphorus and fish communities.     

Landscape Hydrogeology and its Influence on Patterns of Groundwater Flux and Nitrate Removal Efficiency in Riparian Buffers

This study uses data from 46 riparian sites to examine the influence of landscape hydrogeology on patterns of groundwater flux and the buffer width required for effective nitrate removal in humid temperate agricultural regions. There is a considerable imbalance in the research focus on different hydrogeologic settings. More than 40% of the buffers are located in landscapes with surficial sand aquifers, whereas few buffers have been studied in glacial till and weathered bedrock landscapes which cover large areas. Annual groundwater fluxes for 29 of these sites ranged from <20 L/m/day for buffers on flat sand plains and uplands with fine‐textured deposits to 50‐1,200 L/m/day for many sites with upland sand aquifers. Despite a similar range of water fluxes, buffers in gently to moderately sloping landscapes with <4 m depths of sand sediments reached a 90% removal efficiency within 30‐60 m while sites with >4 m depths required a 150‐200 m width. The width for 90% efficiency in buffers with loamy sand and sandy loam sediments also increased from 10‐20 m with <4 m sediment depths to 50‐100 m for >4 m depths. Limited data for buffers with fine‐textured sediments suggest that 90% of the nitrate flux was often depleted in a 10‐20 m width. Groundwater flux did not have a significant relationship with nitrate removal percent per meter buffer width because of the variation in efficiency that occurred in buffers with similar fluxes in different hydrogeologic settings.     

Morphology Characteristics of Southern Appalachian Wilderness Streams1

Zink, Jason M., Gregory D. Jennings, and G. Alexander Price, 2012. Morphology Characteristics of Southern Appalachian Wilderness Streams. Journal of the American Water Resources Association (JAWRA) 48(4): 762‐773. DOI: 10.1111/j.1752‐1688.2012.00647.x Abstract: Watersheds without urbanization or impacts from logging are rare in the southern Appalachian Mountains. The Joyce Kilmer/Slickrock Wilderness of North Carolina and Tennessee contains 24 km² of old‐growth forest, with the balance of the wilderness in a mature second‐growth forest. The watersheds of Little Santeetlah and Slickrock Creek are located within the wilderness. Morphological information, including channel dimensions and longitudinal profiles, was gathered from 14 alluvial stream reaches in these watersheds. The study sites had drainage areas from 0.25 to 41.6 km² and stream slopes from 0.014 to 0.104 m/m. Bankfull cross‐section dimensions of the study stream reaches were strongly correlated to drainage area across the observed range of slopes and bed morphology. Cross‐section area and width relationships for the streams in this study did not differ significantly from regional curves for the mountain physiographic region of North Carolina. Observations of these reaches did not suggest a definitive rule regarding the proportion of steps and riffles in streams. Pools occupied greater than 50% of the length in all stream reaches with slopes less than 0.07 m/m. Significant correlation existed between step height ratio and slope, suggesting that step height can be approximated as the product of channel width and slope. Riffle length and riffle slope ratios were also significantly correlated with slope, though pool spacing was not.     

Sources of Suspended-Sediment Flux in Streams of the Chesapeake Bay Watershed: A Regional Application of the SPARROW Model1

Brakebill, John W., Scott W. Ator, and Gregory E. Schwarz, 2010. Sources of Suspended-Sediment Flux in Streams of the Chesapeake Bay Watershed: A Regional Application of the SPARROW Model. Journal of the American Water Resources Association (JAWRA) 46(4): 757-776. DOI: 10.1111/j.1752-1688.2010.00450.x Abstract: We describe the sources and transport of fluvial suspended sediment in nontidal streams of the Chesapeake Bay watershed and vicinity. We applied SPAtially Referenced Regressions on Watershed attributes, which spatially correlates estimated mean annual flux of suspended sediment in nontidal streams with sources of suspended sediment and transport factors. According to our model, urban development generates on average the greatest amount of suspended sediment per unit area (3,928 Mg/km²/year), although agriculture is much more widespread and is the greatest overall source of suspended sediment (57 Mg/km²/year). Factors affecting sediment transport from uplands to streams include mean basin slope, reservoirs, physiography, and soil permeability. On average, 59% of upland suspended sediment generated is temporarily stored along large rivers draining the Coastal Plain or in reservoirs throughout the watershed. Applying erosion and sediment controls from agriculture and urban development in areas of the northern Piedmont close to the upper Bay, where the combined effects of watershed characteristics on sediment transport have the greatest influence may be most helpful in mitigating sedimentation in the bay and its tributaries. Stream restoration efforts addressing floodplain and bank stabilization and incision may be more effective in smaller, headwater streams outside of the Coastal Plain.     

Predicting the Planform Configuration of the Braided Toklat River,Alaska, With a Suite of Rule‐Based Models1

Abstract: An ensemble of rule‐based models was constructed to assess possible future braided river planform configurations for the Toklat River in Denali National Park and Preserve, Alaska. This approach combined an analysis of large‐scale influences on stability with several reduced‐complexity models to produce the predictions at a practical level for managers concerned about the persistence of bank erosion while acknowledging the great uncertainty in any landscape prediction. First, a model of confluence angles reproduced observed angles of a major confluence, but showed limited susceptibility to a major rearrangement of the channel planform downstream. Second, a probabilistic map of channel locations was created with a two‐parameter channel avulsion model. The predicted channel belt location was concentrated in the same area as the current channel belt. Finally, a suite of valley‐scale channel and braid plain characteristics were extracted from a light detection and ranging (LiDAR)‐derived surface. The characteristics demonstrated large‐scale stabilizing topographic influences on channel planform. The combination of independent analyses increased confidence in the conclusion that the Toklat River braided planform is a dynamically stable system due to large and persistent valley‐scale influences, and that a range of avulsive perturbations are likely to result in a relatively unchanged planform configuration in the short term.     

The Impact of Asynchronicity on Event‐Flow Estimation in Basin‐Scale Hydrologic Model Calibration1

Abstract: The calibration of basin‐scale hydrologic models consists of adjusting parameters such that simulated values closely match observed values. However, due to inevitable inaccuracies in models and model inputs, simulated response hydrographs for multiyear calibrations will not be perfectly synchronized with observed response hydrographs at the daily time step. An analytically derived formula suggests that when timing errors are significant, traditional calibration approaches may generally underestimate the total event‐flow volume. An event‐adaptive time series is developed and incorporated into the Nash‐Sutcliffe Efficiency objective function to diagnose the potential impact of event‐flow synchronization errors. Test sites are the 50 km² Subwatershed I of the Little River Experimental Watershed (LREWswI) in southeastern Georgia, and the 610 km² Little Washita River Experimental Watershed (LWREW) in southwestern Oklahoma, with the Soil and Water Assessment Tool used as the hydrologic model. Results suggest that simulated surface runoff generation is 55% less for LREWswI when the daily time series is used compared with when the event‐adaptive technique is used. Event‐flow generation may also be underestimated for LWREW, but to a lesser extent than it may be for LREWswI, due to a larger portion of the event flow being lateral flow.     

POST‐HARVEST RIPARIAN BUFFER RESPONSE: IMPLICATIONS FOR WOOD RECRUITMENT MODELING AND BUFFER DESIGN1

Despite the importance of riparian buffers in providing aquatic functions to forested streams, few studies have sought to capture key differences in ecological and geomorphic processes between buffered sites and forested conditions. This study examines post‐harvest buffer conditions from 20 randomly selected harvest sites within a managed tree farm in the Cascade Mountains of western Washington. Post‐harvest wind derived treefall rates in buffers up to three years post‐harvest averaged 268 trees/km/year, 26 times greater than competition‐induced mortality rate estimates. Treefall rates and stem breakage were strongly tied to tree species and relatively unaffected by stream direction. Observed treefall direction is strongly biased toward the channel, irrespective of channel or buffer orientation. Fall direction bias can deliver significantly more wood recruitment relative to randomly directed treefall, suggesting that models that utilize the random fall assumption will significantly underpredict recruitment. A simple estimate of post‐harvest wood recruitment from buffers can be obtained from species specific treefall and breakage rates, combined with bias corrected recruitment probability as a function of source distance from the channel. Post‐harvest wind effects may reduce the standing density of trees enough to significantly reduce or eliminate competition mortality and thus indirectly alter bank erosion rates, resulting in substantially different wood recruitment dynamics from buffers as compared to unmanaged forests.     

Extent and Channel Morphology of Unmapped Headwater Stream Segments of the Quabbin Watershed,Massachusetts1

Brooks, Robert T. and Elizabeth A. Colburn, 2011. Extent and Channel Morphology of Unmapped Headwater Stream Segments of the Quabbin Watershed, Massachusetts. Journal of the American Water Resources Association (JAWRA) 47(1):158‐168. DOI: 10.1111/j.1752‐1688.2010.00499.x Abstract: Effective regulatory protection and management of headwater resources depend on consistent and accurate identification and delineation of stream occurrence. Published maps and digital resources fail to represent the true occurrence and extent of headwater streams. This study assessed the accuracy of mapped origins of “blue‐line” streams depicted on U.S. Geological Survey topographic maps, and, if present, the morphological characteristics of unmapped stream segments. We identified 170 mapped stream origins on the Quabbin Reservoir watershed, Massachusetts. Of 30 mapped stream origins, we identified and examined 26 unmapped stream segments above 25, with an average length of 502 m. Twenty unmapped tributaries occurred on 10 of the 26 unmapped segments, with an average length of 127 m. Wetland reaches occurred more frequently and were larger on unmapped than on mapped stream segments. A significant and complex stream network occurs above most mapped stream origins. For the Quabbin watershed, we estimate that there are 85.8 km of unmapped stream upgradient of 314.5 km of mapped streams. Reliance on mapped stream networks for regulatory standards allows for the potential disturbance or even destruction of the unmapped stream resources. Jurisdictional regulations and guidelines should be revised so that the occurrence of streams should require field validation.     

Improved Bankfull Channel Geometry Prediction Using Two‐Year Return‐Period Discharge1

He, Laien and Gregory V. Wilkerson, 2011. Improved Bankfull Channel Geometry Prediction Using Two‐Year Return‐Period Discharge. Journal of the American Water Resources Association (JAWRA) 47(6):1298–1316. DOI: 10.1111/j.1752‐1688.2011.00567.x Abstract:  Bankfull discharge (Q_bf) and bankfull channel geometry (i.e., width, W_bf; mean depth, D_bf; and cross‐section area, A_bf) are important design parameters in stream restoration, habitat creation, mined land reclamation, and related projects. The selection of values for these parameters is facilitated by regional curves (regression models in which Q_bf, W_bf, D_bf, and A_bf are predicted as a function of drainage area, A_da). This paper explores the potential for the two‐year return‐period discharge (Q₂) to improve predictions of W_bf, D_bf, and A_bf. Improved predictions are expected because Q₂ estimates integrate the effects of basin drainage area, climate, and geology. For conducting this study, 29 datasets (each representing one hydrologic region) spanning 14 states in the United States were analyzed. We assessed the utility of using Q₂ by comparing statistical measures of regression model performance (e.g., coefficient of determination and Akaike’s information criterion). Compared to using A_da, Q₂ is shown to be a “clearly superior” predictor of W_bf, D_bf, and A_bf, respectively, for 21, 13, and 25% of the datasets. By contrast, A_da yielded a clearly superior model for predicting W_bf, D_bf, and A_bf, respectively, for 0, 0, and 14% of the datasets. Our conclusion is that it alongside with developing conventional regional curves using A_da it is prudent to develop regional curves that use Q₂ as an independent variable because in some cases the resulting model will be superior.