Refinement and Validation of Habitat Quality Indices and Aquatic Life Use Indices for Application to Assessment and Monitoring of Texas Surface Waters

Kirk Winemiller, Ryan King (Baylor U.), Mark Fisher (TCEQ)

To support monitoring and setting standards for surface water quality, the Texas Commission on Environmental Quality (TCEQ) has adopted Aquatic Life Use Standards (ALUS) that rely on indices of biotic integrity (IBIs) and habitat quality indices (HQIs).  These essential tools allow natural resource managers and regulators to assess the status of ecological systems for evaluation of trends and compliance with established water quality standards.  Following several years of research by the Texas Parks and Wildlife Department (TPWD), the fish IBI used to assess Texas streams underwent major revision to reflect more accurately the major faunal differences among the state’s diverse geologic, climatic, and faunal regions (Linam, G.W., L.J. Kleinsasser, and K.B. Mayes. 2002. Regionalization of the Index of Biotic Integrity of Texas Streams. TPWD River Studies Report No. 17, TPWD, Austin, TX, 26 p).  The fish IBI has been researched more extensively than the HQI, therefore the latter index requires immediate study.  The issue of scale and resolution, both in terms of space (geography, watershed position, siting within stream reach) and time (seasons, hydroperiod, time elapsed since last major disturbance) must be examined quantitatively in order to evaluate the validity and reliability of these assessment tools.  This project examines treams with perennial surface water within the Brazos and Trinity river basins within the Texas-Central Oklahoma Plains, Texas Blackland Prairies, and East Central Texas Plains ecoregions.  Survey locations provide broad geographic coverage, a range of landscape features (including land use), and representation of a range of stream habitat conditions.  Annual surveys will describe and evaluate key metrics of stream habitat in relation to position within stream reach, watershed, and landscape (e.g., geology/soils, topography, land use) as well as fish assemblage structure.  Field surveys will be conducted between April 1 and Oct. 1 in 2006 and 2007.  The project will determine principal relationships between stream habitat features and sources of landscape variation, including indicators of watershed degradation.  The composition and structure of fish assemblages for use as biotic indicators of ecological status will be determined concurrent with stream habitat characterizations.  Statistical analysis of fish and habitat metrics will be conducted to determine their sensitivity to aquatic life uses (ALUs) as well as variation in temporal and spatial scales.  This project aims to determine the most useful (sensitive, reliable) metrics (individual elements, suites of elements, or aggregate variables derived from multiple elements) for creation of improved HQI and IBI for wadeable streams in the Subhumid Agricultural Plains (SAP) aggregate ecoregion.  

Demographic and life history responses of fishes to hydrology and disturbance in floodplain and channel habitats of the Brazos River, Texas

Steven C. Zeug and Kirk O. Winemiller

Several conceptual models predict fish population responses to environmental drivers such as hydrology however, few studies have explicitly tested hypotheses drawn from the predictions of these models (but see King et al. 2003).  The Flood Pulse Concept predicts that fish recruitment is enhanced by connectivity between floodplain and channel habitats provided that floods occur during periods of fish reproduction when water temperatures are warm.  The Low Flow Recruitment Hypothesis states that in systems with unpredictable flood dynamics, recruitment is optimized during periods of low flow when channel and floodplain habitats are isolated but flows are relatively stable and food resources are more abundant. Here we test the hypothesis that fish life history strategy determines the hydrologic regime under which recruitment is maximized, and that life history traits within species will respond to the hydrologic regime of each habitat.  Population demographics and a suite of life history attributes for seven species representing three divergent life history strategies are measured monthly in the Brazos River channel and two oxbow lakes with different connection frequencies to determine how these populations are influenced by environmental drivers (floods, drought).  The results produced by this study will provide a new model of fish recruitment in river-floodplain systems that incorporates spatial heterogeneity and flow variability, and is supported by empirical data.  In the context of instream flow allocation; these data can be used both to make predictions about which species may be affected by a proposed flow regime, and to design flow regimes that support the diversity of life history strategies that exist in these systems.

Association Between Brush Cover and Stream Fish Assemblages in the Pedernales River Basin, Texas

Jenny S. Birnbaum and Kirk O. Winemiller     

One strategy for increasing the water supply in semi-arid rangelands is management of encroaching brush.  Removal of deep-rooted woody species, such as mesquite (Prosopis glandulosa) and juniper (Juniperus ashei), is believed to increase groundwater recharge of streams in areas where annual precipitation exceeds 450 mm.  In fact, the state of Texas is currently spending millions of dollars to subsidize brush management with the goal of increased water yields.  However, scientific examination of this claim is scarce, and studies have yielded conflicting results.  Effects of brush management on water yields are likely to be subtle and localized in accordance with geology, topography, and other landscape features.  This interdisciplinary study incorporates aspects of GIS, hydrogeology, herpetology, riparian vegetation ecology, and stream ecology to explore the issue of juniper management in headwater tributaries of the Pedernales River of central Texas.  We are surveying biota and abiotic characteristics, including fish, benthic macroinvertebrates, water depth, flow, pH, DO, substrate composition, and other physicochemical variables in order to understand how juniper management influences stream fish and benthic macroinvertebrate assemblages.  Additionally, we plan to address fish-macroinvertebrate relationships and macroinvertebrate-microhabitat relationships in on-going work.

Seasonal Variation in Food Web Composition and Structure in a Temperate Tidal Estuary, Mad Island Marsh, Texas

Kirk Winemiller, Senol Akin & Steve Zeug

Seasonal variation in aquatic food web structure at Mad Island Marsh, Matagorda Bay, Texas was examined using dietary information obtained from the analysis of gut contents from large samples of fish and crustacean specimens.  Unique aspects of this study include the use of large samples of consumer gut contents (N= 6,452), long-term sampling (bimonthly surveys over 18 mo), and standard methods of data collection and analysis facilitating comparisons with other aquatic food webs.  Dietary data were partitioned for analysis into summer and winter seasons.  Most consumers fed low in the food web, with trophic levels ranging from about 2 to 3.5 during both summer and winter.  Vegetative detritus was more important in macroconsumer diets than live algae and macrophytes.  Low trophic levels of consumers reflected the important role of abundant detritivores (e.g., striped mullet, Gulf menhaden, and macroinvertebrates) in linking detritus to top predators via short food chains. 

Most food web properties revealed comparatively little seasonal variation.  The summer food web had more nodes (86), more links (562), a higher density of links as indicated by connectance (0.08), and a slightly higher predator/prey ratio (0.51) compared to the winter food web (75 nodes, 394 links, connectance = 0.07, pred./prey ratio= 0.47).  Proportions of top (0.06-0.07) and intermediate (0.75-0.76), and basal (0.19) species did not vary significantly between seasons, but mean trophic level was higher during summer.  Addition of feeding links based on information from the literature increased connectance to 0.13 during the both seasons, however other web parameters had values similar to those obtained for our directly estimated food webs.  Findings were consistent with earlier studies identifying detritus a major pathway from production sources to estuarine consumers.  Seasonal variation in food web structure was influenced by changes in community composition (e.g. influxes of postlarval estuarine-dependent marine fishes during winter), availability of resources (e.g. more submerged macrophytes amphipods during summer), and size structure and ontogenetic diet shifts of dominant consumer taxa. 

Food web structure and major sources of primary production consumed by metafauna of Mad Island Marsh also were compared using stable isotopes and dietary analysis.  Isotopic results revealed two interlinked subwebs– one with algae as a primary production source, and the other with C₄ saltmarsh grasses as a production source, with the latter apparently consumed in the form of detritus and associated heterotrophic bacteria.  Both methods indicate that terrestrial C₃ plants probably contribute relatively little material to the aquatic food web.  The two subwebs are spatially segregated, with most larger fishes and invertebrates associated with deeper areas and algal production.  Isotopic analysis could not reveal the detailed structure of predator-prey interactions at the species level; greater detail of trophic pathways was revealed by the dietary analysis.  Estimates of vertical web structure (species trophic levels) by the two methods were largely concordant.  The exceptions were zooplanktivorous and detritivorous fish species that had higher trophic levels according to nitrogen isotope ratios.  The isotope method more accurately indexed the number of trophic transfers than the dietary method that depends on accurate dietary estimation for all elements of food chains leading to a consumer, and which assumes equal assimilation efficiencies for elements found in stomach contents. 

Resource Use Among Sympatric Lepisosteidae Species in the Middle Brazos River and Associated Oxbows

Clint R. Robertson, Steve C. Zeug and Kirk O. Winemiller

This project examines habitat and diet partitioning among three sympatric gar species (Lepisosteus osseus, L. oculatus and Atractosteus spatula)  in the middle Brazos River and associated oxbow lakes.  We are comparing species relative abundances among the river channel (lotic habitat) and two oxbows that form connections with the river channel at different water levels (lentic habitats).  Fishes are collected using experimental gill nets and standardized beach seine hauls.  Preliminary results indicate that L. osseus is more abundant in channel habitats, with few collected in oxbow habitats.  L. oculatus is more abundant in oxbow habitats, with few collected in channel habitats.  A. spatula was far less frequently collected than the other two species, with one individual caught in the channel and another in an oxbow.  Preliminary diet analysis (volumetric stomach contents analysis) indicates high variability among channel and oxbow habitats, with the diet of L. osseus in the channel dominated by fishes of the families Ictaluridae and Cyprinidae, while the diet of L. oculatus in oxbows is dominated by fishes of the families Centrarchidae, Cyprinidae and Clupeidae.

Resource partitioning among piscivorous fishes of the middle Brazos River, Texas

Steve C. Zeug, Clint R. Robertson and Kirk O. Winemiller

Previous research has shown that piscivory may control the ability of some species to colonize floodplain habitats.  In response to these findings, and in conjunction with other on-going projects in the middle Brazos River-floodplain system, we are examining gut contents of all piscivorous fishes collected during monthly sampling in the main river channel and two oxbows that form connections to the channel at different water levels for a period of one year.  This study should allow us to examine the influence of piscivory on colonization rate in these habitats.

Local vs. regional influences on the structure of fish assemblages in Texas streams

David J. Hoeinghaus, Kirk O. Winemiller and Jenny S. Birnbaum

We examined fish assemblage organization in warm-water streams across Texas using a database of 157 assemblages representing six major river basins.  A multi-scale approach was used to assess patterns in assemblage structure and saturation, similarity of species pools, species co-occurrences and environmental correlations.  The series of analyses was performed using functional groups based on trophic and life-history characteristics as well as the original taxonomic dataset.  Results from our taxonomic analyses revealed distributional patterns resulting from large-scale or historical processes.  Regional environmental factors such as climate extremes and landscape characteristics dominated stream assemblage structure accounting for 67% of the explained variation, whereas local-scale factors explained within sub-basin differences.  Regional species pools differed significantly among sub-basins, and assemblages were saturated.  Most strong negative species associations were observed between species of the same trophic group but different life-history groups, or for predator and prey pairs.  In contrast, analyses using functional groups revealed an approximately equal influence of local and regional-scale factors.  Functional diversity did not follow a geographic trend as observed for the taxonomic dataset.  Our results indicate that addition of functional types to a local community may be possible without reaching a local ceiling, even while taxonomic species saturation may be occurring.  Co-occurrence patterns of functional groups suggest that predation is an important structuring component in these systems, but competition probably has little influence.  Results from our taxonomic analyses support general hierarchical filter-type models of community assembly, whereas our functional analyses suggest models that differentiate between habitats based on environmental stability may be more appropriate.  We suggest that analyses using various kinds of functional groups can more directly infer ecological responses to environmental variation, and may provide a more fruitful avenue for developing and testing ecological theory of community organization. 

Response of Oxbow Lake Biota to Hydrologic Exchanges with the Brazos River Channel

Kirk Winemiller, Tim Bonner (Texas State U.), Steve Zeug, Casey Williams (Texas State U.), Ray Mathews (Texas Water Devel. Bd.), Tim Osting (Texas Water Devel. Bd.) & Fran Gelwick

Fishes and aquatic habitat variables were sampled between June 2003 and September 2004 to obtain information on the ecological dynamics associated with river channel–oxbow lake connectivity in relation to instream flows.  The ecological study complemented a concurrent research effort undertaken by the Texas Water Development Board to document geomorphological and hydrological features that determine degrees of oxbow to channel connectivity.  The ecological study also examined fish population structure and dynamics at two river channel sites in the lower Brazos River upstream and downstream of the site selected for the Allen’s Creek reservoir.  Standardized fish samples were collected using seines and gillnets, with data analyzed separately as catch per unit effort.  Statistical ordination techniques revealed a strong gradient of fish assemblage structure that contrasted oxbow samples from river channel samples.  A secondary gradient was associated with seasonal variation in oxbow lakes.  In contrast to the river channel, oxbow lakes contained high densities of white crappie (Pomoxis annularis), sunfishes (Lepomis spp.), and shads (Dorossoma spp.).  A number of minnow species (e.g., Hybognathus nuchalis, Macrhybopsis hyostoma) appear to be fluvial specialists that always or almost always were collected from the river channel.  Several of these fluvial specialists were more abundant one to two months after periods of peak flow.  For species common in oxbow lakes, density tended to decline following periods of peak flow, which indicates a net export of individuals from oxbows to the river channel during floods that connect these habitats.  Consistent with this view were patterns of higher densities of these species in the river channel following periods of peak flow.  Fluvial specialists appeared in oxbow lakes in low to moderate numbers during periods of peak flow, but these sub-populations generally did not persist more than a month or two.  Densities of phytoplankton, zooplankton, and fish were much higher in oxbow lakes than in the river channel, and more so following prolonged periods of isolation.  Oxbow lakes that were formed more recently and that are located closer to the river channel had lower “control points” in the natural levee, and as a result flooded at lower discharge levels.  It is concluded that oxbow lakes of variable ages and geomorphological structures provide essential habitats that function to increase overall fish species diversity in the lower Brazos River.

Status of Dionda diaboli and established populations of exotic species in Lower San Felipe Creek, Val Verde County, Texas

Hernán López-Fernández and Kirk O. Winemiller

Sampling from April 2001 to March 2003 revealed San Felipe Creek still supports a breeding population of the threatened Devils River minnow (Dionda diaboli).  The species was restricted to creek habitats and was not found in the outflow channels of San Felipe Springs.  We determined that breeding populations of the introduced tropical fish species Hypostomus sp. (a South American armored catfish) and Oreochromis aureus (an African cichlid) are present in San Felipe Creek as well.  Some evidence suggests that the presence of exotic species, particularly Hypostomus sp., might have a negative effect on the Devils River minnow.  We recommend periodic monitoring of fish populations in San Felipe Creek to document future changes in the population of Devils River minnow and other endemic species, and to facilitate design and implementation of conservation plans in San Felipe Creek.

Effects of hydrologic connectivity on food web structure in the Brazos river-floodplain system

Steven Zeug & Kirk O. Winemiller

Off-channel floodplain habitats such as oxbow lakes are recognized as important sources of biological productivity in river-floodplain systems.  Because the flow regimes of these systems are temporally variable, the influence of river-floodplain connectivity on trophic structure is expected to be pulsed.   In this study we examine the influence of hydrologic connectivity on trophic structure in the Brazos River channel and two oxbow lakes with different connection frequencies using stable isotopes of carbon and nitrogen.  Tissue samples from species representing several trophic guilds were collected monthly for a period of one year.  During the study period, each habitat experienced different levels of hydrologic connectivity that may influence trophic structure by transporting organic material, inorganic nutrients and facilitating faunal exchange among habitats.

Big Bend Oxbow,

Brazos River Floodplain

< before flood

during flood >

Habitat attributes influence ontogenetic diet shits of gizzard shad (Dorosoma cepedianum) in a large floodplain river

Zeug, S.C., D. Peretti & K.O. Winemiller

Gizzard shad (Dorosoma cepedianum) are recognized as an important trophic link in many aquatic food webs because of their ability to link detrital carbon sources with higher trophic levels (e.g. piscivorous fishes).  Most research on the feeding ecology of gizzard shad concerns populations stocked in reservoirs whereas populations in more dynamic habitats (river-floodplain systems) have been neglected.  Here we examine ontogenetic diet shifts in gizzard shad inhabiting oxbow lakes and channel habitats of the Brazos River, Texas.  Using stable isotopes of nitrogen and stomach contents we found that oxbow populations show an increase in the consumption and assimilation of primary consumers (zooplankton) during ontogeny resulting in elevated trophic positions despite high diet overlap values calculated using volumetric proportions of stomach contents.  Differences in d¹⁵N values indicate that shad greater than 200 mm standard length have a trophic position approximately one level higher than shad less than 200 mm.  Populations in the river channel maintain similar trophic positions during ontogeny indicative of detritivory.  These results challenge previous models of ontogenetic diet shifts in gizzard shad and similar to other dietary studies using multiple methodologies, suggest that material assimilated may not be proportional to the volume of material consumed.

Updated Wednesday March 01, 2006