Assessing demography, habitat use, and flow regime effects on spawning migrations of Blue Sucker in the lower Colorado River, Texas
Blue Sucker Cycleptus elongatus is widely distributed throughout the United States. Even though it was once a commercially-harvested species, Blue Sucker is currently listed in 19 of 23 states as a species of greatest conservation need, threatened, presumed extirpated, or endangered. They are associated with riffle-run habitat where flows are generally swifter, rendering them vulnerable to alterations to the natural flow regime. Within Texas they are known to occur in three major watersheds; the Red River, Neches-Sabine, and the Colorado River. The population in the lower Colorado River therefore likely represents the southwestern extent of the species range, as the next nearest neighbor occurs in the Sabine-Neches Rivers approximately 400 km east. The lower Colorado River between Austin, Texas and Garwood Dam near Altair, Texas is the only known population of Blue Sucker in the Colorado River and has been presumed to suffer from a lack of recruitment for the last several years. Unfortunately, the lower Colorado River has experienced drastic changes in hydrology with a significant increase in extreme low flow (< 4.53 m3s-1 or 160 ft3s-1) frequency and decrease in small flood pulse (> 685.3 m3s-1 or 24,201 ft3s-1, and < 2874 m3s-1 or 101,494 ft3s-1) frequency following construction of the highland lakes in the late 1930s and 1940s. These alterations and others may be responsible for the current status of Blue Sucker in the lower Colorado River. This grant served as the primary funding support for the Doctoral Dissertation of Matthew Acre and his attached dissertation serves as the bulk of the final report for this project. Relevant results for each primary objective are briefly summarized below with further details provided in the attached dissertation. Acre, Matthew R. 2019. Assessing demography, habitat use, and flow regime effects on spawning migrations of Blue Sucker in the lower Colorado River, Texas. PhD Dissertation. Texas Tech University, Lubbock. (Attachment 1) Methods Habitat use, selection, and movement patterns, including likely spawning migrations were analyzed from data collected on 42 adult Blue Sucker. Fish were tagged in December 2014 and December 2015 with a combined acoustic radio transmitter (CART) tag at Utley (n = 7), Bastrop (n = 12), La Grange (n = 12), Columbus (n = 1), and Altair (n = 10). Fish were then monitored from January 2015-May 2017 which resulted in 1,157 detections during 38 attempts to determine fish locations. These data were used to determine the effects of streamflow, instream temperature, and habitat availability on movements which included spawning migrations. River discharge data was downloaded from USGS gage stations at Austin, Bastrop, Smithville, La Grange, and Wharton. Instream temperature data was collected from nine iButton data loggers mounted throughout the river. Discharge data was assigned to fish depending on the nearest gage station. Submersible ultrasonic receivers were also mounted throughout the river to passively monitor tagged fish. A substrate availability map was created of the 290-rkm study area using side-scan sonar data. The substrate map was used to determine season habitat selection and avoidance, and to determine if locations within the river being used at higher rate could be explained by habitat. To assess population size and the viability of the population to persist into the future, 30 mark-recapture sites were identified and sampled six times over two years (2016-2017) with standard boat-mounted electrofishing techniques. The 30 sites were split in riffle, run, and pool habitat and sampled separately which resulted in 540 samples over two years and 97 hours of electrofishing. This sampling effort resulted in 152 captures of which 15 were recaptures. These data were used to estimate population size and mesohabitat use. When fish were captured a fin ray clip was taken to be used in age estimation analysis. The age estimation data was used to determine back-calculated hatch years which informed the estimated yearly cohort strength. Discharge data was downloaded from the USGS gage station in Bastrop, Texas and temperature data was downloaded from the NOAA Camp Mabry station. These data were then used to determine the effects of abiotic variables on yearly cohort strength. Objectives and Primary Findings Describe movement and habitat use under various experimental releases via radio and acoustic telemetry Greater duration and magnitude of high flows coupled with warmer temperatures likely alter seasonal movement patters and may hinder spawning migrations. Additionally, fish that were in riffle dense areas were less likely to make large movements and more likely to remain in that area. Seasonal movement patterns in the first year of tracking (2015) were typical compared to other Blue Sucker populations across the country. In the winter, most movements were upstream and the mean distance between movements, regardless of direction, was approximately 35 rkm during this time. In the following spring, most movements were downstream, and fish had a mean of 55 rkm between detections. During the summer and fall of 2015 movements were significantly reduced. The next year the pattern begins much the same, but flooding and subsequent duration of high flows during the spring of 2016 may have altered the spawning patterns observed during 2015. In 2017, there was an almost complete cessation of movement that may be partly attributed to increased magnitude and duration of river discharge as 2017 was an exceptionally wet year. River discharge was high leading up to Hurricane Harvey, which brought record high flows below La Grange, Texas in 2017. Throughout this study, fish that were originally tagged in Bastrop, Texas were 2.5x more likely to remain in that location while those tagged both upstream and downstream were more likely to move, particularly during the spawning season. This result further supports the importance of the approximately 15-rkm (9-mi) reach surrounding Bastrop. The timing and to some degree, the magnitude of movements from the manual tracking are supported by the data collected from SUR units. However, SUR data made clear that movement estimates were underestimated as we were missing some movements into tributaries where we did not manually track, particularly in 2015. Additional detail is contained in Chapter 3 (Habitat Use and Selection) of the attached dissertation (Attachment 1).
Determine the timing and extent of use of spawning habitats in relation to streamflow and temperature Migrations to spawning habitat were best predicted by cooler temperatures and discharge between 283-2296 ft3s-1. Variable flows best predicted spawning movements. Additionally, spawning habitat is likely limited to about 34 miles of river near Bastrop, Smithville, and La Grange. Blue Sucker selected boulder habitats more frequently during winter and spring, likely associated with spawning due to this timing. Additionally, Blue Sucker selected cobble and bedrock habitat throughout the year and actively avoided sandy substrates, as well as areas with increased anthropogenic items which included waste material. Spawning movements, which included migrations, were most likely to occur when temperatures were between 13.5-18°C (56-64°F) and river discharge was between 8-65 m3s-1 (283-2296 ft3s-1). When a spawning movement was detected, there was 78% probability the fish moved to riffle or run habitat. Optimized hotspot analysis identified three locations within the lower Colorado River that are likely integral reaches of river for Blue Sucker. These three locations combined represent 55 rkm (34 mi) of the 290-rkm (180-mi) study area and are located around Bastrop, Smithville, and La Grange, Texas. The Bastrop reach was also identified as integral spawning habitat in addition to daily activities. This is further supported by acoustic data which detected 60% of the tagged population utilizing this area and is representative of at least one individual from each tagging location except Columbus where only a single female was tagged. In general, males had a larger linear home range than females. Home ranges increased in size as the percentage of sand and anthropogenic materials increased within their core area. This is an indication that Blue Sucker are moving to specific habitats to complete aspects of their life history and must move further if located in an area with fewer riffle/run habitats available. Additional detail is contained in Chapter 4 (River Discharge, Temperature, and Habitat Effects on Blue Sucker Movements) of the attached dissertation (Attachment 1).
Estimate population size and demographic characteristics
The Blue Sucker population in the lower Colorado River is relatively small and seems to be recruitment-limited. The estimated population size from these data was 679 individuals with a lower boundary (95% confidence interval) of 449 and upper boundary of 1089 individuals. The population has an estimated annual mortality of 15%, therefore, 85% of the population survives annually. However, recruitment in the population has been relatively low or completely lacking from 2009-2017. Multiple gear types sampled from 2015-2017 aimed at capturing young-of-year and sub-adults resulted in zero Blue Sucker captures. In the event recruitment of young individuals back into the adult population does not increase, there is a high probability Blue Sucker will be extirpated from the lower Colorado River. The best models informing what results in a strong recruitment class suggests that greater river discharge in September, lower September and July minimum temperatures, and an increase in high flow pulse frequency best predict strong recruitment years.
More details can be found in Chapter 2 (Blue Sucker Demographics and probability of persistence in the lower Colorado River, Texas) of the attached dissertation (Attachment 1)
Evaluate the effects of streamflow on recruitment and age-0 habitat use. The Blue Sucker population within the lower Colorado River seems to be suffering from reduced recruitment. Limited habitat availability, including spawning grounds, and potentially alteration of environmental cues for migration and spawning may act to limit recruitment. Altered spawning cues may be linked to increased river discharge and an increase in river temperatures, while also correlated with increased duration of high flows. While 2015 had the highest probability of spawning migration and was associated with lower flows than 2016 and 2017, those low flows occurred on the descending leg of flood pulse which were estimated to be approximately 206 m3s-1 (7,274 ft3s-1). The model was built from fish detection that were identified as likely spawning movements or migrations based on peer-reviewed literature. The results from this study identify a lack of flood pulses, increased duration of flood pulses that do occur, an increase in the magnitude of flood pulses, and increased temperatures as likely culprits in the current threatened status of the species within the study area. In addition to alterations of the flow regime from historic standards, there has undoubtedly been a decrease in available habitat post dam construction. The results from this study identify 55 rkm (34 mi) with the 290 rkm (180 mi) between Austin and Altair, Texas (Garwood Dam) that is used at higher rate than the rest of the river. This suggests that habitat may be an additional limiting factor for Blue Sucker in the lower Colorado River. Further detail can be found in Chapter 2 (Blue Sucker Demographics and probability of persistence in the lower Colorado River, Texas) of the attached dissertation (Attachment 1) Recommendations: Continue monitoring population trends of Blue Sucker population in lower Colorado River The observed and estimated recruitment from 2009-2017, which is lower than previous years, is a troubling trend. The results from this study suggest that the population may be in trouble and could be declining. The population viability analysis indicated that if he reduced recruitment continues the population may be extirpated from the river. However, without a continuation of the mark-recapture work it will be difficult to determine how the population is responding to various environmental conditions.
Consider modification timing/duration of flow pulses for Blue Sucker spawning Migration probabilities related to spawning were most likely to occur in January-March and were best predicted by variable flows on the descending leg of flood pulses. River discharge between 283-2296 ft3s-1 preceded by small flood pulse estimated at 7,274 ft3s-1 were estimated from fish detections assumed to be for spawning. A component that may be missing from the current water management plan is variation in flow regime during spawning windows.
Consider adjustment of summer minimum flows to benefit YOY Blue Sucker Estimated yearly cohort strength (recruitment) was best predicted by increased September low flows, increased high pulse frequency (as defined by IHA see Indicators of hydrologic alteration, version 7.1 user’s manual, The Nature Conservancy, 2009), decreased September and July minimum temperatures. This may be an indication that variable hydrology throughout the spawning and the months following hatch date are vital to survival for young-of year Blue Sucker. Start a genetic and early life history study on the population Determine current levels of bottlenecking, genetic uniqueness within the population, and effective population size. A possible explanation for the reduced recruitment throughout this study may be linked to genetics as opposed to the effects of a small population (N = 679). The other possibility for the reduced recruitment may be linked to some aspect of the larval stage. An early life history study focused on what conditions are optimal for successful hatching may illuminate issues with the current habitat and substrate availability, and current flow regime. Dependent on the genetic results actions may include: Modify habitat in the river to increase spawning habitat which results in successful hatching. Modify the water management plan to increase success of early life history stages. Start a captive propagation program to supplement the lack of recruitment. If this action is taken the genetics will inform where broodstock should be gathered and pairings to maximize genetic diversity.