Water-Level Fluctuations and Water Temperature Effects on Young-of-Year Largemouth Bass in a Southwest Irrigation Reservoir

Water management in the arid southwestern United States balances irrigation and municipal demands with spring flows. These demands have competing interests that a limited water supply due to a changing climate is likely to exacerbate. High emission scenarios project annual temperatures could increase throughout the southwestern United States by 5.6°C and precipitation could decrease by as much as 12% by the end of the 21st century (Kunkel et al. 2013). Rising temperatures at the beginning of the 21st century have caused megadrought conditions throughout western North America (Williams et al. 2020). The ability to maintain a volume of water within reservoirs that minimizes evaporative losses will predicate timing and delivery of water to growing populations throughout the Southwest (Friedrich et al. 2018). To reduce evaporative loss, expediting water to consumers minimizes water retention in reservoirs. The resulting water-level fluctuations reduce the quality and quantity of littoral fish habitat (Daugherty et al. 2015).

Fluctuating water levels of reservoirs can affect littoral spawning fishes, including recreationally popular and economically important sportfish such as Largemouth Bass Micropterus salmoides (Sammons et al. 1999). Management is dependent on maintaining viable population abundance throughout high inter- and intra-annual water-level fluctuations. Spawning of Largemouth Bass had a positive relationship to years when reservoirs reached maximum volume in a southeastern U.S. reservoir in Tennessee (Maceina and Bettoli 1998) and a series of Ohio reservoirs (Garvey et al. 2011). Enhanced survival and growth of age-0 Largemouth Bass likely responded to inundation of shoreline habitat during elevated, but stable, water levels. In a tropical reservoir where water temperature is constant, water levels dictated the timing of spawning of the Largemouth Bass population (Waters and Noble 2004). While water levels are important to nesting success of Largemouth Bass, fisheries biologists consider water temperature to be an important extrinsic variable, second only to photoperiod, which affects reproductive readiness in fishes (Bennet and Gibbons 1975). As a temperate species, Largemouth Bass spawns across a range of temperatures (12–24°C; Kramer and Smith 1962; Miller and Storck 1984; Sammons et al. 1999). Peak spawning occurs between 15 and 18°C and hatching, typically within 3–4 d of spawning, depending on water temperature (Kramer and Smith 1962; Chew 1974; Heidinger 1976; Miller and Storck 1984). While native to the Mississippi River basin and the Atlantic drainage, fisheries managers have transplanted the species to the arid southwestern states (Fuller et al. 2021). However, little information is available that describes the effects water temperature and water-level fluctuations have on the timing of reproduction of Largemouth Bass in southwestern reservoirs.

A series of reservoirs constructed throughout the middle Rio Grande basin in the early 19th century serve to manage irrigation needs for agriculture, provide flood control, and generate electricity for New Mexico. Elephant Butte Reservoir represents the largest of these reservoirs and is one of the most important fishing destinations for Largemouth Bass anglers in New Mexico. Water levels in the reservoir are variable due to high evaporation rates and flows reliant on seasonal precipitation and snow pack. Daily water-level fluctuations as large as 15 cm are possible in Elephant Butte Reservoir as water managers attempt to balance evaporative rates with the timing of irrigation needs that coincides with spring spawning of Largemouth Bass. As the annual spring flows diminish, the hydrological cycle of the reservoir undergoes fluctuations detrimental to the nest-building strategies of this fish.

Water managers control the timing and duration of water release through Elephant Butte Reservoir. Water releases are in accordance with downstream irrigation needs. When inflow to the reservoir does not equal or exceed outflow, dropping water levels may be detrimental to the reproduction of littoral spawning fishes such as Largemouth Bass. However, water managers have a tentative agreement to hold water levels steady for 7–10 d to provide a window in which littoral spawning fishes, including Largemouth Bass, would have a chance to spawn. Population surveys indicate Largemouth Bass remain below the statewide relative abundance target of 20–40 fish/h (NMDGF 2016). To improve Largemouth Bass reproduction, managers could regulate water levels to coincide with spring water temperatures optimal for reproduction. Our objectives were to characterize hatch dates, relative abundance, and growth of young-of-year Largemouth Bass over two water years. If we can identify the timing and duration of hatching for Largemouth Bass, then state biologists can communicate a strategic timeframe to water managers for reservoir filling and storage that optimizes conditions for Largemouth Bass reproduction and minimizes conflict with other water user needs.

We conducted the study over 2 y from March 2017 to July 2018 at Elephant Butte Reservoir in Sierra County, New Mexico (Figure 1). The reservoir catchment is approximately 75,000 km² with a surface area averaging 38.4 km² and a total fill capacity of 2.5 km³ (U.S. Bureau of Reclamation 2017). For our study sites, we selected three coves aligned longitudinally from north to south (Kettle Top, McRae Canyon, The Jungles; Figure 1) to conduct monthly fish collections. These sites captured spatial and water temperature differences due to water inflow and outflow. In addition, these coves were on the eastern shore of the reservoir to provide protection from wind and wave action for Largemouth Bass spawning (Nack et al. 1993). Prior observations revealed these coves contained suitable spawning substrate conducive to nest building and reproduction. In addition, submerged terrestrial vegetation and coarse woody debris were present to provide cover for adult and juvenile fish.

We targeted the collection of young-of-year Largemouth Bass to coincide with the onset of nest development throughout the three study coves to retrospectively identify timing of the 2017 and 2018 spawning seasons. In 2017, we collected young-of-year Largemouth Bass biweekly 29 June–15 August (n = 229). In 2018, we collected young-of-year Largemouth Bass biweekly 30 May–8 August, though we captured fewer fish (n = 28). To estimate relative abundance, we reported catch per unit effort as the number of fish captured per hour of electrofishing (fish/h). We electrofished within each cove by walking the entire shoreline with a model 12-B backpack electrofisher (400–700 V, Smith-Root, Vancouver, WA) and we used a pontoon boat (Mini Snout, Demaree Inflatable Boats, Inc.) following the shoreline outfitted with an electrofishing device (60 DC, 20–40%, 5.2-m 5.0 GPP, Smith-Root). We used only Largemouth Bass captured via backpack electrofishing for relative abundance (n = 229 in 2017 and n = 11 in 2018; Data S1, Supplemental Material) because we captured fish opportunistically using boat electrofishing. We immediately euthanized fish in a 200 mg/L aqueous solution of tricaine methanesulfonate (MS-222). We recorded total length (± 1.0 mm) and weight (± 0.1 g) for each fish and the fish was preserved in 95% ethanol.

We modified the procedures for counting daily otolith increments from Miller and Storck (1982). The authors demonstrated greater precision in assigning ages and hatch dates when fish were approximately ≤ 100 mm in total length. Thus, we selected young-of-year fish ≤ 100 mm for 2017 and 2018 collections. From a total of 229 fish collected in 2017, we randomly selected a subset to age (n = 36; Data S2, Supplemental Material). We aged all young-of-year fish collected in 2018 (n = 28; Data S2, Supplemental Material). We removed both left and right sagittal otoliths and affixed them to a microscope slide using Crystalbond™ thermoplastic cement (Aremco, Valley Cottage, NY). We sanded the left sagittal otolith using a grinding/polishing machine (MTI Corporation UNIPOL 1210, Richmond, CA) with 2,000-grit sand paper dampened with ultrapure water and finished with a polishing paste. Two independent readers counted the daily increments for each otolith three times such that they ascertained a mean count. Counting began 0.035 mm from the center of the otolith where the first postlarval daily increment began in fish > 20 mm total length. If the three counts were not within ± 3 daily increments (or no two counts were the same), readers conducted a fourth count and selected and averaged the three counts with the least variance. We randomized otoliths such that the readers counted no individual otoliths consecutively. They did not know total length of the fish. We compared the counts of average daily otolith increments between readers and deemed counts that differed by less than 10%, or by ± 3 daily increments to be acceptable, and averaged the two counts. If disagreement greater than 10% (or ± 3 daily increments) occurred between the readers, both readers recounted daily increments to reach an agreement on which original counts were most accurate. We removed the otolith if an agreement could not be reached between the two readers. We back-calculated hatch dates using the equation h = d − a, where h is the estimated hatch date, d is the calendar day of capture, and a is the number of daily increments counted (Collection Data S2, Supplemental Material).

We calculated mean daily growth rate (millimeters per day) by dividing total length (millimeters) by age (number of days since hatch). We then used a Mann–Whitney U test to determine if mean daily growth (millimeters per day) of young-of-year Largemouth Bass differed between 2017 and 2018. We used an alpha of 0.05 for all analysis to detect statistical differences. We performed all analyses and visualizations with R 4.1.1 (R Core Team 2021) and ggplot2 (version 3.3.5; Wickham 2016) package.

We obtained water-level fluctuations (i.e., rate of change) throughout the 2-y study from reservoir elevation maintained by the U.S. Bureau of Reclamation (Data S3, Supplemental Material). We obtained water temperature from intermittency temperature loggers deployed perpendicular to the shoreline within each cove. Prior to deployment, we modified Hobo Pendant® (Model UA-002-64; Onset Computer Corporation, Bourne, MA) temperature loggers that recorded accurate hourly water temperature (± 0.53°C) and documented when drying events occurred (Chapin et al. 2014). We chose these modified loggers to capture spatial and temporal changes in reservoir volume due to filling and receding. We checked the loggers monthly and retrieved temperature data throughout the entire study. Within each cove, we set 19 loggers apart (0.5–3.0 m) and offset vertically by approximately 0.5 m. The shallowest logger was set at 1.0 m in depth to capture water temperature that would presumably reflect spawning temperatures within each study cove (Nack et al. 1993). We selected hourly water temperatures from only submerged loggers throughout each study cove. We observed that water temperatures were similar across the three study coves for each time interval; thus we combined the water temperatures across the three study coves to obtain an overall median, minimum, and maximum throughout the 2017 and 2018 hatch dates (Data S4, Supplemental Material). We used a two-sample Kolmogorov–Smirnov test to determine if the water temperature distributions differed between 2017 and 2018. We used mean hourly water temperatures within the hatch dates in each year to perform the analysis.

Mean catch per unit effort for young-of-year Largemouth Bass in 2017 was 21.7 fish/h (n = 229). In 2018, mean catch per unit effort was 6.8 fish/h (n = 11). The retrospective age analysis from otoliths revealed similar hatch dates in 2017 (14 April–29 May; n = 36) and 2018 (13 April–28 May; n = 28) for young-of-year Largemouth Bass across the three study coves. Median growth within the 2017 hatch dates was 1.02 mm/d with upper (75%) quartile of 1.19 mm/d and lower (25%) quartile of 0.94 mm/d (Figure 2). Median growth within the 2018 hatch dates was 0.82 mm/d with upper (75%) quartile of 0.93 mm/d and lower (25%) quartile of 0.77 mm/d (Figure 2). Distributions of growth for young-of-year Largemouth Bass were significantly different between 2017 (n = 36) and 2018 (n = 28; Figure 3; Mann–Whitney U = 847, P = 0.0001).

The water level rate-of-change during the 2017 hatch dates was positive or increasing (+3.1 to +13.1 cm/d) while the rate of change during the 2018 hatch dates was negative or receding (−8.5 to −0.6 cm/d; Figure 4). Median water temperature within the 2017 hatch dates was 19.0°C with upper (75%) quartile of 20.1°C and lower quartile (25%) of 18.0°C (Figure 5). The temperature range within the 2017 hatch dates was 14.3–24.4°C. Median water temperature within the 2018 hatch dates was 17.6°C with upper (75%) quartile of 19.4°C and lower (25%) quartile of 15.9°C (Figure 5). The temperature range within the 2018 hatch dates was 13.5–21.7°C. We determined water temperature within the 2017 and 2018 hatch dates had significantly different distributions (Figure 6; D = 0.4079, P = 0.0001).

The scarcity of water in the southwest United States will continue to pose a challenge to state and federal agencies tasked with the priority of meeting water demands. These demands appear in direct conflict with natural resource agencies tasked with managing New Mexico's littoral spawning fishes, including popular and economically important sportfish. Following 2017, when precipitation was above average throughout the southwestern United States, the ensuing drought in 2018 provided an opportunity to characterize the effects of variable reservoir levels on the timing of hatch in a Largemouth Bass population (Dai and National Center for Atmospheric Research Staff 2019). While we conducted this study across only 2 y, water temperature and reservoir level were in contrast during the hatch dates both years. Previous work suggests that water temperature is important in the timing of spawning for Largemouth Bass (Sammons et al. 1999). The median water temperature differed by 1.4°C during the hatch dates between 2017 (19.0°C) and 2018 (17.6°C) and the overall temperature range between the 2 y (13.8–23.0°C) was similar to that of Largemouth Bass populations throughout the southeastern United States, which spawn from mid-April to May (12–24°C; Kramer and Smith 1962; Miller and Storck 1984; Sammons et al. 1999). Thus, the differences in water temperatures did not appear to alter the spawning window of Largemouth Bass in Elephant Butte Reservoir, nor did they exceed temperatures considered optimum for growth in the species (25°C; Nimi and Beamish 1974).

The magnitude and duration of water-level fluctuations may dictate the success or failure of nests (Ozen and Noble 2002). Strong year classes of Largemouth Bass were associated with water levels at or above full capacity of southeastern reservoirs (Maceina and Bettoli 1998; Sammons and Bettoli 2000). Kohler et al. (1993) also observed a correlation between strong year classes of Largemouth Bass and high-water years. Elevated but stable water levels inundated vegetation in littoral areas that provided young-of-year Largemouth Bass with increased cover from predation as well as increased availably of prey (Parkos et al. 2011). We observed the median daily growth rate in young-of-year Largemouth Bass during the hatch dates was greater when Elephant Butte Reservoir was increasing in 2017 (1.02 mm/d) than in 2018 (0.82 mm/d). This suggests that the timing in rising reservoir levels and an increase in water temperature benefited hatch and growth of young fish.

Although fluctuating water levels (+0.88 to −3.94 cm/d) in Lake Mead, Arizona, did not affect hatching success of the Largemouth Bass population, the author of the study predicted that if water level declined by 6 cm/d, then 85% of the nests would have been exposed to wind and wave action (Morgensen 1983). While monitoring nest success and larval survival was beyond the scope of our study, Elephant Butte Reservoir experienced declining water levels throughout hatch dates by as much as 8.5 cm/d in 2018. This observation corresponded with a reduction of relative abundance and reduced growth of young-of-year Largemouth Bass. Although we recognize the limitations of a 2-y study, our results suggest that as long as water temperature is optimal throughout the final stages of ovulation and spawning, then the timing of environmental factors that include reservoir filling or receding may contribute to the success or failure of the reproductive cycle in the Largemouth Bass population.

In light of a changing climate and increased user conflicts, our findings have implications for the management of southwestern U.S. reservoirs that experience large fluctuations in water levels. We recognize that two water years does not necessarily provide a strong inference on time of hatch for the Largemouth Bass population. However, hatch dates were similar to those reported in other regions of the United States and reflect the timing of the annual reproductive cycle of Largemouth Bass might be similar across the species range. Temperate fishes such as Largemouth Bass rely on changes in day length (i.e., photophase; Rosenblum et al. 1994) and water temperature (Peter and Hontela 1978) to trigger reproductive hormonal changes to time spawning events. Similar hatch dates between water years in which rates of change in the reservoir were contrasting reflected that the timing of hatch in a Largemouth Bass population in a southwestern reservoir experiences the same reproductive cues to those in southeastern and northeastern regions of the United States where the species originated.

If water managers can hold water levels steady in Elephant Butte Reservoir beginning mid-April through to the end of May, then an adequate window would provide sufficient opportunity for egg development, hatch, and emergence of fry from the nests. Maintaining stable water levels in a southwestern irrigation reservoir during critical periods of hatch seems unlikely, but not impossible. Texas Parks and Wildlife partnered with the Brazos River Authority to identify minimum water levels in the Brazos River Reservoir, Texas, to provide littoral habitat for fish (Daugherty et al. 2015). While natural resource and water managers need additional work to identify minimum water level that would ensure available littoral fish habitat while meeting water demand, we identified a potential time frame during which these managers could work proactively to balance water allocation with spawning of a nest-building fish species in an arid-lands irrigation reservoir.

Please note: The Journal of Fish and Wildlife Management is not responsible for the content or functionality of any supplemental material. Queries should be directed to the corresponding author for the article.

Data S1. Catch per unit effort (CPUE) includes the year and date (2017–2018) in which young-of-year Largemouth Bass Micropterus salmoides were collected from three study coves in Elephant Butte Reservoir, New Mexico. Total length (mm) and weight (g) are presented. Effort represents time in seconds in which young-of-year were collected.

Available: https://doi.org/10.3996/JFWM-21-071.S1 (7 KB CSV)

Data S2. Hatch dates for 2017 and 2018 young-of-year Largemouth Bass Micropterus salmoides from three study coves in Elephant Butte Reservoir, New Mexico, fish identification or unique number assigned to the fish, and hatch dates from retrospective analysis of daily rings in both left and right sagittal otoliths.

Available: https://doi.org/10.3996/JFWM-21-071.S2 (2 KB CSV)

Data S3. Water level rate of change (ROC) includes date of the calculated ROC at midnight (0:00). Storage volume of Elephant Butte Reservoir (km³) obtained from the website maintained by the U.S. Bureau of Reclamation (https://water.usbr.gov/query.php?sites=elephantbuttedam; accessed 20 May 2019). Elevation represents storage (m) for the date. ROC was obtained by subtracting elevation of day 2 from elevation of day 1 and multiplying by 100 to get ROC in centimeters.

Available: https://doi.org/10.3996/JFWM-21-071.S3 (33 KB CSV)

Data S4. Water temperature (°C) includes year of collection (2017–2018), date and time of water temperature collection in Greenwich Mean Time (−6), and sites (JU, The Jungles; MRO, McRae outer cove; MRI, McRae inner cove; KT, Kettle Top) of three study coves in Elephant Butte Reservoir, New Mexico. Continuous temperature is provided in hourly increments.

Available: https://doi.org/10.3996/JFWM-21-071.S4 (1.765 MB CSV)

Reference S1. Kunkel KE, Stevens LE, Stevens SE, Sun L, Janssen E, Wuebbles D, Kruk MC, Thomas DP, Shulski MD, Umphlett NA, Hubbard KG, Robbins K, Romolo L, Akyuz A, Pathak TB, Bergantino TR, Dobson JG. 2013. Regional climate trends and scenarios for the U.S. National Climate Assessment Part 4. Climate of the US Great Plains. Washington, D.C.: NOAA Technical Report NESDIS 142-4.

Available: https://doi.org/10.3996/JFWM-21-071.S5 (5.124 MB PDF)

Reference S2.[NMDGF] New Mexico Department of Game and Fish. 2016. Statewide fisheries management plan. Sante Fe: New Mexico Department of Game and Fish.

Available: https://doi.org/10.3996/JFWM-21-071.S6 (6.845 MB PDF)

Funding was provided by the New Mexico Department of Game and Fish. The authors thank E. Mammoser, K. Gardner, E. Enriquez, and J. Miller for equipment, assistance, and technical advice. Extensive field and laboratory support was provided by A. Guerrero, C. Sartin, A. Davis, J. Johnson, and K. Warden. Additional funding was provided by Bass Pro Shops. We thank R. Patiño, D. Shoup, and three anonymous reviewers, including the Associate Editor, for helpful comments on earlier drafts of the manuscript. The project was conducted in accordance with the ethical standards of New Mexico State University's Institutional Animal Care and Use Committee (Project 2016-030) and New Mexico Department of Game and Fish Scientific Collection Permit (no. 3033). All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. The authors declare that they have no conflicts of interest.

Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.