Bull. Southern California Acad. Sci. 116(3), 2017, pp. 174-192 © Southern California Academy of Sciences, 2017 A Longitudinal Temperature Profile of the Los Angeles River from June through October 2016 Jennifer Mongolo,* 1 Nina Trusso,1 Rosi Dagit,1* Andres Aguilar,2 and Sabrina L. Drill3 ]RCD of the Santa Monica Mountains, 540 S. Topanga Canyon BlvdTopanga, CA 90290 2 Department of Biological Sciences, California State University, Los Angeles, CA 90032 3 University of California Cooperative Extension, 669 County Square Drive, Ventura, CA 93003 Abstract.—This pilot study developed a longitudinal temperature profile of the Los Angeles River by deploying temperature loggers throughout the watershed between June and October 2016. The watershed was divided into zones based on river system component, urbaniza¬ tion, and channelization. Channelized sites recorded the highest temperatures, tributaries recorded the lowest, and the estuary showed the most fluctuation. Overall, temperatures were too warm to support re-introduction of native fish but currently support non-native fish species. Temperature mitigation is needed for native species to re-establish. Albeit limited in scope, this study establishes a baseline of summer/fall temperatures in the Los Angeles River. The 82-kilometer-long Los Angeles River (LAR) is an urban river that flows through 14 cities and unincorporated areas in Los Angeles County, California. Approximately 77.25 km of the 82 km main stem of the river is contained in concrete flood control channels, leaving three miles of river with natural channel bottom along the main stem. The soft bottom reaches occur at three locations: the estuary in Long Beach between Willow Street Bridge and the Long Beach Harbor, the Sepulveda Flood Control Basin, and the Glendale Narrows. Tributaries in the upper watershed within the Angeles National Forest, upper Arroyo Seco and Upper Tujunga Wash, remain in a fairly natural state with natural substrate and riparian vegetative cover. When these reaches flow into more urbanized areas, however, they are often channelized as well, with variable levels of natural channel bottom and riparian vegetation remaining depending on flood risk to nearby urban areas. Currently, native fish species only reside in the upper reaches of the watershed at Big Tujunga Wash and the Arroyo Seco, as well as in the estuary. The obligate freshwater community found in the upper reaches includes arroyo chub (Gila orcutti), Santa Ana speckled dace (Rhinichthys osculus ssp.), and Santa Ana sucker (Catastomus santanae). The freshwater life history form of rainbow trout (Oncorhynchus mykiss) is also present in the upper watershed in the Arroyo Seco, but the federally endangered anadromous southern steelhead form of Oncorhynchus mykiss was last observed in the LAR watershed in the 1940s (Swift et al. 1993). The Recovery Plan for the extirpated unarmored threespine stickleback (Gasterosteus aculeatus williamsoni) calls for reintroduction in the LAR watershed1, while the now state and federally listed endangered Pacific lamprey (Entosphenus tridentatus), which was historically found in rivers throughout * Corresponding author: [email protected] 1 US Fish and Wildlife Service. 1985. Unarmored threespine stickleback recovery plan. USFWS Portland, Oregon. 174 LOS ANGELES RIVER TEMPERATURE PROFILE 175 southern California, has not been observed in the LAR in many years2 3 4. Currently, numerous non-native fish species, including common carp (Cyprinus carpio), Nile tilapia (Oreochromis niloticus), green sunfish (Lepomis cyanellus), and largemouth bass (Micropterus salmoides) are found in the limited soft-bottom areas of the watershed and represent the dominant ichthyofauna of the river reaches that are targeted for major restoration efforts3,4. The Los Angeles River is headed for an extraordinary restoration effort5, but the form and direction of restoration is still under development. Underpinning restoration of the river is the need to understand how the contemporary aquatic community will respond to restorative actions, and to identify barriers to re-establishing native species. Without accurate characterization of the existing instream conditions in priority restoration reaches, the benefits of the multi-billion- dollar effort to revitalize the river will be difficult to determine. Among the suite of factors that influence distribution and abundance of fish species, water temperature is one of the most important. As ectotherms, the body temperature of a fish is linked to the temperature of the water in which it resides. This means that growth, metabolism, feeding rate, reproduction, and rearing are all tied directly to water temperature. Furthermore, most aquatic organisms, such as benthic macroinvertebrates, that fish rely upon as food sources are poikilotherms, and are also limited by water temperature. Data on critical thermal temperatures for native fishes historically found in the LAR watershed are limited in the current literature. While increased summer water temperatures tend to be a major limiting factor for most salmonids in other areas, multiple studies conducted in southern California show that rainbow trout (O. ruykiss) demonstrate more flexibility in their temperature range and an ability to acclimate to higher temperatures within the southern extent of their range (Boughton et al. 2007; Myrick and Cech 2000; Myrick and Cech 2005; Spina 2007). Critical thermal maxima (CTM) ranging from 23°C to 31.5°C have been reported for O. mykiss in southern California creeks (Bell 1986; Dagit et al. 2009; Sloat and Osterback 2012). A detailed study of the unarmored threespine stickleback (G. aculeaius williamsoni) in the Santa Clara River found CTM for this species was 30.4CC when individuals were acclimated to 8°C, and 34.6°C when acclimated to 22.7°C (Feldmeth and Baskin 1976). Unfortunately, among historically native LAR fishes, these are the only species for which detailed experimental studies on CTM have been published. While not experimentally derived, field observed temperatures from studies of native fish in the Los Angeles basin and other southern California watersheds can be utilized as indicators of the thermal requirements of target species relative to the conditions currently found in the LAR. For instance, field observations of Santa Ana sucker (C. santanae) published by Feeney and Swift (2008) show that larvae may bask in slower flowing areas where temperatures reach 24°C, while juveniles may retreat from warm summer flows (up to 30°C), congregating in cooler areas (15-22°C) near tributary or groundwater sources. Saiki et al. (2007) observed juvenile Santa Ana suckers in June in the Santa Ana River when daytime temperatures averaged 25.3°C, and 2 ~ CalFish. Accessed on 15 August 2017. http://www.calfish.org/FisheriesManagement/SpeciesPages/ PacificLamprey.aspx. 3 Swift, C. C. and S. L. Drill. 2008. State of the River 2 - The Fish Study. Friends of the Los Angeles River (FoLAR). Los Angeles, CA. 4 Friends of the Los Angeles River (FoLAR). 2016. State of the River 3: The Long Beach Fish Study. Los Angeles, CA. ' City of Los Angeles, Department of Public Works, Bureau of Engineering and US Army Corps of Engineers, Los Angeles District, Planning Division. 2007. Final programmatic environmental impact report/programmatic environmental impact statement. Los Angeles River Revitalization Master Plan. Los Angeles, CA. 176 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES ranged to 30.8=C, as well as in September in the San Gabriel River when temperatures averaged 19.6CC (range 15.0-22.9°C). USFWS6 reported mortality events for Santa Ana suckers when temperatures exceeded 32.8°C in the Santa Ana River and 26.7°C in Big Tujunga Creek.6 In a survey of the upper San Gabriel River from 2007 and 2008, O’Brien et al. (2011) reported mean daily temperatures of ~21°C in the north and east forks of the San Gabriel River and ~20CC in the west fork. This study reported Santa Ana sucker, rainbow trout, and Santa Ana speckled dace in all three forks of the San Gabriel River, while arroyo chub were only found in the east and west forks (O’Brien et al. 2011). While arroyo chub are physiologically adapted to survival in habitats with wide temperature fluctuations (Castleberry and Cech 1986), they are most commonly found in low gradient streams where water temperatures do not exceed 28CC, and where spawning temperatures ranging from 14-22°C are available (O’Brien 2009; Moyle et al. 2015). Moyle et al. (1995) found that Santa Ana speckled dace prefer perennial streams fed by cool springs that maintain summer water temperatures below 20° C. Even though much of the L AR has been channelized, there are still areas with natural substrate that could potentially provide suitable habitat for native fish (i.e. Glendale Narrows, Sepulveda Basin). The lack of concrete lining at these locations accommodates groundwater upwelling, which provides refugia habitat that currently support both non-native fishes as well as native amphibians absent in concrete reaches of the LAR7. Since water temperature is so closely tied to the distribution and abundance of fish species at various life stages and with that of their prey animals, a longitudinal temperature profile of the river can be used as an indicator of habitat quality on the watershed scale (Poole et al. 2001a). Determining where water temperature in the LAR is currently suitable for native fish is an important first step for any proposed restoration effort. If temperatures are in fact suitable for native species, then future efforts can focus on targeted in-stream and riparian habitat restoration, non-native species management, or other non-temperature related actions. If temperatures in the river are not suitable for native species, future restoration efforts should be developed with a focus on improving the temperature regime of the river for native fishes. A study to capture a detailed thermal profile of the LAR was initiated in early 2016, with installation of continuously recording temperature data loggers at 13 sites throughout the water¬ shed. Temperature data was recorded from June through October 2016. The intent of the study was threefold: to characterize temperatures throughout the watershed; to document current base¬ line conditions at representative locations during the most stressful summer conditions; and to identify opportunities for restoration of native fish habitat. While limited in scope, the present study provides an initial, albeit incomplete, picture of baseline summer/fall temperatures in the LAR against which future studies and conditions can be compared. Materials and Methods The study area includes the main stem and three major tributaries of the LAR, from its headwaters in the Angeles National Forest and western San Fernando Valley, to the estuary in Long Beach (Fig. 1). For comparison purposes, the watershed is divided into six zones based on river component (tributaries: A, C, E; mainstem: B, D, and estuary: F). Tributaries within 6 US Fish and Wildlife Service. 2014. Draft recovery plan for the Santa Ana sucker. USFWS Pacific Southwest Region, Sacramento, CA. 7 Swift, C.C. and J. Seigel. 1993. The past and present freshwater fish fauna of the Los Angeles River, southern California, with particular reference to the area of Griffith Park, in The biota of the Los Angeles River: an overview of the historical and present plant and animal life of the Los Angeles River drainage. (K. Garrett, ed.) Los Angeles Natural History Museum Foundation, Los Angeles, CA, 28 pp. LOS ANGELES RIVER TEMPERATURE PROFILE 177 Valley, L\ SanA«,,,,XJ.V<::'-V F i(nn*l Mt. Wilson CBS _ ' L iAI >i»11,1 m Van Nuys ^ t ' : v -' V''} .- ' , notice & ,, AirfortX Tillman Water Cl « ,Mi\j.Recl(!imafidn Plant F2TT ffi A2 (cid:9632) Agour a BT^IZSepulveda Dam BZ^-. ,,E285. PaC2^'^:pasadepa21,t w ation Plant Ateadia ' *‘i1 % Beverly Hills Angejem VVest. Covina \ Al !* Los Angeles Downtown i Inglewood Long Beach Daugherty Norwalk 2016 Temperature Study Los Angeles River Watershed Fuller Los Angeles River Mainstream Redondo Sott-Bottom Reach Beach Tributary streams LDoanugg hBeeratcyh F ield (1) 2016 Temperature Logger ® Garden Deployment Sites Rancho » Grove1: Weather Station Palos Long vVerdes Stream Gauge Beadi Fig. 1. Study Area showing zones and temperature logger locations. a geographic region are ascribed their own zones and the main stem was further divided into natural (B) and channelized sections (D). Sites representative of depth and canopy cover were chosen within each zone and temperature loggers were installed at 13 locations in May 2016 (Fig. 2). Sites were selected based on the following criteria: accessibility, safety, location with respect to soft bottom reaches, depth, canopy cover, and tributary inputs, and distribution along the river (Table 1). Water temperature data was collected from June through October 2016 using a combination of ONSET HOBO TidbiT v2 Water Temperature Data Loggers and HOBO Pendant Temperature Data Loggers (collectively, HOBOs) programmed to record time, date, and temperature. The TidbiT v2 has an accuracy of ± 0.21 °C and the Pendant has ±0.53°C accuracy; both are designed for use in outdoor and underwater environments. The study period was selected to align with southern California’s dry season, with the highest air temperatures and lowest precipitation, when thermal stress on fish would be most likely to occur. All loggers were prepared for deployment in the water column with appropriate site-specific materials to anchor them in place depending on site conditions (substrate, vegetation, access, etc.) and water depth. At all sites, the loggers were crimped to one end of an approximately one-meter long line of 90 lb-test stainless steel trolling wire using 1.40 mm leaden sleeves. The method used to anchor the other end of the wire to the stream channel varied depending on site 178 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES D4 De Forest Park El Compton Creek FI Willow St Bridge (concrete) F2 Willow St. Bridge (soft bottom) Fig. 2. Photograph panel of Los Angeles River study sites June-October 2016. conditions. For natural flow areas with mature vegetation, the devices were secured to a tree trunk, root, boulder or other stable object at the water’s edge in such a way as to allow the logger to hang near the bottom of the water column while keeping it out of plain sight in order to reduce incidents of vandalism. For concrete channels and other areas where the previous method was not feasible, loggers were bolted to the channel wall. Weights were added to all wires to help prevent loggers from being swept up on shore during high flow events. Data loggers were not enclosed in protective housing, were of similar coloration to the surrounding substrate, and were only protected from direct sunlight where sufficient riparian vegetation was present to provide shading. Locations in Zone A were managed by the Council for Watershed Health through their Los Angeles River Watershed-wide Monitoring Program. Both sites in Zone A had one HOBO in LOS ANGELES RIVER TEMPERATURE PROFILE 179 oo'5ooo3)'s_ _ ^ _ oooooo<^oo^oo dddddccdddddd 3333333333333 3s dd d d d b b -E -E fc* frl 3d33 3d33 * 3 3 3 u O t3 e3 (J U U g 6 | 3 d d X3> 3p rHS rH2 'S3 ‘2c3 •Hc 3-Hc> -4235 23 23 5 HI- 2I 2 s 6 O (N '(cid:9632)O 11 3g 25 nO0\nNini^-Oiin^’'^ftfO'N>OfDnO^fidOn'^ftpSdHo-O 6OoOd <u O 60 ^ q60 3H3 s a e s e s o o o o o o o o o o o o o CQ CQ QQ 03 CQ CQ CQ <d <t3 cd 4S cK 4S cd CoO Wo Wo Mo Mo Wo Wo s 60 •8 (d o> Jo3 *«0 X5>0 ^ "ug ^ q -c ‘s ^ si ^ Is 8 cq oq co § £ 3d Q o h 23 -38 g ©c on c/5 O>, ^d 3o gd <U CQ -8 c/5 ^ 8 K 5 > 8 2 x1>3 --3P --©OO£ c»©Oo J35 "^. ”>©> O3 go 5® 5® <*3 W3 CQCQ<(^<JEQ(J^ <n m rt —i *-i ?n 2 O Na> ac J* 2 U 2 2 H, a, 11 £o £o O Q U U W fe 180 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Table 2. Summary of potential and missing data points (148 days) at each site for the study period of June 4, 2016 at 12:00 to October 30, 2016 at 18:00. A1 A2 B1 B2 Cl C2 D1 D2 D3 D4 El FI F2 Potential Data Points 886 886 7117 7117 7117 7117 7117 7117 7117 7117 7117 7117 7117 Missing Data Points 99 20 197 196 98 1 3515 5184 7 21 2140 1 57 Missing Data Days 12 4 4 4 2 0 73 108 0 0 45 0 1 the water recording temperature every four hours and one HOBO outside the water recording air temperature at the same four-hour interval. Data recorded at these sites were provided in a Microsoft Excel file for comparative analysis. Locations in Zones B-F had one HOBO per site recording water temperature at 30-minute intervals. Each site was visited monthly by trained citizen science volunteers to download the recorded data, ensure loggers were secure, and photograph site conditions. Data from each logger was offloaded using a HOBO U-DTW-1 Waterproof Shuttle Data Transporter in the field, which was subsequently uploaded to a computer using Hoboware PRO software, then compiled in a Microsoft Excel database. Data points available at each site varied due to environmental factors affecting temperature readings (e.g. dry-downs, washouts, etc.), theft or vandalism, and equipment malfunction. These factors affected both the thermometer’s ability to record data and its ability to take data repre¬ sentative of river conditions. Of all potential data points, less than 15 percent were absent for the entire study period across all sites. The majority of missing data occurred at sites Dl, D2, and El (Table 2). Water temperature data recorded from study reaches were summarized to establish a daily maximum, minimum, and mean temperature for each site. These daily metrics were combined to establish monthly mean, maxima, and minima. Temperature metrics were compared between study sites: 1) to examine differences between concrete and natural bottom locations; 2) to examine differences between sites in the main stem and tributaries; 3) to calculate the frequency, time of day and duration when temperatures exceed thermal limits for target native fish species; and 4) to map the changes in temperature throughout the river. A quality assurance/quality control process to ensure data accuracy included several levels of review. The first level occurred when HOBO readings were imported into Microsoft EXCEL, and included completeness and examination for unusual outliers or missing information. Then, difference in temperature readings between consecutive data points was analyzed in an effort to differentiate between natural extreme changes in temperature, unnatural extreme changes in temperature representative of river conditions, and unnatural extreme changes in temperature that are not representative of river conditions (HOBO being handled or out of water during temperature recording). Table 2 summarizes the completeness of data collected between June and October 2016. Precipitation and daily minimum and maximum air temperature records were obtained from five different NOAA weather stations throughout the Los Angeles basin. Daily flow data was obtained from Los Angeles County Department of Public Work’s eight gauging stations in the Los Angeles basin. Weather stations and stream gauges are mapped in Fig. 1. The data obtained was examined for relationships with water temperature. Correlations between daily maximum water temperatures and daily maximum air temperatures and flow measurements were determined independently for each site. Daily maximum air temperatures showed a high degree of collinearity (not shown), so subsequent analyses utilized daily maximum air temperature from the Mount Wilson weather station. The sole exception was site Cl, at which daily maximum LOS ANGELES RIVER TEMPERATURE PROFILE 181 water temperatures showed the highest correlation to air temperature data from the Long Beach weather station. Data from the flow and weather station with the highest correlation coefficient for each site was used in a multiple regression with daily maximum water temperature as the dependent variable. All analyses were performed in R8. Results Between June and October 2016, the highest daily maximum water temperatures occurred in mainstem concrete bottom reaches (D2, D3, D4), while lower temperatures occurred in tributary reaches with more natural substrate and riparian vegetation (Al, A2, Cl, C2, El) (Fig. 3). Site D1, along the main stem with natural substrate and concrete banks, exhibited the highest daily maximum temperatures particularly in June and July. Less urbanized sites B1 and B2 in the Sepulveda Basin showed moderate daily maximum temperatures compared to other sites, while the estuary sites (FI and F2) demonstrated more variability in maximum daily temperatures across the season than other sites, probably due to tidal influence. At site F2, overnight high temperatures are in close coordination with late night high tides. The maximum water temperatures observed in all study sites are shown by month in Table 3. Monthly maximums were lowest in tributaries (Zones A, C, E). Highly urbanized main stem sites recorded the highest maximums (Zone D), and less urbanized main stem areas were mid-range (Zone B). Maximum temperatures showed the widest range in June, with readings ranging from 20.6°C to 36.8°C. Channelized sites D1 and D2 reached their highest temperatures in the month of June, while channelized sites D3 and D4 reached their highest temperatures in July. The highest single temperature reading of the season was recorded at the Long Beach Estuary site F1 in August. The other estuary site F2 also experienced its highest temperature during the month of August. Compton Creek El was the only site to record its highest maximum temperature in September. Zone B sites had the most consistent monthly maximums. All other sites showed more variation in maximums from month to month. The most extreme monthly maximum variation occurred at site Cl with monthly maximums of 20.9°C, 22.1°C, 36.7°C, 25.4°C, 33.3°C occurring in June through October respectively. However, this was the shallowest site, and dry downs were a continuous issue, requiring relocation of the logger on multiple occasions throughout the study period. The highest temperatures recorded at C1, therefore, are most likely due to water receding to such extent that the logger recorded air temperature for some time before being relocated to a deeper pool. Overall, monthly minimum water temperatures show less variation across sites than monthly maximums (Fig. 4). Sites B1 and B2 in Sepulveda Basin had the highest minimum temperatures, but also had the smallest monthly ranges (3.6-7.1°C) and recorded relatively cool maximums compared to other sites. In sites with concrete bottoms, the range between monthly maximum and minimum is greater (13.2-20.1GC). Zone A sites in the Angeles National Forest consistently had the lowest minimums throughout the season followed by site C1 (a fairly remote and natural tributary reach just downstream of Al). A short distance downstream, site C2 also recorded relatively low minimums during June and July. The lowest mean water temperatures were recorded in tributary sites A1, A2, C1, C2, and E1. Sites B1 and B2 in the less heavily urbanized Sepulveda Basin portion of the study area had the highest monthly averages and the smallest ranges of temperatures between monthly maximum and minimum. R Core Team. 2016. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. 182 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Jun Jul Aug Sep Oct Nov Date Jun jul Aug Sep Oct Nov Date ‘tQerPr '-S.C rtft Jun Jul Aug Sep Oct Nov Date Fig. 3. Daily maximum temperatures at all sites plotted by date between June - October 2016. Smoothed lines are shown. LOS ANGELES RIVER TEMPERATURE PROFILE 183 Table 3. Maximum water temperatures (max), minimum water temperatures (min), and range between maxi¬ mum and minimum water temperatures each month (range). Highest maximum water temperatures for each month shown in bold; highest maximum water temperature for each site underlined. June July Aug. Sept. Oct. Site Max Min Rng Max Min Rng Max Min Rng Max Min Rng Max Min Rng Al* 23.7 17.2 6.5 26.3 16.7 9.6 25.8 15.8 10.0 23.5 13.7 9.8 19.9 13.8 ‘ 6. .1 A2* 20.6 13.7 7.0 19.9 14.0 5.8 19.9 13.8 6.1 18.5 13.8 4.8 17.7 13.5 4. .2 Bl* 28.4 21.2 7.1 28.7 25.1 3.6 29.3 23.8 5.5 26.7 21.7 5.0 24.8 18.4 6. .3 B2* 29.9 22.8 7.0 30.5 25.9 4.6 30.8 25.5 5.4 29.3 23.3 6.0 26.3 19.9 6. .4 Cl* 20.9 15.2 5.7 22.1 17.8 4.4 36.7 16.0 20.7 25.4 14.4 11.0 33.3 13.4 20. .0 C2* 31.5 14.4 17.1 31.3 17.8 13.5 26.2 19.7 6.5 21.6 17.2 4.4 31.0 15.1 15. .9 Dl* 36.8 17.1 19.7 36.5 19.8 16.8 35.5 19.1 16.5 - - - - - - D2 33.2 20.0 13.2 31.5 23.5 8.0 - - - - . _ - - - D3 35.7 17.2 18.6 36.4 20.6 15.7 35.6 19.8 15.8 33.8 17.4 16.4 31.3 17.0 14. .3 D4 35.6 16.7 18.9 35.7 20.4 15.3 34.9 19.5 15.4 33.3 17.0 16.3 33.4 13.2 20, .1 El* 26.4 16.5 9.9 25.0 19.7 5.4 26.8 19.6 7.2 29.5 17.7 11.8 - - - FI 33.3 20.9 12.4 34.9 20.3 14.6 36.1 19.6 16.5 32.1 17.5 14.6 28.0 17.5 10, .5 F2* 34.4 21.3 13.1 34.0 20.2 13.8 37.0 18.6 18.4 30.6 17.5 13.1 28.6 15.4 13, .1 * indicates natural bottom location Hourly variation is shown in Fig. 5. Throughout the study period, the coolest temperatures in the LAR were recorded in the early morning, between 06:00 and 08:00, except for site F2 whose coolest hour on average was 11:00 (Table 4). The highest temperatures occurred between 13:00 and 20:00, with the majority of sites peaking between 14:00 and 16:00. Greater diurnal variation occurred in highly urbanized zones at D2, D3, El, and FI, while diurnal variation was much diminished in more natural sites with soft bottoms and riparian vegetation such as sites Al, A2, Cl, and C2. In the less urbanized Sepulveda Basin in the San Fernando Valley region, sites B1 and B2 were warmer overnight throughout the whole season. Average nighttime (17:00- 05:00) temperatures were 0.53°C warmer than average daytime (05:00-17:00) temperatures at site B1 and 0.86c warmer at site B2. This pattern of warmer overnight temperatures was also observed at the estuary sites FI and F2. Nighttime temperatures were 3.07°C warmer than daytime temperatures at FI, and 0.95°C warmer at nighttime than daytime at F2 throughout the study period. The highest seasonal water temperatures occurred in the most heavily developed portions of the watershed, namely Dl, D2, D3, and D4, all with average maximum temperatures for the season topping 30°C (Table 5). Site Dl had the highest temperatures of the season (average maximum T = 34.1°C). This site has a natural substrate bottom but concrete lined banks. This site also demonstrated the largest difference between average maximum and average minimum temperatures during the study period. The lowest temperatures in the watershed were recorded in Zone A, a relatively natural portion of the watershed found within the Angeles National Forest. The main stem channel reaches in the Sepulveda Basin (B1 and B2) showed the most stability with a 1.4C and 2.3° difference between seasonal average maximum and seasonal average minimum temperatures. Fig. 6 illustrates these temperatures in a longitudinal profile throughout the watershed to highlight thermal barriers to movement of native fishes from headwaters to the ocean and vice versa. The changes in water temperature along the longitudinal continuum of the river from head¬ waters to estuary are illustrated in Fig. 7. The range of temperatures at each site are plotted