Shedding Light on the Role of Artificial Light at Night in Lake Washington, WA, USA

Shedding Light on the Role of Artificial Light at Night in Lake Washington, WA, USA | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Shedding Light on the Role of Artificial Light at Night in Lake Washington, WA, USA Jennifer Schulien, Tessa Code This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6406457/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Artificial light at night (ALAN) poses a threat to ecosystems globally. It includes both direct and indirect light, or skyglow, which occurs when ALAN scatters in the atmosphere, extending beyond its source. We analyzed ALAN trends in Lake Washington, WA, from 2014 to 2023 using Visible Infrared Imaging Radiometer Suite (VIIRS) nighttime light measurements, evaluated the relationship between in-situ and satellite measurements, and assessed juvenile salmon predation risk due to ambient light. We observed significant increases in ALAN in all open water regions, while nearshore regions primarily showed no or negative trends, revealing the significant role of skyglow on open water light levels. Although spatial correspondence between the satellite and in-situ measurements was observed, VIIRS did not capture the changes in yellow and blue light. Juvenile salmon at the shallow southern open water site experience light levels 28 times brighter than fish at the northern site. In the Ship Canal, a narrow corridor for outmigrating salmon, light levels are 65 times brighter than at the southern site. These contrasts in light exposure highlight the impact of local conditions on predation risk by visual foraging predators and emphasize the need for effective mitigation efforts targeting both nearshore and distant light sources. Biological sciences/Ecology Earth and environmental sciences/Ecology Earth and environmental sciences/Environmental sciences Earth and environmental sciences/Limnology Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Artificial light at night (ALAN) has increasingly been recognized as a global threat to ecosystems 1,2 and continues to increase in both intensity and spatial coverage. Since the mid-20th century, ALAN has increased 6% per year on average (0–20% range) 3 , with currently lit areas brightening at a rate of 2.2% annually 4 . ALAN includes both direct light and indirect light, or skyglow, which occurs when ALAN scatters in the atmosphere, extending beyond its original source 5 . Skyglow can have significant ecological impacts, as even small increases in light can significantly alter environmental conditions, cueing biological responses 6,7 . As urbanization increases, with more than half of the global population living near freshwater shorelines 8 , rivers and lakes are increasingly exposed to light pollution. ALAN significantly alters both the quality and quantity of light in aquatic environments 9 , penetrating well below the surface and disrupting natural light gradients. Fish are particularly susceptible to changes in light intensity as they use daily light cycles to cue foraging and diel vertical migration 10 . Moreover, ALAN influences fish behavior and trophic relationships, as demonstrated in numerous laboratory and field studies 11–13 . In Lake Washington, located adjacent to Seattle, WA, ALAN significantly affects predator–prey dynamics, as observed for the piscivorous cutthroat trout ( Oncorhynchus clarki ) 14 and sculpins ( Cottus sp.) 13 , both of which are voracious predators of juvenile salmon species 15,16 . Understanding the intricate interplay between ALAN, underwater light dynamics, and biota, is thus essential for identifying and mitigating both the direct and far-reaching effects of ALAN on aquatic ecosystems. Measuring spectral nighttime light presents significant challenges due to the high cost and operational complexity of the required sensors 1 . While broadband sensors are more affordable and widely used for nighttime light measurements, they come with limitations 17 . The widely used Sky Quality Meter (SQM), for example, reports measurements in the astronomical units of magnitudes per square arc second, a unit that is instrument specific 18 . Calibrated, spectrally discrete, radiometric units are necessary to address species-specific questions and to engage government agencies and the lighting industry. Tamir et al. ( 2017 ) were the first to study in-situ spectral ALAN measurements in the Gulf of Aqaba using a profiling reflectance radiometer 19 . These measurements were used to quantify light intensity and spectra affecting offshore regions. Our study is the first assessment of ALAN in a freshwater system, shedding light on its spatial and spectral impacts on predator–prey dynamics in an urban waterbody. Here we examined spatiotemporal trends in ALAN in both the nearshore and open water regions of Lake Washington, WA (Fig. 1 ) using both 1) Visible Infrared Imaging Radiometer Suite Day-Night Band (VIIRS/DNB) reflectance (L VIIRS ) data and 2) surface downwelling irradiance (E d (λ,0 + )) measurements at 18 wavelengths collected using a Compact Optical Profiling System (C-OPS; Biospherical Instruments Inc., San Diego, CA). We calculated nighttime light (from surface light and the downwelling diffuse light attenuation coefficient, K d (λ), measurements) at the depths where high fish densities were detected using hydroacoustic surveys and discrete depth trawl nets. Results Inter-annual Nighttime Light Trends We found significant trends over the past decade during the summer period (June-September) in 20 of the 30 regions of interest (ROIs) evaluated (Fig. 2 ; S.Table 1 ). ALAN increased significantly in every open water ROI. In contrast, in the nearshore ROIs, negative or no trends were observed, except for the northwest, north Mercer Island, and I-90 bridge ROIs, where positive trends were also observed. Interestingly, several ROIs with notably high average pixel reflectance, such as the Ship Canal, Renton, and Kirkland, exhibited primarily decreasing trends in ALAN. Table 1 Statistical summary of the exponential model relating monthly cloud cover to VIIRS/DNB nighttime reflectance (N = 12), including p-values, F-statistics, and adjusted R 2 values. Region of Interest P-value F-statistic R 2 Northern pelagic 9.80E-10 312 0.569 Middle pelagic north 3.52E-09 240 0.537 Southwest pelagic 3.80E-10 378 0.0869 Middle pelagic middle 1.96E-11 687 0.614 Middle pelagic south 9.40E-10 314 0.339 Southeast pelagic 3.53E-11 611 0.553 Monthly Skyglow The seasonal pattern in cloud cover means (based on hourly observations from Renton Municipal Airport) paralleled that of VIIRS/DNB reflectance spanning from 2022 through 2023 (Fig. 3 ). A significant exponential relationship for all of the six ROIs evaluated was observed (Table 1 ). Spatial Variability in Nighttime Irradiance Nighttime downwelling irradiance at 589 nm, E d (589), generally increased from north to south in the lake (Fig. 4 a). Four of the sites in the Ship Canal and three in Renton exhibited significantly higher E d (589) values compared to the other site locations (Fig. 4 a). The Renton site just offshore of Gene Coulon Park was brightest, with E d (589) equal to 8.0x10 -4 µW cm -2 nm -1 . This is 2–4 times greater than E d (589) measured at the two starred sites closer to the Cedar River mouth, (3.9x10 -4 and 2.1x10 -4 µW cm -2 nm -1 ). For reference, this is about an order of magnitude brighter than the nearshore region north of Gene Coulon Park where E d (589) is ~ 4x10 -5 µW cm -2 nm -1 . E d (589) in the Ship Canal at the four starred sites ranged from 1.4x10 -4 to 5.4x10 -4 µW cm -2 nm -1 . Comparison of Satellite and In-situ Nighttime Light Measurements Consistent with the spatial correspondence, the brightest regions of the lake identified using the C-OPS reference sensor were significantly linearly related to the VIIRS/DNB reflectance measurements when all the data are considered (Fig. 4 b; p-value = 0.0016). The largest deviations from the linear response occurred in the Ship Canal and at the three Renton sites, where the spectrally weighted E d values are higher than expected for the corresponding VIIRS/DNB measurement. The highest nighttime downwelling irradiance intensities were recorded at 443 nm and 589 nm across both open water and nearshore regions. Since light at 443 nm falls outside the VIIRS/DNB spectral sensitivity and was excluded from the weighted E d calculation, the variability at this wavelength does not affect model performance. However, 589 nm, while within the VIIRS/DNB range, is measured with lower sensitivity (VIIRS/DNB sensitivity peaks at 700 nm). Consequently, variability at this wavelength may not be fully accounted for by VIIRS/DNB measurements. The model residuals were significantly correlated with the E d (589) peak (Fig. 4 c; r = 0.95, p-value = 3.20E-52), indicating that understanding the variability in the yellow portion of the spectrum is critical for reconciling differences between in-situ and satellite-based ALAN measurements. Light Attenuation, Hydroacoustic, and Discrete Depth Net Samples To understand the light environment experienced by juvenile salmon and their predators, we combined daytime K d (λ,z) with nocturnal surface nighttime light to model changes in spectral intensity through the water column at three sites representing variable lake environments (Fig. 5 ). The ‘Kirkland Open Water’ and ‘Renton Open Water’ sites were selected for their proximity to concurrent hydroacoustic and midwater bar trawl measurements, which were used to assess the depth distribution of planktivores, including juvenile salmon species, across the lake. Planktivores were caught in nets fishing at approximately 15 m, a depth within the thermocline. However, acoustic data identified the highest proportion of planktivore targets (all planktivores excluding transparent sub-yearling longfin smelt) to be different at each site: 21 m at ‘Renton Open Water’ and 31 m at ‘Kirkland Open Water’. Planktivores at the shallow southern open water site experienced light levels 28 times brighter than fish at the northern open water site. At the ‘Ship Canal’, the brightest and shallowest of the three sites, fish near the bottom (8 m) are exposed to light levels 65 times greater than fish at 21m at the southern site. Discussion Although our analysis revealed significant increasing trends in ALAN across all open water ROIs, nearly all nearshore ROIs showed either no trend or a significant negative trend. This decoupling in light trends between nearshore and limnetic regions suggests that local mitigation efforts targeting nearshore light pollution may not effectively address the broader impacts of distant light sources on open water environments. Consequently, efforts to mitigate the impact of light pollution on the biota of Lake Washington must also consider the effects of light from distant sources. A significant relationship was found between monthly cloud cover and VIIRS/DNB reflectance, further demonstrating the role of clouds in amplifying light in the limnetic zone 20 . The effect of skyglow is particularly pronounced during the winter months, when both cloud cover and VIIRS/DNB values peak. Skyglow impacts are greater in open water areas, as overall light levels are lower in the middle of the lake. Comparing VIIRS/DNB data with in-situ spectrally-resolved nighttime light measurements revealed important discrepancies in quantifying ALAN from satellite imagery. VIIRS/DNB consistently underestimated the light intensity at the nearshore sites due to several factors, including pixel resolution, strong directionality of the light field, and the spectral quality of LEDs 21 . Nearshore pixels, which span water, shoreline, and terrestrial habitats, exhibit high variability within each pixel. Our sampling, conducted at a single point within each pixel, did not reflect the average value measured by VIIRS/DNB. Further exacerbating this issue, Li et al. ( 2019 ) demonstrated that VIIRS/DNB measurements are highly sensitive to the direction, or angle, of the light source 22 . In-situ measured surface light consistently peaked in the blue channel (443 nm), a wavelength pronounced in LED lighting but not captured by VIIRS/DNB. Hung et al. ( 2021 ) measured sky brightness before and after a streetlight LED retrofit using a photometrically calibrated camera system 21 and found despite a decrease in measured VIIRS/DNB values, the retrofit caused skyglow to become brighter and extend higher in the sky. The authors attributed this discrepancy to factors including a significant increase in light emitted at wavelengths shorter than 500 nm. Thus, knowledge of the spectral quality of ALAN, which is not fully captured by VIIRS/DNB, is critical for accurately assessing the actual light experienced by fish and understanding its trophic impacts. Fish must balance tradeoffs in the water column, evident in their depth distribution within open water habitats. Fish adjust their depth to optimize their environment, balancing temperature limitations, food availability, and predation risk 23 . Hydroacoustic surveys revealed that fish were concentrated at shallower depths at the ‘Renton Open Water’ site (21 m), where the maximum depth is 33 m, whereas at ‘Kirkland Open Water,’ they were observed in high densities at 31m (50 m total depth). Planktivores in Renton reside suspended off the benthos to avoid predation by benthic piscivores 24 , allowing them to access the zooplankton community, which is densest in the 0–20 m layer. However, this depth choice also exposes these fish to bright water conditions, increasing their vulnerability to foraging pelagic predators. At ‘Kirkland Open Water’, the ambient light level at 31 m was 28 times lower than at 21 m in ‘Renton Open Water,’ effectively reducing pelagic predator effectiveness by approximately 784-fold 14 . This mismatch between foraging opportunities and prey availability could be hindering fish growth. These contrasts in depth distribution and light exposure highlight the significant impact of local conditions on predation risk, illustrating how light serves as a compounding stressor that can profoundly affect the survival of juvenile salmon. ALAN particularly impacts juvenile salmon at times of migration. The Ship Canal, one of the brightest regions identified, serves as the sole migration route for anadromous salmon from the Lake Washington basin to the marine waters of Puget Sound and ultimately the Pacific Ocean. Here, the combination of high surface light, relatively shallow depths (~ 9 m), and confined channel widths poses a significant threat to salmon survival, as increased light levels enhance piscivore effectiveness under bright ambient conditions and increase visibility of juvenile salmon to invasive “sit and wait” predators like smallmouth and largemouth bass. During outmigration, half of the diet of a smallmouth bass can be attributed to migrating salmon smolts 25 . Shallow depths and confined widths of the channel provide no reprieve from the light experienced by migrating salmon. Comparing light intensities between the ‘Renton Open Water' site at 21 m, the depth at which most planktivores were observed, and 8 m at the ‘Ship Canal’ site, the darkest conditions a smolt would experience in the Ship Canal, reveals significant differences. The ambient light at the ‘Ship Canal’ site is 65 times brighter than that at 21 m at ‘Renton Open Water’, a staggering contrast given that the effect of light on predation risk is squared 14 . These artificial nighttime light conditions exacerbate predation vulnerability inherent in transiting through a shallow channel with abundant predators. Urban waterbodies are significantly impacted by nearshore and regional light sources which are brightening each year. Understanding factors affecting this variability in light intensity and quality is imperative to create effective mitigation strategies. When considering the predation vulnerability and light relationship 14 , reducing light intensity results in a squared reduction effect, providing an encouraging incentive to reduce ambient light to reduce pressure on the local fish community. Each contributing light source in the region can make easy changes to reduce light emissions. For instance, adding motion sensors or timers to security lights would reduce residential and industrial light pollution during off hours. Municipal efforts like reducing light intensity or altering the spectral composition of streetlights (city or personally maintained) would additionally decrease ambient light both in the dense urban centers and outside the city bounds 26 . Vegetative cover along migration corridors and shorelines adjacent to tributaries can help to block ALAN from the water. Focusing light to where and when it is needed will help reduce the impact of ALAN on urban waterways and fish species. As light becomes more accessible, thoughtful measures will be necessary to limit the effect on urban ecosystems. Methods Regions of Interest Thirty regions of interest (ROIs) were selected with the aim of isolating direct, distinct sources of ALAN affecting Lake Washington (Fig. 1 ). Each ROI was strategically delineated to encompass key areas of potential light pollution. For example, prominent structures such as the bridges spanning Lake Washington, namely SR-520 and I-90, were individually identified and assessed for their light emissions. The Ship Canal was also delineated and evaluated. C-OPS Surface and In-water Irradiance Measurements A C-OPS downwelling surface irradiance reference sensor was used to collect nighttime E d (λ,0 + ) (µW cm -2 nm -1 ) measurements on August 5–7 & 14, 2024. Measurements occurred after the end of astronomical twilight when the moon was beneath the horizon. Cloud cover remained less than 20% for all sampling periods. The C-OPS measures irradiance across 19 channels (380, 395, 412, 443, 465, 490, 510, 532, 555, 560, 565, 589, 625, 665, 670, 683, 694, and 710 nm) and over the spectral range of Photosynthetic Active Radiation (PAR), at a frequency of 150 Hz. The instrument package was mounted to a height above the highest point on the boat to ensure that the sensor was not obstructed by the boat shadow and all navigation lights were turned off. Sensors were additionally oriented the same, with respect to the boat and the shoreline, for every measurement. Data acquisition was conducted at 10 Hz, with frames ( i.e. , raw spectra) averaged over 10-second intervals, for a total of one minute, to maximize the signal-to-noise ratio. Instrument noise, determined from the dark spectrum measurement nearest in time, was subtracted from each spectrum, and data with pitch or roll exceeding 5º were excluded from the analysis. The dark spectra collected over the 1-minute period were averaged to produce a site-specific spectrum. Dark spectra were collected about every hour. We calculated a single weighted value from the C-OPS ALAN spectra to compare the nighttime E d (λ,0 + ) measurements to VIIRS/DNB reflectance (nW cm -2 sr -1 ). To do this, we first created a digitized VIIRS/DNB spectral response curve (scaled from 0–1) using DataThief 27 , which extracts data points from a graph. Biospherical Instruments provides with the C-OPS a spectral response for every channel scaled from 0–1. For every C-OPS channel, we 1) interpolated the in-situ response to match the VIIRS/DNB wavelength range (Matlab v2021; ’interp1’,”linear”)(MathWorks, Inc., Natick, MA), 2) calculated a weighted response by multiplying the interpolated in-situ response by the VIIRS/DNB response, and 3) calculated a weighted response by integrating the area under the curve over the VIIRS/DNB wavelength range (‘trapz’). C-OPS spectra were multiplied by this response curve, for each channel, then the values at each channel were summed to produce the final weighted E d value for comparison to VIIRS/DNB reflectance. Profiles of daytime downwelling irradiance (E d (λ,z)) were collected at seven sites on July 24, 2024 in cloudless conditions using the C-OPS profiler. Three profiles were collected at each site. Instrument noise, determined from the average of the dark measurements, was first removed from each frame, and data with pitch or roll exceeding 5º were excluded from the analysis. Next, a 50-cell running mean was applied to each profile to smooth the data. To average the three profiles collected at each site, each profile was interpolated (‘interp1’, “spline”) from the minimum to maximum depths at 0.5-m depth intervals. Downwelling diffuse light attenuation rates (K d (λ; m -1) ) were calculated for every wavelength and depth (z) interval as: $$\:{K}_{d}(z,\lambda\:)=\frac{1}{{z}_{2}-{z}_{1}}*ln\frac{{E}_{d}(\lambda\:,{z}_{1})}{{E}_{d}(\lambda\:,{z}_{2})}$$ 1 . These measurements were used to model ALAN through the water column, focusing on the areas where juvenile salmon are seasonally observed in the greatest concentrations. To account for Fresnel reflection and transmittance across the air-water interface, we multiplied surface E d (λ) measurements by 0.97 28 . VIIRS/DNB Data Daily VIIRS/DNB data were obtained from the Earth Observation Group (EOG) Colorado School of Mines web portal ( https://eogdata.mines.edu/nighttime_light/nightly/rade9d/?C=N;O=D ; accessed August 8 & 15, 2024). These data were spatially aligned with E d (λ,0 + ) measurements, with sampling sites located at each pixel center. This allowed us to understand the relationship between VIIRS/DNB reflectance, which measures light from 500–900 nm (peak sensitivity at 700 nm) in radiance units at 500 m spatial resolution, and C-OPS irradiance, which is spectrally sensitive in the visible portion of the electromagnetic spectrum relevant for fish vision and behavior. We obtained monthly VIIRS/DNB composites from the EOG College of Mines web portal ( https://eogdata.mines.edu/nighttime_light/monthly/v10/ ; accessed March 29, 2022 & April 5, 2024). We used the “vcmsl” version, which includes pixels with reflectance values that have undergone the stray-light correction procedure. This allowed for coverage of our region over the full annual cycle. Stray light-corrected image files are not available for the 2012–2013 period, so these years were not included in our analyses. Cloud Cover We retrieved hourly cloud cover observations spanning from 2022 to 2023 from the Renton Municipal Airport ( https://mesonet.agron.iastate.edu/request/download.phtml?network=WA_ASOS ; accessed on February 13, 2024). The data categorize cloud cover as CLR (clear skies), FEW (few clouds), SCT (scattered clouds), BKN (broken clouds), or OVC (overcast conditions). To facilitate analysis, we assigned numerical values ranging from 0 to 4, with 0 representing clear skies and 4 indicating overcast conditions. Statistical Methods The monthly composites were binned to focus on the summer season (June-September) for trend analysis, as this period represents the critical rearing stage for juvenile salmon species. This approach effectively isolates the key period of interest within the annual cycle. During winter, fish are minimally influenced by ALAN due to reduced activity and a tendency to inhabit deeper, darker layers of the water column. Moreover, winter ALAN levels are elevated due to (1) increased contributions from sources such as holiday lighting and (2) greater scattering caused by cloud cover. To evaluate trends from 2014 to 2023, we applied a Generalized Linear Regression Model (‘fitglm’,”linear”). Reflectance values were summed within each ROI before analysis. This regression method was also used to evaluate the relationship between L VIIRS and spectrally-weighted E d . The Pearson correlation coefficient was calculated using the ‘corr’ function. Additionally, we used the ‘fitnlm’ function to fit an exponential model to the monthly cloud cover and L VIIRS datasets. A significance level of ⍺ = 0.05 was applied to all analyses. In-situ Fish Collection We deployed the midwater bar trawl net (3m wide x 7m deep x 18m long) at discrete depths informed by concurrent hydroacoustic measurements (Biosonics 200kHz 7° split-unit DTX) to target depths with high fish density. The bar trawl net has been proven to effectively capture fish of sizes 20–180 mm and is the standard sampling method to complement hydroacoustic assessment measurements for limnetic planktivores. We compared planktivore depth distribution in Renton and Kirkland, two sites with distinct light conditions. The third site, located in the Ship Canal, was chosen to represent the conditions experienced by all nocturnally out-migrating salmon smolts. Declarations Handling of vertebrates was conducted under the auspices of the Institutional Animal Care and Use Committee of the U.S. Geological Survey, Western Fisheries Research Center IACUC protocols #2008-57. Competing Interest Statement: The authors declare no competing interests. Author Contribution Both authors conceived the study. J.S. wrote the MATLAB scripts and performed all statistical analyses. T.C. collected the nighttime data. Both authors collected the daytime data. J.S. wrote the initial draft of the paper and obtained valuable contributions from T.C. Acknowledgement We thank Dr David Beauchamp for his invaluable contributions to this research project and Luke Valleli for his help with the data collection. The material is based on work supported by the Washington WRIA 8 Salmon Recovery Group (grant # 4.8.22.009 ) and the National Science Foundation Graduate Research Fellowship under Grant No. DGE-2140004. King County WRIA8 Cooperative Watershed Management Grant Program, and the Washington State Legislature provided funding through the King County Flood Control District. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government. Data Availability Scientific information and data developed as a result of this proposal are subject to applicable USGS Fundamental Science Practices (FSP) review, approval, and release requirements, which are available in Survey Manual Chapter (SMC) 502.4, Fundamental Science Practices: Review, Approval, and Release of Information Products. The USGS is required to provide timely public access to the results of scientific information and data that does not contain sensitive protected information. Data and associated metadata will be open format and publicly accessible in the USGS ScienceBase repository. The data and metadata will also be open access and machine readable in accordance with USGS FSP requirements available in SMC 502.7, Fundamental Science Practices: Metadata for USGS Scientific Information Products Including Data and SMC 502.8, Fundamental Science Practices: Review and Approval of Scientific Data for Release. References Jechow, A. & Hölker, F. How dark is a river? Artificial light at night in aquatic systems and the need for comprehensive night-time light measurements. Wiley Interdisciplinary Reviews: Water , 6 (6), e1388 (2019). Morgan-Taylor, M. Regulating light pollution: more than just the night sky. Science 380 (6650), 1118–1120 (2023). Hölker, F. et al. The dark side of light: a transdisciplinary research agenda for light pollution policy. Ecol. Soc. , 15 (4) (2010). Kyba, C. C. et al. Artificially lit surface of Earth at night increasing in radiance and extent. Sci. Adv. 3 (11), e1701528 (2017). Kyba, C. C., Ruhtz, T., Fischer, J. & Hölker, F. Cloud coverage acts as an amplifier for ecological light pollution in urban ecosystems. PloS one , 6 (3), e17307 (2011). Dyer, A. et al. Insect communities under skyglow: diffuse night-time illuminance induces spatio-temporal shifts in movement and predation. Philosophical Trans. Royal Soc. B . 378 (1892), 20220359 (2023). Hölker, F., Jechow, A., Schroer, S., Tockner, K. & Gessner, M. O. Light pollution of freshwater ecosystems: principles, ecological impacts and remedies. Philosophical Trans. Royal Soc. B . 378 (1892), 20220360 (2023). Kummu, M., De Moel, H., Ward, P. J. & Varis, O. How close do we live to water? A global analysis of population distance to freshwater bodies. PloS one , 6(6), p.e20578 (2011). Mobley, C. D. et al. Comparison of numerical models for computing underwater light fields. Appl. Opt. 32 (36), 7484–7504 (1993). Scheuerell, M. D. & Schindler, D. E. Diel vertical migration by juvenile sockeye salmon: empirical evidence for the antipredation window. Ecology 84 (7), 1713–1720 (2023). Riley, W. D., Bendall, B., Ives, M. J., Edmonds, N. J. & Maxwell, D. L. Street lighting disrupts the diel migratory pattern of wild Atlantic salmon, Salmo salar L., smolts leaving their natal stream. Aquaculture 330 , 74–81 (2012). Becker, A., Whitfield, A. K., Cowley, P. D., Järnegren, J. & Næsje, T. F. Potential effects of artificial light associated with anthropogenic infrastructure on the abundance and foraging behaviour of estuary-associated fishes. J. Appl. Ecol. 50 (1), 43–50 (2013). Tabor, R. A. et al. Predation of juvenile Chinook salmon by predatory fishes in three areas of the Lake Washington basin (US Fish and Wildlife Service, Western Washington Fish and Wildlife Office, 2004). Mazur, M. M. & Beauchamp, D. A. Linking piscivory to spatial–temporal distributions of pelagic prey fishes with a visual foraging model. J. Fish. Biol. 69 (1), 151–175 (2006). Beauchamp, D. A. Spatial and temporal dynamics of piscivory: implications for food web stability and the transparency of Lake Washington. Lake Reserv. Manag . 9 (1), 151–154 (1994). Tabor, R. A. et al. Smallmouth bass and largemouth bass predation on juvenile Chinook salmon and other salmonids in the Lake Washington basin. N Am. J. Fish. Manag . 27 (4), 1174–1188 (2007). Kyba, C. C., Ruhtz, T., Fischer, J. & Hölker, F. Red is the new black: how the colour of urban skyglow varies with cloud cover. Mon Not R Astron. Soc. 425 (1), 701–708 (2012). Cinzano, P. Night sky photometry with sky quality meter. ISTIL Int. Rep. , 9 (1) (2005). Tamir, R., Lerner, A., Haspel, C., Dubinsky, Z. & Iluz, D. The spectral and spatial distribution of light pollution in the waters of the northern Gulf of Aqaba (Eilat). Sci. Rep. 7 (1), 42329 (2017). Kyba, C. C., Ruhtz, T., Fischer, J. & Hölker, F. Cloud coverage acts as an amplifier for ecological light pollution in urban ecosystems. PloS one , 6 (3), e17307 (2011). Hung, L. W., Anderson, S. J., Pipkin, A. & Fristrup, K. Changes in night sky brightness after a countywide LED retrofit. J. Environ. Manage. 292 , 112776 (2021). Li, X. et al. Anisotropic characteristic of artificial light at night–Systematic investigation with VIIRS DNB multi-temporal observations. Remote Sens. Environ. 233 , 111357 (2019). Hansen, A. G., Beauchamp, D. A. & Baldwin, C. M. Environmental constraints on piscivory: insights from linking ultrasonic telemetry to a visual foraging model for cutthroat trout. Trans. Am. Fish. Soc. 142 (1), 300–316 (2013). Tabor, R. A., Warner, E. J., Fresh, K. L., Footen, B. A. & Chan, J. R. Ontogenetic diet shifts of prickly sculpin in the Lake Washington basin, Washington. Trans. Am. Fish. Soc. 136 (6), 1801–1813 (2007). Tabor, R. A. et al. Smallmouth bass and largemouth bass predation on juvenile Chinook salmon and other salmonids in the Lake Washington basin. North Am. J. Fish. Manag. 27 (4), 1174–1188 (2007). Barentine, J. C. et al. Recovering the city street lighting fraction from skyglow measurements in a large-scale municipal dimming experiment. J. Quant. Spectrosc. Radiative Transf. 253 , 107120 (2020). Tummers, B. & DataThiefIII (2006). https://datathief.org/ Doxaran, D. et al. Optical characterisation of suspended particles in the Mackenzie River plume (Canadian Arctic Ocean) and implications for ocean colour remote sensing. Biogeosciences 9 (8), 3213–3229 (2012). Additional Declarations No competing interests reported. Supplementary Files schuliencodenaturecommsupplement.pdf Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6406457","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":443301835,"identity":"acff2477-b59e-4ff3-96d0-4376ea227980","order_by":0,"name":"Jennifer Schulien","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2klEQVRIiWNgGAWjYBACxgYgkWDDwMDPwMDGwFBAtJY0BgbJBpAWA6LtAmoxOECsFub23mMPHiTY5BufP/zsMY+BnZw5A/PDRzfwOaznXLpBQkKa5bYbaebGPAbJxpYNbMbGOfi0zMgxk0j8cdjA7AYPmzSPwYHEDQeADLxa5r8xk0hI+G9g3H+GWC0zeEBaDhgYMOQQq6UnB6Ql2UDiRpqZ5BygXwwOE/CLYfsZM8kfCXYG/P2Hn0m8qbCTMzje/PAxXi0NGELMeJSDgDwB+VEwCkbBKBgFDAwA8nlEf/E+eQ0AAAAASUVORK5CYII=","orcid":"","institution":"Schulien Consulting","correspondingAuthor":true,"prefix":"","firstName":"Jennifer","middleName":"","lastName":"Schulien","suffix":""},{"id":443301836,"identity":"2ee82709-1e14-476c-ba40-d4726e7dfacf","order_by":1,"name":"Tessa Code","email":"","orcid":"","institution":"U.S. Geological Survey Western Fisheries Research Center","correspondingAuthor":false,"prefix":"","firstName":"Tessa","middleName":"","lastName":"Code","suffix":""}],"badges":[],"createdAt":"2025-04-08 21:53:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6406457/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6406457/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":83355220,"identity":"896c0a3f-a819-4444-86de-a09dcc448974","added_by":"auto","created_at":"2025-05-23 15:02:42","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":222619,"visible":true,"origin":"","legend":"\u003cp\u003ea) Map of the Pacific Northwest with the study area, Lake Washington, WA, USA, outlined. b) Log-transformed VIIRS/DNB nighttime reflectance from August 6, 2024 mapped for Lake Washington and the Seattle area with the 30 regions of interest (ROIs) outlined. Sites where spectral surface downwelling irradiance (E\u003csub\u003ed\u003c/sub\u003e(λ,0\u003csup\u003e+\u003c/sup\u003e)) was measured are shown in the corresponding colored markers. The ‘Renton Open Water’, ‘Kirkland Open Water’, and ‘Ship Canal’ sites are identified with a black circle.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6406457/v1/abb220fd027328e4d7f19e10.png"},{"id":83355229,"identity":"3da16b94-d12b-4760-a6c7-b7067296eea2","added_by":"auto","created_at":"2025-05-23 15:02:42","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":476745,"visible":true,"origin":"","legend":"\u003cp\u003eMay 2014 monthly VIIRS/DNB nighttime reflectance composite mapped for the study region. Significant annual trends from 2014 through 2023 for the summer period (June-September) are shown. In the red shaded regions, the significant trend is positive, while negative trends are shown in blue. Regions where no significant trend was observed are shown in cyan.\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6406457/v1/5b97c26e55cb282cdb96e275.jpeg"},{"id":83356836,"identity":"096e30f3-3ec1-4cfe-9309-5d9b0b91b278","added_by":"auto","created_at":"2025-05-23 15:18:42","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":174545,"visible":true,"origin":"","legend":"\u003cp\u003eMonthly cloud cover and VIIRS/DNB nighttime light.\u003cstrong\u003e \u003c/strong\u003ea) Monthly averages of hourly cloud observations, shown in black circles, for the Middle pelagic middle (Mid pel mid) ROI identified in Figure 1. The averaged monthly VIIRS/DNB reflectance values are shown in blue squares. b) Scatterplot of the averaged cloud cover and VIIRS/DNB reflectance for the same ROI with the fit exponential model shown in red. The R\u003csup\u003e2\u003c/sup\u003e value is listed.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-6406457/v1/f1c01986533f0eae3d9ac46c.png"},{"id":83356297,"identity":"9eb6bb31-c371-4e96-a9d9-eeb9cb296d3b","added_by":"auto","created_at":"2025-05-23 15:10:42","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":447818,"visible":true,"origin":"","legend":"\u003cp\u003ea) Spectral surface downwelling irradiance at 589 nm (E\u003csub\u003ed\u003c/sub\u003e(589)) across Lake Washington and in the Ship Canal are shown. Seven sites are marked with red stars. These sites have irradiance levels brighter than 10x10\u003csup\u003e-5 \u003c/sup\u003eµW cm\u003csup\u003e-2\u003c/sup\u003e nm\u003csup\u003e-1\u003c/sup\u003e. b) Scatterplot of VIIRS/DNB reflectance (L\u003csub\u003eVIIRS\u003c/sub\u003e) and the spectrally weighted E\u003csub\u003ed\u003c/sub\u003e\u003c/p\u003e","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6406457/v1/5ac29b72a8bc4ee734602a01.jpeg"},{"id":83355226,"identity":"5bcbac77-82da-48f9-80b9-30eeb2d12c8a","added_by":"auto","created_at":"2025-05-23 15:02:42","extension":"jpeg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":537474,"visible":true,"origin":"","legend":"\u003cp\u003eSpectral surface downwelling irradiance (E\u003csub\u003ed\u003c/sub\u003e(λ,0\u003csup\u003e+\u003c/sup\u003e)) is shown for the a) 'Kirkland Open Water’ site, b) ‘Renton Open Water’ site, and c) the ‘Ship Canal’ site. Spectral and vertically resolved downwelling irradiance, E\u003csub\u003ed\u003c/sub\u003e, is shown for d) ‘Kirkland Open Water’, e) ‘Renton Open Water’, and f) ‘Ship Canal’. Juvenile salmon are shown at the depths where they are observed in greatest abundances for the open water sites.\u003c/p\u003e","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6406457/v1/5f9a4f09d22caf6a2b8af510.jpeg"},{"id":91116506,"identity":"70454963-62c6-491d-a390-40cf11427a81","added_by":"auto","created_at":"2025-09-11 17:46:37","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2496566,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6406457/v1/e89ecaa7-2abe-4001-a24e-da75d83eb7a6.pdf"},{"id":83356295,"identity":"8ea1f53f-f984-46c8-96aa-acb282f9dc43","added_by":"auto","created_at":"2025-05-23 15:10:42","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":91583,"visible":true,"origin":"","legend":"","description":"","filename":"schuliencodenaturecommsupplement.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6406457/v1/e26c1bde95065acb1970a92a.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Shedding Light on the Role of Artificial Light at Night in Lake Washington, WA, USA","fulltext":[{"header":"Introduction","content":"\u003cp\u003eArtificial light at night (ALAN) has increasingly been recognized as a global threat to ecosystems\u003csup\u003e1,2\u003c/sup\u003e and continues to increase in both intensity and spatial coverage. Since the mid-20th century, ALAN has increased 6% per year on average (0\u0026ndash;20% range)\u003csup\u003e3\u003c/sup\u003e, with currently lit areas brightening at a rate of 2.2% annually\u003csup\u003e4\u003c/sup\u003e. ALAN includes both direct light and indirect light, or skyglow, which occurs when ALAN scatters in the atmosphere, extending beyond its original source\u003csup\u003e5\u003c/sup\u003e. Skyglow can have significant ecological impacts, as even small increases in light can significantly alter environmental conditions, cueing biological responses\u003csup\u003e6,7\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAs urbanization increases, with more than half of the global population living near freshwater shorelines\u003csup\u003e8\u003c/sup\u003e, rivers and lakes are increasingly exposed to light pollution. ALAN significantly alters both the quality and quantity of light in aquatic environments\u003csup\u003e9\u003c/sup\u003e, penetrating well below the surface and disrupting natural light gradients. Fish are particularly susceptible to changes in light intensity as they use daily light cycles to cue foraging and diel vertical migration\u003csup\u003e10\u003c/sup\u003e. Moreover, ALAN influences fish behavior and trophic relationships, as demonstrated in numerous laboratory and field studies\u003csup\u003e11\u0026ndash;13\u003c/sup\u003e. In Lake Washington, located adjacent to Seattle, WA, ALAN significantly affects predator\u0026ndash;prey dynamics, as observed for the piscivorous cutthroat trout (\u003cem\u003eOncorhynchus clarki\u003c/em\u003e)\u003csup\u003e14\u003c/sup\u003e and sculpins (\u003cem\u003eCottus\u003c/em\u003e sp.)\u003csup\u003e13\u003c/sup\u003e, both of which are voracious predators of juvenile salmon species\u003csup\u003e15,16\u003c/sup\u003e. Understanding the intricate interplay between ALAN, underwater light\u003c/p\u003e \u003cp\u003edynamics, and biota, is thus essential for identifying and mitigating both the direct and far-reaching effects of ALAN on aquatic ecosystems.\u003c/p\u003e \u003cp\u003eMeasuring spectral nighttime light presents significant challenges due to the high cost and operational complexity of the required sensors\u003csup\u003e1\u003c/sup\u003e. While broadband sensors are more affordable and widely used for nighttime light measurements, they come with limitations\u003csup\u003e17\u003c/sup\u003e. The widely used Sky Quality Meter (SQM), for example, reports measurements in the astronomical units of magnitudes per square arc second, a unit that is instrument specific\u003csup\u003e18\u003c/sup\u003e. Calibrated, spectrally discrete, radiometric units are necessary to address species-specific questions and to engage government agencies and the lighting industry. Tamir et al. (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) were the first to study in-situ spectral ALAN measurements in the Gulf of Aqaba using a profiling reflectance radiometer \u003csup\u003e19\u003c/sup\u003e. These measurements were used to quantify light intensity and spectra affecting offshore regions. Our study is the first assessment of ALAN in a freshwater system, shedding light on its spatial and spectral impacts on predator\u0026ndash;prey dynamics in an urban waterbody.\u003c/p\u003e \u003cp\u003eHere we examined spatiotemporal trends in ALAN in both the nearshore and open water regions of Lake Washington, WA (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) using both 1) Visible Infrared Imaging Radiometer Suite Day-Night Band (VIIRS/DNB) reflectance (L\u003csub\u003eVIIRS\u003c/sub\u003e) data and 2) surface downwelling irradiance (E\u003csub\u003ed\u003c/sub\u003e(λ,0\u003csup\u003e+\u003c/sup\u003e)) measurements at 18 wavelengths collected using a Compact Optical Profiling System (C-OPS; Biospherical Instruments Inc., San Diego, CA). We calculated nighttime light (from surface light and the downwelling diffuse light attenuation coefficient, K\u003csub\u003ed\u003c/sub\u003e(λ), measurements) at the depths where high fish densities were detected using hydroacoustic surveys and discrete depth trawl nets.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eInter-annual Nighttime Light Trends\u003c/h2\u003e \u003cp\u003eWe found significant trends over the past decade during the summer period (June-September) in 20 of the 30 regions of interest (ROIs) evaluated (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e; S.Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). ALAN increased significantly in every open water ROI. In contrast, in the nearshore ROIs, negative or no trends were observed, except for the northwest, north Mercer Island, and I-90 bridge ROIs, where positive trends were also observed. Interestingly, several ROIs with notably high average pixel reflectance, such as the Ship Canal, Renton, and Kirkland, exhibited primarily decreasing trends in ALAN.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eStatistical summary of the exponential model relating monthly cloud cover to VIIRS/DNB nighttime reflectance (N\u0026thinsp;=\u0026thinsp;12), including p-values, F-statistics, and adjusted R\u003csup\u003e2\u003c/sup\u003e values.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRegion of Interest\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eP-value\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eF-statistic\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eR\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNorthern pelagic\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e9.80E-10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e312\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.569\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMiddle pelagic north\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.52E-09\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e240\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.537\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSouthwest pelagic\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.80E-10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e378\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.0869\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMiddle pelagic middle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.96E-11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e687\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.614\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMiddle pelagic south\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e9.40E-10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e314\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.339\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSoutheast pelagic\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.53E-11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e611\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.553\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eMonthly Skyglow\u003c/h3\u003e\n\u003cp\u003eThe seasonal pattern in cloud cover means (based on hourly observations from Renton Municipal Airport) paralleled that of VIIRS/DNB reflectance spanning from 2022 through 2023 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). A significant exponential relationship for all of the six ROIs evaluated was observed (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eSpatial Variability in Nighttime Irradiance\u003c/h3\u003e\n\u003cp\u003eNighttime downwelling irradiance at 589 nm, E\u003csub\u003ed\u003c/sub\u003e(589), generally increased from north to south in the lake (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea). Four of the sites in the Ship Canal and three in Renton exhibited significantly higher E\u003csub\u003ed\u003c/sub\u003e(589) values compared to the other site locations (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea). The Renton site just offshore of Gene Coulon Park was brightest, with E\u003csub\u003ed\u003c/sub\u003e(589) equal to 8.0x10\u003csup\u003e-4\u003c/sup\u003e \u0026micro;W cm\u003csup\u003e-2\u003c/sup\u003e nm\u003csup\u003e-1\u003c/sup\u003e. This is 2\u0026ndash;4 times greater than E\u003csub\u003ed\u003c/sub\u003e(589) measured at the two starred sites closer to the Cedar River mouth, (3.9x10\u003csup\u003e-4\u003c/sup\u003e and 2.1x10\u003csup\u003e-4\u003c/sup\u003e \u0026micro;W cm\u003csup\u003e-2\u003c/sup\u003e nm\u003csup\u003e-1\u003c/sup\u003e). For reference, this is about an order of magnitude brighter than the nearshore region north of Gene Coulon Park where E\u003csub\u003ed\u003c/sub\u003e(589) is ~\u0026thinsp;4x10\u003csup\u003e-5\u003c/sup\u003e \u0026micro;W cm\u003csup\u003e-2\u003c/sup\u003e nm\u003csup\u003e-1\u003c/sup\u003e. E\u003csub\u003ed\u003c/sub\u003e(589) in the Ship Canal at the four starred sites ranged from 1.4x10\u003csup\u003e-4\u003c/sup\u003e to 5.4x10\u003csup\u003e-4\u003c/sup\u003e \u0026micro;W cm\u003csup\u003e-2\u003c/sup\u003e nm\u003csup\u003e-1\u003c/sup\u003e.\u003c/p\u003e\n\u003ch3\u003eComparison of Satellite and In-situ Nighttime Light Measurements\u003c/h3\u003e\n\u003cp\u003eConsistent with the spatial correspondence, the brightest regions of the lake identified using the C-OPS reference sensor were significantly linearly related to the VIIRS/DNB reflectance measurements when all the data are considered (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb; p-value\u0026thinsp;=\u0026thinsp;0.0016). The largest deviations from the linear response occurred in the Ship Canal and at the three Renton sites, where the spectrally weighted E\u003csub\u003ed\u003c/sub\u003e values are higher than expected for the corresponding VIIRS/DNB measurement.\u003c/p\u003e \u003cp\u003eThe highest nighttime downwelling irradiance intensities were recorded at 443 nm and 589 nm across both open water and nearshore regions. Since light at 443 nm falls outside the VIIRS/DNB spectral sensitivity and was excluded from the weighted E\u003csub\u003ed\u003c/sub\u003e calculation, the variability at this wavelength does not affect model performance. However, 589 nm, while within the VIIRS/DNB range, is measured with lower sensitivity (VIIRS/DNB sensitivity peaks at 700 nm). Consequently, variability at this wavelength may not be fully accounted for by VIIRS/DNB measurements. The model residuals were significantly correlated with the E\u003csub\u003ed\u003c/sub\u003e(589) peak (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec; r\u0026thinsp;=\u0026thinsp;0.95, p-value\u0026thinsp;=\u0026thinsp;3.20E-52), indicating that understanding the variability in the yellow portion of the spectrum is critical for reconciling differences between in-situ and satellite-based ALAN measurements.\u003c/p\u003e\n\u003ch3\u003eLight Attenuation, Hydroacoustic, and Discrete Depth Net Samples\u003c/h3\u003e\n\u003cp\u003eTo understand the light environment experienced by juvenile salmon and their predators, we combined daytime K\u003csub\u003ed\u003c/sub\u003e(λ,z) with nocturnal surface nighttime light to model changes in spectral intensity through the water column at three sites representing variable lake environments (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). The \u0026lsquo;Kirkland Open Water\u0026rsquo; and \u0026lsquo;Renton Open Water\u0026rsquo; sites were selected for their proximity to concurrent hydroacoustic and midwater bar trawl measurements, which were used to assess the depth distribution of planktivores, including juvenile salmon species, across the lake. Planktivores were caught in nets fishing at approximately 15 m, a depth within the thermocline. However, acoustic data identified the highest proportion of planktivore targets (all planktivores excluding transparent sub-yearling longfin smelt) to be different at each site: 21 m at \u0026lsquo;Renton Open Water\u0026rsquo; and 31 m at \u0026lsquo;Kirkland Open Water\u0026rsquo;. Planktivores at the shallow southern open\u003c/p\u003e \u003cp\u003ewater site experienced light levels 28 times brighter than fish at the northern open water site. At the \u0026lsquo;Ship Canal\u0026rsquo;, the brightest and shallowest of the three sites, fish near the bottom (8 m) are exposed to light levels 65 times greater than fish at 21m at the southern site.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eAlthough our analysis revealed significant increasing trends in ALAN across all open water ROIs, nearly all nearshore ROIs showed either no trend or a significant negative trend. This decoupling in light trends between nearshore and limnetic regions suggests that local mitigation efforts targeting nearshore light pollution may not effectively address the broader impacts of distant light sources on open water environments. Consequently, efforts to mitigate the impact of light pollution on the biota of Lake Washington must also consider the effects of light from distant sources. A significant relationship was found between monthly cloud cover and VIIRS/DNB reflectance, further demonstrating the role of clouds in amplifying light in the limnetic zone\u003csup\u003e20\u003c/sup\u003e. The effect of skyglow is particularly pronounced during the winter months, when both cloud cover and VIIRS/DNB values peak. Skyglow impacts are greater in open water areas, as overall light levels are lower in the middle of the lake.\u003c/p\u003e \u003cp\u003eComparing VIIRS/DNB data with in-situ spectrally-resolved nighttime light measurements revealed important discrepancies in quantifying ALAN from satellite imagery. VIIRS/DNB consistently underestimated the light intensity at the nearshore sites due to several factors, including pixel resolution, strong directionality of the light field, and the spectral quality of LEDs\u003csup\u003e21\u003c/sup\u003e. Nearshore pixels, which span water, shoreline, and terrestrial habitats, exhibit high variability within each pixel. Our sampling, conducted at a single point within each pixel, did not reflect the average value measured by VIIRS/DNB. Further exacerbating this issue, Li et al. (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) demonstrated that VIIRS/DNB measurements are highly sensitive to the direction, or angle, of the light source\u003csup\u003e22\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn-situ measured surface light consistently peaked in the blue channel (443 nm), a wavelength pronounced in LED lighting but not captured by VIIRS/DNB. Hung et al. (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) measured sky brightness before and after a streetlight LED retrofit using a photometrically calibrated camera system\u003csup\u003e21\u003c/sup\u003e and found despite a decrease in measured VIIRS/DNB values, the retrofit caused skyglow to become brighter and extend higher in the sky. The authors attributed this discrepancy to factors including a significant increase in light emitted at wavelengths shorter than 500 nm. Thus, knowledge of the spectral quality of ALAN, which is not fully captured by VIIRS/DNB, is critical for accurately assessing the actual light experienced by fish and understanding its trophic impacts.\u003c/p\u003e \u003cp\u003eFish must balance tradeoffs in the water column, evident in their depth distribution within open water habitats. Fish adjust their depth to optimize their environment, balancing temperature limitations, food availability, and predation risk\u003csup\u003e23\u003c/sup\u003e. Hydroacoustic surveys revealed that fish were concentrated at shallower depths at the \u0026lsquo;Renton Open Water\u0026rsquo; site (21 m), where the maximum depth is 33 m, whereas at \u0026lsquo;Kirkland Open Water,\u0026rsquo; they were observed in high densities at 31m (50 m total depth). Planktivores in Renton reside suspended off the benthos to avoid predation by benthic piscivores\u003csup\u003e24\u003c/sup\u003e, allowing them to access the zooplankton community, which is densest in the 0\u0026ndash;20 m layer. However, this depth choice also exposes these fish to bright water conditions, increasing their vulnerability to foraging pelagic predators. At \u0026lsquo;Kirkland Open Water\u0026rsquo;, the ambient light level at 31 m was 28 times lower than at 21 m in \u0026lsquo;Renton Open Water,\u0026rsquo; effectively reducing pelagic predator effectiveness by approximately 784-fold\u003csup\u003e14\u003c/sup\u003e. This mismatch between foraging opportunities and prey availability could be hindering fish growth. These contrasts in depth distribution and light exposure highlight the significant impact of local conditions on predation risk, illustrating how light serves as a compounding stressor that can profoundly affect the survival of juvenile salmon.\u003c/p\u003e \u003cp\u003eALAN particularly impacts juvenile salmon at times of migration. The Ship Canal, one of the brightest regions identified, serves as the sole migration route for anadromous salmon from the Lake Washington basin to the marine waters of Puget Sound and ultimately the Pacific Ocean. Here, the combination of high surface light, relatively shallow depths (~\u0026thinsp;9 m), and confined channel widths poses a significant threat to salmon survival, as increased light levels enhance piscivore effectiveness under bright ambient conditions and increase visibility of juvenile salmon to invasive \u0026ldquo;sit and wait\u0026rdquo; predators like smallmouth and largemouth bass. During outmigration, half of the diet of a smallmouth bass can be attributed to migrating salmon smolts\u003csup\u003e25\u003c/sup\u003e. Shallow depths and confined widths of the channel provide no reprieve from the light experienced by migrating salmon. Comparing light intensities between the \u0026lsquo;Renton Open Water' site at 21 m, the depth at which most planktivores were observed, and 8 m at the \u0026lsquo;Ship Canal\u0026rsquo; site, the darkest conditions a smolt would experience in the Ship Canal, reveals significant differences. The ambient light at the \u0026lsquo;Ship Canal\u0026rsquo; site is 65 times brighter than that at 21 m at \u0026lsquo;Renton Open Water\u0026rsquo;, a staggering contrast given that the effect of light on predation risk is squared\u003csup\u003e14\u003c/sup\u003e. These artificial nighttime light conditions exacerbate predation vulnerability inherent in transiting through a shallow channel with abundant predators.\u003c/p\u003e \u003cp\u003eUrban waterbodies are significantly impacted by nearshore and regional light sources which are brightening each year. Understanding factors affecting this variability in light intensity and quality is imperative to create effective mitigation strategies. When considering the predation vulnerability and light relationship\u003csup\u003e14\u003c/sup\u003e, reducing light intensity results in a squared reduction effect, providing an encouraging incentive to reduce ambient light to reduce pressure on the local fish community. Each contributing light source in the region can make easy changes to reduce light emissions. For instance, adding motion sensors or timers to security lights would reduce residential and industrial light pollution during off hours. Municipal efforts like reducing light intensity or altering the spectral composition of streetlights (city or personally maintained) would additionally decrease ambient light both in the dense urban centers and outside the city bounds\u003csup\u003e26\u003c/sup\u003e. Vegetative cover along migration corridors and shorelines adjacent to tributaries can help to block ALAN from the water. Focusing light to where and when it is needed will help reduce the impact of ALAN on urban waterways and fish species. As light becomes more accessible, thoughtful measures will be necessary to limit the effect on urban ecosystems.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eRegions of Interest\u003c/h2\u003e \u003cp\u003eThirty regions of interest (ROIs) were selected with the aim of isolating direct, distinct sources of ALAN affecting Lake Washington (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Each ROI was strategically delineated to encompass key areas of potential light pollution. For example, prominent structures such as the bridges spanning Lake Washington, namely SR-520 and I-90, were individually identified and assessed for their light emissions. The Ship Canal was also delineated and evaluated.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eC-OPS Surface and In-water Irradiance Measurements\u003c/h2\u003e \u003cp\u003eA C-OPS downwelling surface irradiance reference sensor was used to collect nighttime E\u003csub\u003ed\u003c/sub\u003e(λ,0\u003csup\u003e+\u003c/sup\u003e) (\u0026micro;W cm\u003csup\u003e-2\u003c/sup\u003e nm\u003csup\u003e-1\u003c/sup\u003e) measurements on August 5\u0026ndash;7 \u0026amp; 14, 2024. Measurements occurred after the end of astronomical twilight when the moon was beneath the horizon. Cloud cover remained less than 20% for all sampling periods. The C-OPS measures irradiance across 19 channels (380, 395, 412, 443, 465, 490, 510, 532, 555, 560, 565, 589, 625, 665, 670, 683, 694, and 710 nm) and over the spectral range of Photosynthetic Active Radiation (PAR), at a frequency of 150 Hz. The instrument package was mounted to a height above the highest point on the boat to ensure that the sensor was not obstructed by the boat shadow and all navigation lights were turned off. Sensors were additionally oriented the same, with respect to the boat and the shoreline, for every measurement. Data acquisition was conducted at 10 Hz, with frames (\u003cem\u003ei.e.\u003c/em\u003e, raw spectra) averaged over 10-second intervals, for a total of one minute, to maximize the signal-to-noise ratio. Instrument noise, determined from the dark spectrum measurement nearest in time, was subtracted from each spectrum, and data with pitch or roll exceeding 5\u0026ordm; were excluded from the analysis. The dark spectra collected over the 1-minute period were averaged to produce a site-specific spectrum. Dark spectra were collected about every hour.\u003c/p\u003e \u003cp\u003eWe calculated a single weighted value from the C-OPS ALAN spectra to compare the nighttime E\u003csub\u003ed\u003c/sub\u003e(λ,0\u003csup\u003e+\u003c/sup\u003e) measurements to VIIRS/DNB reflectance (nW cm\u003csup\u003e-2\u003c/sup\u003e sr\u003csup\u003e-1\u003c/sup\u003e). To do this, we first created a digitized VIIRS/DNB spectral response curve (scaled from 0\u0026ndash;1) using DataThief\u003csup\u003e27\u003c/sup\u003e, which extracts data points from a graph. Biospherical Instruments provides with the C-OPS a spectral response for every channel scaled from 0\u0026ndash;1. For every C-OPS channel, we 1) interpolated the in-situ response to match the VIIRS/DNB wavelength range (Matlab v2021; \u0026rsquo;interp1\u0026rsquo;,\u0026rdquo;linear\u0026rdquo;)(MathWorks, Inc., Natick, MA), 2) calculated a weighted response by multiplying the interpolated in-situ response by the VIIRS/DNB response, and 3) calculated a weighted response by integrating the area under the curve over the VIIRS/DNB wavelength range (\u0026lsquo;trapz\u0026rsquo;). C-OPS spectra were multiplied by this response curve, for each channel, then the values at each channel were summed to produce the final weighted E\u003csub\u003ed\u003c/sub\u003e value for comparison to VIIRS/DNB reflectance.\u003c/p\u003e \u003cp\u003eProfiles of daytime downwelling irradiance (E\u003csub\u003ed\u003c/sub\u003e(λ,z)) were collected at seven sites on July 24, 2024 in cloudless conditions using the C-OPS profiler. Three profiles were collected at each site. Instrument noise, determined from the average of the dark measurements, was first removed from each frame, and data with pitch or roll exceeding 5\u0026ordm; were excluded from the analysis. Next, a 50-cell running mean was applied to each profile to smooth the data. To average the three profiles collected at each site, each profile was interpolated (\u0026lsquo;interp1\u0026rsquo;, \u0026ldquo;spline\u0026rdquo;) from the minimum to maximum depths at 0.5-m depth intervals. Downwelling diffuse light attenuation rates (K\u003csub\u003ed\u003c/sub\u003e(λ; m\u003csup\u003e-1)\u003c/sup\u003e) were calculated for every wavelength and depth (z) interval as:\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$$\\:{K}_{d}(z,\\lambda\\:)=\\frac{1}{{z}_{2}-{z}_{1}}*ln\\frac{{E}_{d}(\\lambda\\:,{z}_{1})}{{E}_{d}(\\lambda\\:,{z}_{2})}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\u003c/div\u003e.\u003c/p\u003e \u003cp\u003eThese measurements were used to model ALAN through the water column, focusing on the areas where juvenile salmon are seasonally observed in the greatest concentrations. To account for Fresnel reflection and transmittance across the air-water interface, we multiplied surface E\u003csub\u003ed\u003c/sub\u003e(λ) measurements by 0.97\u003csup\u003e28\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eVIIRS/DNB Data\u003c/h2\u003e \u003cp\u003eDaily VIIRS/DNB data were obtained from the Earth Observation Group (EOG) Colorado School of Mines web portal (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://eogdata.mines.edu/nighttime_light/nightly/rade9d/?C=N;O=D\u003c/span\u003e\u003cspan address=\"https://eogdata.mines.edu/nighttime_light/nightly/rade9d/?C=N;O=D\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e; accessed August 8 \u0026amp; 15, 2024). These data were spatially aligned with E\u003csub\u003ed\u003c/sub\u003e(λ,0\u003csup\u003e+\u003c/sup\u003e) measurements, with sampling sites located at each pixel center. This allowed us to understand the relationship between VIIRS/DNB reflectance, which measures light from 500\u0026ndash;900 nm (peak sensitivity at 700 nm) in radiance units at 500 m spatial resolution, and C-OPS irradiance, which is spectrally sensitive in the visible portion of the electromagnetic spectrum relevant for fish vision and behavior.\u003c/p\u003e \u003cp\u003eWe obtained monthly VIIRS/DNB composites from the EOG College of Mines web portal (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://eogdata.mines.edu/nighttime_light/monthly/v10/\u003c/span\u003e\u003cspan address=\"https://eogdata.mines.edu/nighttime_light/monthly/v10/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e ; accessed March 29, 2022 \u0026amp; April 5, 2024). We used the \u0026ldquo;vcmsl\u0026rdquo; version, which includes pixels with reflectance values that have undergone the stray-light correction procedure. This allowed for coverage of our region over the full annual cycle. Stray light-corrected image files are not available for the 2012\u0026ndash;2013 period, so these years were not included in our analyses.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eCloud Cover\u003c/h2\u003e \u003cp\u003eWe retrieved hourly cloud cover observations spanning from 2022 to 2023 from the Renton Municipal Airport (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://mesonet.agron.iastate.edu/request/download.phtml?network=WA_ASOS\u003c/span\u003e\u003cspan address=\"https://mesonet.agron.iastate.edu/request/download.phtml?network=WA_ASOS\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e; accessed on February 13, 2024). The data categorize cloud cover as CLR (clear skies), FEW (few clouds), SCT (scattered clouds), BKN (broken clouds), or OVC (overcast conditions). To facilitate analysis, we assigned numerical values ranging from 0 to 4, with 0 representing clear skies and 4 indicating overcast conditions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Methods\u003c/h2\u003e \u003cp\u003eThe monthly composites were binned to focus on the summer season (June-September) for trend analysis, as this period represents the critical rearing stage for juvenile salmon species. This approach effectively isolates the key period of interest within the annual cycle. During winter, fish are minimally influenced by ALAN due to reduced activity and a tendency to inhabit deeper, darker layers of the water column. Moreover, winter ALAN levels are elevated due to (1) increased contributions from sources such as holiday lighting and (2) greater scattering caused by cloud cover.\u003c/p\u003e \u003cp\u003eTo evaluate trends from 2014 to 2023, we applied a Generalized Linear Regression Model (\u0026lsquo;fitglm\u0026rsquo;,\u0026rdquo;linear\u0026rdquo;). Reflectance values were summed within each ROI before analysis. This regression method was also used to evaluate the relationship between L\u003csub\u003eVIIRS\u003c/sub\u003e and spectrally-weighted E\u003csub\u003ed\u003c/sub\u003e. The Pearson correlation coefficient was calculated using the \u0026lsquo;corr\u0026rsquo; function. Additionally, we used the \u0026lsquo;fitnlm\u0026rsquo; function to fit an exponential model to the monthly cloud cover and L\u003csub\u003eVIIRS\u003c/sub\u003e datasets. A significance level of ⍺ = 0.05 was applied to all analyses.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eIn-situ Fish Collection\u003c/h2\u003e \u003cp\u003eWe deployed the midwater bar trawl net (3m wide x 7m deep x 18m long) at discrete depths informed by concurrent hydroacoustic measurements (Biosonics 200kHz 7\u0026deg; split-unit DTX) to target depths with high fish density. The bar trawl net has been proven to effectively capture fish of sizes 20\u0026ndash;180 mm and is the standard sampling method to complement hydroacoustic assessment measurements for limnetic planktivores.\u003c/p\u003e \u003cp\u003eWe compared planktivore depth distribution in Renton and Kirkland, two sites with distinct light conditions. The third site, located in the Ship Canal, was chosen to represent the conditions experienced by all nocturnally out-migrating salmon smolts.\u003c/p\u003e \u003c/div\u003e "},{"header":"Declarations","content":"\u003cp\u003eHandling of vertebrates was conducted under the auspices of the Institutional Animal Care and Use Committee of the U.S. Geological Survey, Western Fisheries Research Center IACUC protocols #2008-57.\u003c/p\u003e\n\u003ch2\u003eCompeting Interest Statement:\u003c/h2\u003e \u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e \u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eBoth authors conceived the study. J.S. wrote the MATLAB scripts and performed all statistical analyses. T.C. collected the nighttime data. Both authors collected the daytime data. J.S. wrote the initial draft of the paper and obtained valuable contributions from T.C.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe thank Dr David Beauchamp for his invaluable contributions to this research project and Luke Valleli for his help with the data collection. The material is based on work supported by the Washington WRIA 8 Salmon Recovery Group (grant # 4.8.22.009 ) and the National Science Foundation Graduate Research Fellowship under Grant No. DGE-2140004. King County WRIA8 Cooperative Watershed Management Grant Program, and the Washington State Legislature provided funding through the King County Flood Control District. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eScientific information and data developed as a result of this proposal are subject to applicable\u0026nbsp;USGS Fundamental Science Practices (FSP) review, approval, and release requirements, which are available in\u0026nbsp;Survey Manual Chapter (SMC) 502.4, Fundamental Science Practices: Review, Approval, and Release of Information Products. The USGS is required to provide timely public access to the results of scientific information and data that does not contain sensitive protected information. Data and associated metadata will be open format and publicly accessible in the USGS ScienceBase repository. The data and metadata will also be open access and machine readable in accordance with USGS FSP requirements available in\u0026nbsp;SMC 502.7, Fundamental Science Practices: Metadata for USGS Scientific Information Products Including\u0026nbsp;Data\u0026nbsp;and\u0026nbsp;SMC 502.8, Fundamental Science Practices: Review and Approval of Scientific\u0026nbsp;Data\u0026nbsp;for Release.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eJechow, A. \u0026amp; H\u0026ouml;lker, F. How dark is a river? Artificial light at night in aquatic systems and the need for comprehensive night-time light measurements. \u003cem\u003eWiley Interdisciplinary Reviews: Water\u003c/em\u003e, \u003cb\u003e6\u003c/b\u003e(6), e1388 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorgan-Taylor, M. Regulating light pollution: more than just the night sky. \u003cem\u003eScience\u003c/em\u003e \u003cb\u003e380\u003c/b\u003e (6650), 1118\u0026ndash;1120 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eH\u0026ouml;lker, F. et al. The dark side of light: a transdisciplinary research agenda for light pollution policy. \u003cem\u003eEcol. Soc.\u003c/em\u003e, \u003cb\u003e15\u003c/b\u003e(4) (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKyba, C. C. et al. Artificially lit surface of Earth at night increasing in radiance and extent. \u003cem\u003eSci. Adv.\u003c/em\u003e \u003cb\u003e3\u003c/b\u003e (11), e1701528 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKyba, C. C., Ruhtz, T., Fischer, J. \u0026amp; H\u0026ouml;lker, F. Cloud coverage acts as an amplifier for ecological light pollution in urban ecosystems. \u003cem\u003ePloS one\u003c/em\u003e, \u003cb\u003e6\u003c/b\u003e(3), e17307 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDyer, A. et al. Insect communities under skyglow: diffuse night-time illuminance induces spatio-temporal shifts in movement and predation. \u003cem\u003ePhilosophical Trans. Royal Soc. B\u003c/em\u003e. \u003cb\u003e378\u003c/b\u003e (1892), 20220359 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eH\u0026ouml;lker, F., Jechow, A., Schroer, S., Tockner, K. \u0026amp; Gessner, M. O. Light pollution of freshwater ecosystems: principles, ecological impacts and remedies. \u003cem\u003ePhilosophical Trans. Royal Soc. B\u003c/em\u003e. \u003cb\u003e378\u003c/b\u003e (1892), 20220360 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKummu, M., De Moel, H., Ward, P. J. \u0026amp; Varis, O. How close do we live to water? A global analysis of population distance to freshwater bodies. \u003cem\u003ePloS one\u003c/em\u003e, 6(6), p.e20578 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMobley, C. D. et al. Comparison of numerical models for computing underwater light fields. \u003cem\u003eAppl. Opt.\u003c/em\u003e \u003cb\u003e32\u003c/b\u003e (36), 7484\u0026ndash;7504 (1993).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eScheuerell, M. D. \u0026amp; Schindler, D. E. Diel vertical migration by juvenile sockeye salmon: empirical evidence for the antipredation window. \u003cem\u003eEcology\u003c/em\u003e \u003cb\u003e84\u003c/b\u003e (7), 1713\u0026ndash;1720 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRiley, W. D., Bendall, B., Ives, M. J., Edmonds, N. J. \u0026amp; Maxwell, D. L. Street lighting disrupts the diel migratory pattern of wild Atlantic salmon, Salmo salar L., smolts leaving their natal stream. \u003cem\u003eAquaculture\u003c/em\u003e \u003cb\u003e330\u003c/b\u003e, 74\u0026ndash;81 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBecker, A., Whitfield, A. K., Cowley, P. D., J\u0026auml;rnegren, J. \u0026amp; N\u0026aelig;sje, T. F. Potential effects of artificial light associated with anthropogenic infrastructure on the abundance and foraging behaviour of estuary-associated fishes. \u003cem\u003eJ. Appl. Ecol.\u003c/em\u003e \u003cb\u003e50\u003c/b\u003e (1), 43\u0026ndash;50 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTabor, R. A. et al. \u003cem\u003ePredation of juvenile Chinook salmon by predatory fishes in three areas of the Lake Washington basin\u003c/em\u003e (US Fish and Wildlife Service, Western Washington Fish and Wildlife Office, 2004).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMazur, M. M. \u0026amp; Beauchamp, D. A. Linking piscivory to spatial\u0026ndash;temporal distributions of pelagic prey fishes with a visual foraging model. \u003cem\u003eJ. Fish. Biol.\u003c/em\u003e \u003cb\u003e69\u003c/b\u003e (1), 151\u0026ndash;175 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBeauchamp, D. A. Spatial and temporal dynamics of piscivory: implications for food web stability and the transparency of Lake Washington. \u003cem\u003eLake Reserv. Manag\u003c/em\u003e. \u003cb\u003e9\u003c/b\u003e (1), 151\u0026ndash;154 (1994).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTabor, R. A. et al. Smallmouth bass and largemouth bass predation on juvenile Chinook salmon and other salmonids in the Lake Washington basin. \u003cem\u003eN Am. J. Fish. Manag\u003c/em\u003e. \u003cb\u003e27\u003c/b\u003e (4), 1174\u0026ndash;1188 (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKyba, C. C., Ruhtz, T., Fischer, J. \u0026amp; H\u0026ouml;lker, F. Red is the new black: how the colour of urban skyglow varies with cloud cover. \u003cem\u003eMon Not R Astron. Soc.\u003c/em\u003e \u003cb\u003e425\u003c/b\u003e (1), 701\u0026ndash;708 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCinzano, P. Night sky photometry with sky quality meter. \u003cem\u003eISTIL Int. Rep.\u003c/em\u003e, \u003cb\u003e9\u003c/b\u003e(1) (2005).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTamir, R., Lerner, A., Haspel, C., Dubinsky, Z. \u0026amp; Iluz, D. The spectral and spatial distribution of light pollution in the waters of the northern Gulf of Aqaba (Eilat). \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e7\u003c/b\u003e (1), 42329 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKyba, C. C., Ruhtz, T., Fischer, J. \u0026amp; H\u0026ouml;lker, F. Cloud coverage acts as an amplifier for ecological light pollution in urban ecosystems. \u003cem\u003ePloS one\u003c/em\u003e, \u003cb\u003e6\u003c/b\u003e(3), e17307 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHung, L. W., Anderson, S. J., Pipkin, A. \u0026amp; Fristrup, K. Changes in night sky brightness after a countywide LED retrofit. \u003cem\u003eJ. Environ. Manage.\u003c/em\u003e \u003cb\u003e292\u003c/b\u003e, 112776 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, X. et al. Anisotropic characteristic of artificial light at night\u0026ndash;Systematic investigation with VIIRS DNB multi-temporal observations. \u003cem\u003eRemote Sens. Environ.\u003c/em\u003e \u003cb\u003e233\u003c/b\u003e, 111357 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHansen, A. G., Beauchamp, D. A. \u0026amp; Baldwin, C. M. Environmental constraints on piscivory: insights from linking ultrasonic telemetry to a visual foraging model for cutthroat trout. \u003cem\u003eTrans. Am. Fish. Soc.\u003c/em\u003e \u003cb\u003e142\u003c/b\u003e (1), 300\u0026ndash;316 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTabor, R. A., Warner, E. J., Fresh, K. L., Footen, B. A. \u0026amp; Chan, J. R. Ontogenetic diet shifts of prickly sculpin in the Lake Washington basin, Washington. \u003cem\u003eTrans. Am. Fish. Soc.\u003c/em\u003e \u003cb\u003e136\u003c/b\u003e (6), 1801\u0026ndash;1813 (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTabor, R. A. et al. Smallmouth bass and largemouth bass predation on juvenile Chinook salmon and other salmonids in the Lake Washington basin. \u003cem\u003eNorth Am. J. Fish. Manag.\u003c/em\u003e \u003cb\u003e27\u003c/b\u003e (4), 1174\u0026ndash;1188 (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarentine, J. C. et al. Recovering the city street lighting fraction from skyglow measurements in a large-scale municipal dimming experiment. \u003cem\u003eJ. Quant. Spectrosc. Radiative Transf.\u003c/em\u003e \u003cb\u003e253\u003c/b\u003e, 107120 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTummers, B. \u0026amp; DataThiefIII (2006). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://datathief.org/\u003c/span\u003e\u003cspan address=\"https://datathief.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDoxaran, D. et al. Optical characterisation of suspended particles in the Mackenzie River plume (Canadian Arctic Ocean) and implications for ocean colour remote sensing. \u003cem\u003eBiogeosciences\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e (8), 3213\u0026ndash;3229 (2012).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-6406457/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6406457/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eArtificial light at night (ALAN) poses a threat to ecosystems globally. It includes both direct and indirect light, or skyglow, which occurs when ALAN scatters in the atmosphere, extending beyond its source. We analyzed ALAN trends in Lake Washington, WA, from 2014 to 2023 using Visible Infrared Imaging Radiometer Suite (VIIRS) nighttime light measurements, evaluated the relationship between in-situ and satellite measurements, and assessed juvenile salmon predation risk due to ambient light. We observed significant increases in ALAN in all open water regions, while nearshore regions primarily showed no or negative trends, revealing the significant role of skyglow on open water light levels. Although spatial correspondence between the satellite and in-situ measurements was observed, VIIRS did not capture the changes in yellow and blue light. Juvenile salmon at the shallow southern open water site experience light levels 28 times brighter than fish at the northern site. In the Ship Canal, a narrow corridor for outmigrating salmon, light levels are 65 times brighter than at the southern site. These contrasts in light exposure highlight the impact of local conditions on predation risk by visual foraging predators and emphasize the need for effective mitigation efforts targeting both nearshore and distant light sources.\u003c/p\u003e","manuscriptTitle":"Shedding Light on the Role of Artificial Light at Night in Lake Washington, WA, USA","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-23 15:02:38","doi":"10.21203/rs.3.rs-6406457/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"a68b3966-4269-4a77-aa82-8c2840fcacd6","owner":[],"postedDate":"May 23rd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":47197363,"name":"Biological sciences/Ecology"},{"id":47197364,"name":"Earth and environmental sciences/Ecology"},{"id":47197365,"name":"Earth and environmental sciences/Environmental sciences"},{"id":47197366,"name":"Earth and environmental sciences/Limnology"}],"tags":[],"updatedAt":"2025-09-11T17:38:29+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-23 15:02:38","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6406457","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6406457","identity":"rs-6406457","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}