(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .



Application of watershed-scale habitat modeling and decision-support tools for reservoir reoperations in coastal northern California [1]

['Doug Chalmers', 'Us Centre', 'Stockholm Environment Institute', 'Davis', 'California', 'United States Of America', 'Marisa Escobar', 'Laura G. Forni', 'Jason Nishijima', 'Watersheds Stewardship']

Date: 2023-07

Water managers must often balance the needs of both aquatic habitat and human water supply. However, they frequently only have the tools to manage water delivery alone. Existing modeling tools for habitat have gaps in providing detailed biological estimates at a watershed scale and in simulating water supply operations and habitat suitability simultaneously. A new modeling platform and calculation framework, Aquatic Habitat Assessment, was applied in a case study to quantify habitat suitability and fish passage at a watershed scale for local species of Chinook salmon (Oncorhynchus tshawytscha) and steelhead trout (Oncorhynchus mykiss). Aquatic Habitat Assessment was coupled with a suite of tools, including HEC-RAS used for hydraulics, WEAP for water allocations, and Tableau for visualization. The tools ensemble was used to simulate the operations of a water utility system near San Francisco Bay in California to evaluate the effects of reservoir reoperations on both human water supply and aquatic habitat. The suite of tools was successful in bringing a range of conflicting parties to coalesce around a common solution for reservoir operations. Two sets of alternative reservoir operation schemes were developed, which largely involve higher winter and lower summer releases, aligning more closely with natural Mediterranean patterns and functional flows connected to the biological needs of local aquatic species. Results quantify tradeoffs across reservoir operation schemes, with modeled fish passage habitat suitability increased under the reservoir reoperations, but human water supply delivery decreased. The modeled spawning, incubation, and rearing habitat suitability increased during the winter release period under reoperations, but decreased during the summer release period.

Competing interests: I have read the journal’s policy and the authors of this manuscript have the following competing interests: Two authors from the funding organization were among several collaborators who established the study design. The authors from the funding organization additionally validated the data analysis and provided editing of the manuscript.

Funding: Study funded by Agreement A3939Xa awarded to ME from the Santa Clara Valley Water District: https://www.valleywater.org/ . Two authors from the funding organization were among several collaborators who established the study design. The authors from the funding organization additionally validated the data analysis and provided editing of the manuscript. The funder had no role in the decision to publish.

Copyright: © 2023 Chalmers et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Central California Coast steelhead trout were listed as threatened in 1997 under the federal Endangered Species Act, and as high concern in the 2017 State of Salmonids report [ 33 ]. The current population is believed to be less than 10% of historical numbers, with an estimate from the National Marine Fisheries Service of 14,100 annual spawners remaining in 2006 [ 33 ]. Central California Coast steelhead trout appear naturally in the Three Creeks. Urbanization and dam construction have reduced natural habitat in Coyote Creek by 49%, Guadalupe River by 21%, and Stevens Creek by 54% [ 33 ].

As of 2023, Central Valley fall-run Chinook are a federal species of concern, a state species of special concern, and listed as high concern in the 2017 State of Salmonids report [ 33 ]. Fall-run Chinook were historically the most abundant anadromous fish run in California [ 34 ], with estimates of one million annual spawners in the Central Valley [ 35 ]. However, by the early 1900s, human activity during the gold rush era had reduced the population to approximately 10% of historical abundance [ 35 ]. Since the twentieth century, dams have blocked approximately 60% of fall-run Chinook spawning habitat [ 36 ], resulting in further decline. Being a prominent species within state hatcheries, hatchery fall-run Chinook have been observed establishing spawning runs in streams not previously occupied by natural populations [ 34 ]. Starting in the mid-1980s, Central Valley fall-run Chinook of hatchery origin established runs of 100–200 spawners in the Guadalupe and Coyote Creek watersheds of the Three Creeks system [ 34 ]. Archeological evidence of native Chinook salmon has been attributed to several San Francisco Bay tributaries, but no conclusive evidence of native Chinook salmon runs in the Three Creek systems has been found [ 37 ].

The habitat assessment was completed for two local fish species, Central Valley fall-run Chinook salmon (Oncorhynchus tshawytscha, often simplified to “Chinook” in this paper) and Central California Coast steelhead trout (Oncorhynchus mykiss) populations. Chinook salmon and steelhead trout are anadromous species, which migrate from the ocean up freshwater streams to spawn.

The current broad objectives of Valley Water include delivering reliable water supply to its constituents, providing flood protection, protecting aquatic habitat in the local watershed, and enhancing groundwater recharge to local aquifers. There are municipal & industrial and domestic demands throughout the watershed plus agricultural demands surrounding Coyote Creek and watersheds to the south. The groundwater aquifers provide supply for the municipal & industrial, domestic, and agricultural uses, as well as support groundwater dependent ecosystems. The aquifers are replenished by natural recharge largely from precipitation, and Valley Water’s managed aquifer recharge operations that use surface water from the system via percolation ponds and seepage through the stream channels. To protect fish passage, habitat, and groundwater recharge, there are several existing instream flow requirements.

The case study described in this paper was performed within the managed streams of Valley Water’s jurisdiction. Valley Water’s jurisdiction encompasses the entirety of Santa Clara County– 3,400 km 2 –serving 15 cities and 2 million residents. The scope of the habitat assessment applies to the managed portion downstream of local reservoirs of the Three Creeks watersheds ( Fig 2 ), consisting of Coyote Creek, Stevens Creek, and Guadalupe River along with its tributaries: Alamitos, Calero, Guadalupe, and Los Gatos Creeks. The Three Creeks run from south to north, originating from the undeveloped hillsides south of San Jose and draining to the San Francisco Bay. Typical of urban areas in coastal California, Valley Water relies on a combination of local and imported water supplies. Santa Clara County receives an average of 18.4 inches of annual precipitation, with significant variability and susceptibility to drought. As a result, local supplies are supplemented with imported water from a portfolio including the Central Valley Project, California State Water Project, Hetch Hetchy Project, and Semitropic Water Bank.

In the FAHCE study, AHA was applied to help design and assess the reservoir reoperation plan alternatives. AHA was run as a plugin within SEI’s Water Evaluation And Planning (WEAP) [ 27 ] water allocation model of the water utility system. AHA produced a direct quantitative comparison of water supply and ecological results at the watershed scale across the water management scenarios. These results were connected to a Decision Support Visualization (DSV) tool using the Tableau software platform to provide the detailed representation across spatiotemporal scales needed to evaluate potential tradeoffs [ 28 – 32 ] between meeting water supply and ecological objectives. The direct numerical comparison of management scenarios was useful to resolve the cognitive conflicts–technical disagreements of how to interpret results [ 1 ]–across the parties, and ultimately allowed for the development of the alternative reservoir reoperation plans to best balance water supply and ecological objectives.

This paper describes how AHA was applied in a case study with the Santa Clara Valley Water District (also known simply as Valley Water) to compare the effects of various reservoir reoperation scenarios on habitat suitability. Valley Water is a local government agency providing water resources management, flood protection, and stream stewardship in Santa Clara County, California within the Silicon Valley region surrounding the city of San Jose. The FAHCE (Fish and Aquatic Habitat Collaborative Effort) Settlement Agreement detailed in this study was initiated to provide a path to resolve an administrative water rights complaint filed against Valley Water at the California State Water Resources Control Board. The FAHCE Settlement Agreement required reoperations of Valley Water’s reservoirs to better protect downstream habitat and certain non-flow measures to improve fish habitat and migration. These flow and non-flow measures are required to be implemented in an adaptive manner through the FAHCE Adaptive Management Program to effectively mitigate adverse impacts on fisheries and habitat for two local species (detailed in section 1.6), resulting from Valley Water’s water supply facilities and operations. The FAHCE case study detailed in this paper consisted of the reservoir reoperation design, completed through a collaborative process with Valley Water, the Stockholm Environment Institute (SEI) and other consultants, as well as the TWG (Technical Working Group), a body comprised of experts from each party in that initialed the FAHCE Settlement Agreement, responsible for overseeing and providing technical input on the implementation of the flow and non-flow measures in the Agreement. Valley Water’s objectives for reoperations were to maximize winter habitat, as was desired by the TWG, while minimizing any adverse impacts to summer rearing habitat, water supply, or groundwater recharge.

This paper presents the use of a new platform, Aquatic Habitat Assessment (AHA), which enables water managers to connect water operations to both supply and habitat objectives simultaneously at the watershed scale. Pictured in Fig 1 below, AHA encompasses a system of methods to create a simulation modeling platform which integrates hydro-ecological processes within a water operations model. The case study used a particular set of modeling tools for each portion of AHA, but the general methodology may be applied using other models as well. AHA relates physical stream characteristics to fish species biological suitability curves during each life stage. The tool can explore questions like: How many acres of suitable habitat are available for spawning salmonids at any given moment? How frequently can juveniles successfully emigrate? What is the most upstream passage extent of migrating adult salmonids?

An additional research gap exists in current approaches for integrating both human and ecological inputs and outputs together. Simulating water supply operations and aquatic habitat suitability integrated together allows for habitat management goals to influence upstream controlled releases dynamically in the model [ 25 ] and also can produce quantified performance results to study the tradeoffs between water supply and habitat. The AQUATOOL platform, a decision-support system which integrates water operations simulation together with water quality and habitat evaluation [ 26 ] has been applied by Pardo-Loaiza et al., 2022 and Paredes-Arquiola et al., 2014 to complete habitat evaluations at the basin scale. However, these studies relied on flow-based evaluation methods and only a small number of river reaches, lacking a detailed, comprehensive assessment across the entire watershed.

Habitat suitability methods perform a more detailed assessment of aquatic habitat conditions, connecting depth, velocity, and other physical stream parameters to the biological habitat preferences of aquatic species during particular life stages. Popular habitat suitability modeling platforms include MesoHABSIM [ 17 , 18 ], the mesohabitat evaluation model [ 19 ], and CASiMiR [ 20 ]. These more detailed characterizations require significant input data at small spatial scales, such as the geomorphic unit or mesoscale, defined as 1 to 100 m [ 16 , 17 ]. The extensive data requirements makes application of these methods across larger scales, such as the reach scale (100 m to 10 km) [ 21 ], often impractical due to time and cost limitations [ 14 , 22 , 23 ]. As a result, most studies are limited in scope to a small number of river reaches rather than a watershed-scale. Two recent studies by Spurgeon et al., 2019 [ 24 ] and Parasiewicz et al., 2018 [ 23 ] have been observed to apply MESOHABSIM at a more comprehensive watershed scale, but the studies did not consider water management alternatives or performance- the second research gap discussed below.

Generally, riverine aquatic habitat may be evaluated using either flow-based metrics, or habitat suitability metrics related to hydraulic stream parameters including depth and velocity. Popular flow-based aquatic habitat evaluation methods include Indicators of Hydrologic Alteration [ 8 ], the functional flows approach [ 9 ], and Hydroecological Integrity Assessment Process [ 10 ]. These methods evaluate aquatic habitat by comparing the streamflow regime between altered and natural conditions, and have been used to conduct habitat evaluations at a watershed scale [ 7 , 11 – 13 ]. However, such approaches lack the capacity to fully represent the hydraulic, geomorphic, and ecologic interactions of physical habitat [ 13 – 16 ].

Existing approaches to integrate ecosystem considerations in water planning face two general limitations in their application for water utilities: they lack the ability to provide sufficiently detailed biological results while still being able to analyze at a watershed scale, and the ability to integrate both human and ecological inputs and outputs together.

In order to incorporate ecosystem needs into water utility planning, water managers need access to new tools which can integrate both human and habitat demands alongside each other, as well as resolve allocation conflicts [ 7 ] inherent in water-scarce regions. This work presents a case study of applying such a new tool.

Studies show that ecological services–benefits arising from the ecological functions of healthy ecosystems–are central for sustainable water resource protection [ 4 ]. When ecosystem health is not considered in water supply planning, human diversions can decimate aquatic habitat by draining wetlands, hindering fish migration, and increasing saltwater intrusion. Climate change intensifies the impacts with warming waters, ocean acidification, and extreme weather. Addressing these effects–and choosing sustainable policy options–means integrating ecosystem needs into water management. However, water managers often only have the tools to consider human uses of the water system [ 5 – 7 ].

As global population grows and precipitation patterns change, water suppliers increasingly must stretch their limited resources to meet various needs. Water suppliers face an interest conflict [ 1 ] between human water demands for urban, agricultural, industrial, and energy production versus ecological demands for maintaining healthy aquatic habitat. However, in urban areas, current water supply operations often focus exclusively on human needs [ 2 , 3 ]. Ecosystem health is a crucial missing element.

2. Methods

2.1. Management scenarios The study compares the modeling results across six management scenarios. In each scenario, one of three reservoir operation schemes (Base Case, FAHCE, and FAHCE Plus) was run under one of two temporal conditions (2015 and 2035). Base Case represents the then-current reservoir operations at the onset of the project. FAHCE represents the alternative reservoir operations originally specified under the FAHCE Settlement Agreement. The FAHCE operation rules were designed with objectives to increase winter habitat for spawning, incubation, and passage while minimizing any detriment to rearing and avoiding any unreasonable impacts to water supply and groundwater recharge. While the operation rules of FAHCE were already specified in the FAHCE Settlement Agreement, its performance related to the objectives were assessed using AHA. The FAHCE operations implement the following for the Three Creeks reservoirs: Winter base flow releases to improve winter and spring habitat (November to April in Coyote and Guadalupe watersheds, January to April in Stevens Creek).

Spring pulse flow releases from February to April of 1.4 CMS (cubic meters per second) for 5 consecutive days up to twice per year to improve fish passage.

Summer base flow releases from May to October to provide rearing habitat for steelhead. The summer base flows are made as part of the proposed cold-water management program, which controls the volume and temperature of releases to maintain maximum daily average temperatures of 18–19°C downstream of Anderson, Guadalupe, and Stevens Creek reservoirs from May to October for summer rearing habitat. The FAHCE Plus operation rules were designed under an iterative, collaborative process between SEI, Valley Water, consultants, and the TWG to refine the reservoir operations rule curves specified in the FAHCE Settlement Agreement. The FAHCE Plus operations attempt to further maximize passage reliability, while minimizing any detriment to the gains made in spawning and incubation under the FAHCE operation rules and avoiding any unreasonable impacts to water supply or groundwater recharge. The FAHCE Plus operations provide the following additional refinements to the FAHCE operations: Lower releases during some portions of winter base periods to conserve water for pulse flow opportunities and make summer rearing flows more reliable.

The cold-water management program raised the maximum temperature range of the reservoir storage which could be considered for cold-water releases to provide greater summer flows.

Additional flexibility in pulse flow timing and magnitude implemented to provide more frequent pulses and provide pulses in a greater frequency of years. The date range of pulse flows was modified from February to April to be December to April. When sufficient volume for summer releases would remain available, safeguard pulses are provided each March if other migration pulses have not yet been provided that year to provide a greater frequency of years with steelhead migration and spawning opportunities. Lastly, an emigration-specific pulse was added in mid-April to provide more frequent emigration opportunities that better align with expected timing of the life stage of the species. The 2015 conditions reflect the then-current set of demands and infrastructure. The 2035 conditions reflect future projected demands and imports, as well as infrastructure changes, most notably including structural improvements to the reservoirs against seismic danger to allow for greater storage. Because of the greater management flexibility and closer reflection of potential impacts of future operations associated with the increased reservoir storage, this paper focuses on the results of the 2035 scenarios. A representative sample of the 2035 Base Case, FAHCE, and FAHCE Plus seasonal reservoir operations is shown in Fig 3. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 3. Representative seasonal reservoir operations for Base Case, FAHCE, and FAHCE plus. https://doi.org/10.1371/journal.pwat.0000069.g003

2.2. Water allocation model The operations of Valley Water’s water utility system in the Three Creeks watershed under the water management scenarios described above were simulated in a daily water allocation model using the SEI’s WEAP [27] platform as part of the overall AHA modeling toolkit shown in Fig 1. Several model platforms, including AQUATOOL, MODSIM, RIBASIM, WARGI-SIM, are common in water resources planning [12] and could potentially be applied in place of WEAP. In this case, WEAP was chosen based on its ability to simulate water allocations during each daily time step based on user-defined priority [27] rather than using pre-defined allocation of other models such as WARGI-SIM or RIBASIM [12] as well as the ability to build off the utility’s existing monthly WEAP model. The hydrologic portion of the model was populated from 1990–2010 using historical observed inflows above the reservoirs, local climate, and modeled rainfall-runoff inflows beneath the reservoirs. Estimates of demands and calibrated algorithms of operations were integrated together with the hydrologic components to simulate daily streamflow, reservoir storage, and water allocations across the water utility system. The modeled streamflow was validated against observed streamflow gauge data in the Three Creeks, as shown in S1 Table. The streamflow performance criteria were based on daily watershed-scale streamflow model criteria from Moriasi et al., 2015 [38] shown in Table 1. The calibration model runs entered observed reservoir releases as direct inputs to isolate the performance of the downstream operations and runoff hydrology independent of reservoir operation modeling performance. In the calibration, all 8 point of interest (POI) locations (shown in Fig 2) showed satisfactory NSE values, 4 of 8 showed satisfactory PBIAS values, with another 4 of 8 showing PBIAS values still within the threshold of acceptable performance. The validation model runs included the full set of modeled water utility operations. For the 16 validation POI locations, 10 showed satisfactory NSE and PBIAS, 5 showed NSE and 4 showed PBAS within the acceptable threshold, and 1 showed NSE and 2 showed PBIAS with unacceptable performance. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Table 1. Daily watershed model performance criteria for streamflow. https://doi.org/10.1371/journal.pwat.0000069.t001 Although the model shows locations with not satisfactory and unacceptable validation streamflow performance, Moriasi et al., 2007 describes how model performance criteria may be relaxed based on the nature of the model [39]. Given the difficulty in modeling water operations decisions compared to a purely physical hydrology model as supported by Brown et al., 2015 [40], lower performance thresholds can be expected. In accordance with Legates and McCabe, 1999 and Moriasi et al., 2007 which support the importance of graphical techniques in evaluating hydrologic models [39, 41], seasonal box plots and exceedance curves of observed versus modeled flows were also used to validate model performance. Those locations where the historical flow validation results are not acceptable show improved performance under the hydrology calibration. This attributes the source of error in modeled daily streamflow to the reservoir releases. It was difficult to precisely model the daily timing of transfers due to the variability and uncertainty in the decisions taken by reservoir operators around the frequency and volume of these transfers. However, this source of error may not be relevant in the Base and FAHCE model scenarios, which represent present and future, rather than historical reservoir operations. The present and future reservoir operations can be modeled more accurately with better and more current information. The hydrology calibration results reveal that there are still differences in the modeled and observed results unrelated to reservoir operations. These differences are likely attributed to inaccuracies in the historical amount and timing of imported water entering the system. In addition to the flow, the model also simulated average daily stream temperature, a critical parameter in the habitat assessment. A statistical regression approach, shown in the below equations where a, b, c, d, and e are calibrated constants was used to calculate daily average water temperature at 31 of the 39 POI and river reach locations (described further in section 2.4) where observed temperature calibration data was available. The regression approach was selected because of the incomplete calibration data and discrete nature of the model with a daily time step and temperature calculated at the POI locations. Eq 1 below shows how reservoir storage was used as a proxy for reservoir release temperature. (1) (2) (3) The regression coefficients at each POI were calibrated with unique daily values using a least-squares regression to minimize the sum of squared errors between the predicted water temperature and the historical observed water temperature from stream logger data during the 2000–2014 calibration period. The validation performance between daily average observed and modeled stream temperatures shown in S2 Table shows an average root mean square error (RMSE) of 2.2°C and mean absolute error (MAE) of 1.7°C across all locations. While stream temperature does not possess strict performance criteria thresholds [42] these values fall within a range of performance values for other published daily water temperature streamflow models. Published MAE values from Kamarianakis et al., 2016 [43] range from 0.4–1.7°C, with a value of 0.7°C from Ficklin et al., 2012 [42] falling within this range. Published RMSE values for daily stream temperature models range from 0.6°C found in Kamarianakis et al., 2016 [43] to 2.5°C found in Barnhart et al., 2014 [44], with additional values from Ficklin et al., 2012, Khorsandi et al., 2022, and Piccolroaz et al., 2016 [42, 45, 46] falling within this range. Likely sources of error include discrepancies in reservoir release and streamflow volumes, limited temperature calibration data for some POIs, and simplification of the temperature into a one-dimensional model.

2.4. Habitat assessment To connect the water utility operations to the effects on aquatic habitat, depth, velocity, and temperature results generated from the coupled water allocation and hydraulic models were used to quantify daily habitat suitability and fish passage using the novel AHA framework shown in Fig 1. In doing so, AHA can quantify the effect of system operations on the biological viability of aquatic species during each of the species’ life stages. While the habitat assessment results using the AHA framework may be calculated using different tools, AHA was run as a plugin within the WEAP water allocation model to calculate most of the habitat results directly in WEAP. To access the WEAP plugin for AHA, WEAP may be downloaded and ran for free (https://www.weap21.org/Download/), then AHA is available as a plugin download. Access to AHA is maintained as part of the overall software maintenance of WEAP, and AHA may be updated in future WEAP versions. While this section describes how AHA was applied in the case study, the AHA framework is highly customizable and can be modified to conduct habitat assessment for other species, locations, or parameters. In the case study, AHA was applied to estimate habitat suitability for Chinook salmon and steelhead trout for six life stages: immigration, spawning, incubation, fry rearing, juvenile rearing, and emigration. In the life cycles of these anadromous species, adults immigrate from the ocean to cold headwater streams to spawn. The eggs incubate, then the hatched fry rear in the headwater areas, growing into juveniles. Once large enough, the juveniles emigrate out to the ocean. The habitat suitability curves and thresholds used to quantify the habitat assessment results using AHA are summarized in Table 3 below, with more detailed suitability curves provided in S3 Table. The criteria, documented in detail within the Appendix N–Fisheries Habitat Availability Estimation Methodology of the publicly available FAHCE Draft Program Environmental Impact Report (DEIR), accessed through https://www.valleywater.org/, were provided by Valley Water based on its local aquatic stream population surveys and an extensive literature review on Chinook salmon and steelhead trout habitat suitability in California. The literature review, detailed within the FAHCE DEIR, incorporated studies from Williams 2006, Merz et al., 2016, Moyle 2002, Satterthwaite et al., 2010, Sogard et al., 2012, Beakes et al., 2010, Merz et al., 2013, Zeug et al., 2014, Satterthwaite et al., 2009 [34, 67–74], California Department of Fish & Game, California Department of Fish & Wildlife, the United States Environmental Protection Agency, United States Bureau of Reclamation, and the San Francisco Public Utilities Commission. The criteria are based on fish population survey data, with optimal suitability values representing stream conditions where the maximum number of fish were observed, corresponding to habitats with ideal, healthy conditions. Uninhabitable values represent stream conditions where zero fish were observed, corresponding to habitat conditions that the species would avoid traveling to or staying in, and where they would otherwise suffer from reduced fitness over time from staying in. The habitat results for each life stage are only quantified during the corresponding active life stage timing period in Table 3. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Table 3. Aquatic Habitat Assessment habitat suitability curves and thresholds used. https://doi.org/10.1371/journal.pwat.0000069.t003 Habitat suitability results in AHA for the spawning, embryo incubation, and rearing life stages when the fish species are reproducing and living in the freshwater streams are quantified at each river reach, broken out by the 39 POIs in the Three Creeks, using three types of indices: habitat suitability index (HSI), composite habitat suitability index (CHSI), and habitat availability index (HAI). The three indices are differentiated according to their level of aggregation. The HSI is the least-aggregated result value, which is calculated daily for each combination of river reach, stream parameter (depth, velocity, temperature, substrate embeddedness, suitable substrate), habitat type (riffle, run, and pool), life stage and species (steelhead and Chinook). Each HSI relates a single stream parameter to a biological suitability value ranging from 0 (not habited) to 1 (optimal) using the habitat suitability curves (Table 3, S3 Table). The depth and velocity inputs were calculated by relating the daily streamflow of the water allocation model to the flow-depth-velocity relationships of the given representative cross-section from the hydraulic model (section 2.3), corresponding to the average depth and velocity across the cross-section. The temperature inputs were calculated from the water allocation model streamflow results coupled with the temperature regression equations (section 2.2). The cover, substrate embeddedness, and suitable substrate inputs are static values for each habitat type of each river reach, representing the structural habitat component. Cover, suitable substrate, and substrate embeddedness input values were based off field survey data gathered by Valley Water. Detailed survey techniques and complete documentation of source information for the structural habitat components is found in the FDEIR Appendix N and largely adapted from the 2010 California Salmonid Stream Habitat Restoration Manual of the California Department of Fish & Game by Flosi et al. Cover was defined as the areal percent of a habitat unit which contained the categories of cover defined in Flosi et al., 2010 the: large cobble, medium cobble, boulder, root mass, large woody debris, bedrock ledges, and undercut bank. Because of the differing habitat preferences of salmonids during the winter versus summer months as noted in the FDEIR Appendix N, the summer cover was defined as the areal proportion of small cobble. Suitable substrate, representing those surfaces where fish are most likely to spawn, was defined as the aerial proportion of a habitat unit containing suitably-sized substrate (12–102 mm for steelhead, 12–152 mm for Chinook) based on report data from several government agencies documented in FDEIR Appendix N. Substrate embeddedness, representing the percent of substrate buried in fine sediment which would reduce their utility for spawning, was estimated by visual techniques adapted from Flosi et al., 2010 documented in the FDEIR Appendix N. Using the example habitat suitability curve shown in Fig 6, a modeled velocity of 0.40 m/s at the given reach and habitat type would correspond to an HSI value of 0.40 for steelhead spawning, representing habitable, but sub-optimal habitat area. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 6. Example habitat suitability curve for steelhead spawning velocity. https://doi.org/10.1371/journal.pwat.0000069.g006 The HSI values are aggregated across all stream parameters to calculate a daily CHSI for the given life stage, species, river reach, and habitat type. Eqs 8 and 9 show CHSI calculations for spawning and rearing respectively. CHSI values are then aggregated across all habitat types to calculate the daily HAI for each life stage, species, and river reach as shown in Eq 10. Although calculated in units of area, the HAI represents the general capacity of the system to provide habitat. (8) (9) (10) The embryo incubation results in AHA are also expressed using HAI area, but are calculated differently using accumulated temperature units (ATU). For each day in the active life stage period for incubation, a new daily cohort of embryos undergo an ATU evaluation at each river reach and habitat type for each species. For each daily embryo cohort, an ATU sum accumulates, adding on the average daily water temperature for each day where the incubation conditions for depth and temperature are satisfied. If the ATU units reach 582°C for steelhead or 898°C (FDEIR Appendix N) for Chinook before the depth or temperature criteria are failed, then the incubation CHSI is assigned a value of 1 for the day when incubation began, representing likely successful embryo incubation through hatching. Otherwise, the CHSI for that day is assigned a value of 0, representing embryos which likely failed to survive through hatching. The CHSI results are then aggregated into HAI units of area as shown in Eq 10. Incubation results were calculated using two sets of temperature thresholds, named upper habitable and upper optimal in Table 3. The incubation results in this paper use the upper habitable results, developed based on feedback from the TWG that exceeding the upper habitable threshold for short periods of time may not always result in embryo mortality. The habitat results for fish passage for the life stages of emigration and immigration in AHA rely on binary evaluations. The depth evaluation was performed using the thalweg (rather than average) depth at the critical passage-limiting riffle cross-sections for each POI, distinct from the representative riffle, run, and pool cross-sections used for the other life stages. As noted in Table 3 above, the threshold values of depth and temperature were evaluated to indicate a binary result of either 0 (no passage) or 1 (passage) at each POI. Upstream adult immigration is evaluated at each POI, with passage at upstream POIs only possible if all downstream POIs also have successful passage. Juvenile emigration is evaluated only at the most upstream POI of each creek, with successful emigration only possible if downstream passage is possible for all downstream POIs connecting to the Bay.

[END]
---
[1] Url: https://journals.plos.org/water/article?id=10.1371/journal.pwat.0000069

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.