Peterson, S.H. Ackerman, J.T. 20260227 spreadsheet (CSV) n/a U.S. Geological Survey data release https://doi.org/10.5066/P13NYGAI Sarah H. Peterson Joshua T. Ackerman Brady L. Fettig Andrew C. Greenawalt C. Alex Hartman Meghan P. Keating Mark P. Herzog Unknown publication Ornithological Applications Unknown We conducted a multi-year study (2016-2019) to identify elements of wetland habitat management that could potentially benefit breeding ducks. Specifically, we examined the influence of habitat variables in explaining habitat selection of managed wetlands near upland nesting sites by radio-tagged Mallard (Anas platyrhynchos) and Gadwall (Mareca strepera) ducklings in a brackish marsh (Suisun Marsh, California, USA). Each duckling location was associated with a unique wetland identification (after moving from their nest in upland habitats to a nearby wetland) and the corresponding wetland attributes for that year and the timing within the year (wetland type, dominant wetland vegetation, the percent of the wetland that contained vegetation cover, surface water salinity concentration, and flooded wetland area). For the habitat selection analysis, we also generated available locations in flooded wetlands on the landscape for each radio-tagged brood and associated the corresponding wetland attributes to those available locations. Additionally, we used daily locations of radio-tagged ducklings to quantify the number of wetlands used by ducklings, the frequency of inter-wetland movements, and duckling dispersal from the nest relative to species, age, and year. Managed wetlands that are flooded during the spring and summer provide critical habitat for breeding waterfowl populations and wetlands in close proximity to upland nesting habitat can be especially important for duckling broods. Broods may prefer flooded wetlands with certain habitat characteristics; consequently, habitat management actions that increase the availability of preferred habitats near to upland nesting areas could potentially increase duckling survival rates and improve population recruitment. Habitat selection analyses were conducted separately by species. 2016 2019 ground condition Complete None planned Suisun Marsh, California -122.0993 -121.8480 38.2630 38.0448 ISO 19115 Topic Category biota USGS Thesaurus birds ornithology ecology wetland ecosystems habitats Common Geographic Areas California USGS Metadata Identifier USGS:68631b39d4be025653d31f22 Integrated Taxonomic Information System (ITIS) Anas platyrhynchos Mareca strepera Salicornia Phragmites australis Schoenoplectus Bolboschoenus maritimus Typha Distichlis spicata Frankenia salina Mallard Gadwall pickleweed saltgrass bulrush cattail U.S. Geological Survey 2013 Online Database https://doi.org/10.5066/F7KH0KBK www.itis.gov expert identifier Domain Eukaryota Kingdom Animalia Subkingdom Bilateria Infrakingdom Deuterostomia Phylum Chordata Subphylum Vertebrata Infraphylum Gnathostomata Superclass Tetrapoda Class Aves Order Anseriformes Family Anatidae Genus Anas Species Anas platyrhynchos TSN: 175063 Genus Mareca Species Mareca strepera TSN: 1192581 Kingdom Plantae Subkingdom Viridiplantae Infrakingdom Streptophyta Superdivision Embryophyta Division Tracheophyta Subdivision Spermatophytina Class Magnoliopsida Superorder Caryophyllanae Order Caryophyllales Family Amaranthaceae Genus Salicornia TSN: 20646 Family Frankeniaceae Genus Frankenia Species Frankenia salina TSN: 502648 Superorder Lilianae Order Poales Family Poaceae Genus Phragmites Species Phragmites australis TSN: 41072 Genus Distichlis Species Distichlis spicata TSN: 40662 Saltgrass Family Cyperaceae Genus Schoenoplectus TSN: 500920 Genus Bolboschoenus Species Bolboschoenus maritimus TSN: 508145 Family Typhaceae Genus Typha TSN: 42324 No access constraints. Please see 'Distribution Info' for details. No use constraints. Questions pertaining to appropriate use or assistance with understanding limitations or interpretation of the data are to be directed to the individuals/organization listed in the Point of Contact section. U.S. Geological Survey, Western Ecological Research Center Data Manager mailing and physical

3020 State University Drive, Modoc Hall, Suite 4004

Sacramento CA 95819 USA 279-782-0904 gs-b-werc_data_management@usgs.gov This research was supported by the California Department of Water Resources and the U.S. Geological Survey Ecosystems Mission Area. R is a free software environment for statistical computing and graphics. It compiles and runs on a wide variety of UNIX platforms, Windows and MacOS. https://cran.r-project.org/mirrors.html You can choose your preferred CRAN mirror to visit a site to download R. To download previous versions of R, visit https://cran.r-project.org/bin/windows/base/old/ For analyses, we only included live duckling locations (excluding locations from the telemetry flights) and we only included one location per brood per day for each unique wetland. In the few cases (n = 7 broods) where the two radio-tagged brood mates were tracked to different wetlands, the locations for the duckling that survived longer were retained and the locations of the other duckling were censored (n = 2 broods). If radio-tagged brood mates were located in different wetlands and then neither duckling was located again, we retained the location of first duckling that was located for that day and censored the location of the second duckling (n = 5 broods). Data were visually inspected to check for outliers. Data were checked for ranges and duplication. Not all the locations used by ducklings are in both the Duckling_Movement_Dataset.csv and the Duckling_Habitat_Selection_Dataset.csv. The Duckling_Movement_Dataset.csv includes some broods (2018_11AS-1_1, 2018_13L_2, 2016_10NW_2) that were excluded from the Duckling_Habitat_Selection_Dataset.csv. Two broods that left Grizzly Island were excluded from the habitat analysis dataset because they were resighted at a lower frequency than other broods and were outside of the area where landscape-level habitat data were quantified. Winter-flooded seasonal wetlands were excluded from analyses for Gadwall ducklings because they were not available for Gadwall ducklings (Gadwall ducks nest later in the year than Mallard ducks). The one Gadwall brood that had a single location in a rapidly drying out winter-flooded seasonal wetland was excluded. We deployed very high frequency (VHF) radio transmitters that contained thermistors (model BD-2T, Holohil Systems, Ontario, Canada) on two randomly chosen ducklings from each brood within hours of hatching. See the larger work for more information on duckling selection. We did not include the aircraft-derived duckling locations in our analyses due to the increased uncertainty in correctly assigning the location to a wetland unit. The aircraft was able to search methodically over a wide area and locate transmitters that had gone missing, enabling the ground crew to either continue tracking those ducklings or recover the transmitters. Radio transmitters were recovered when the duckling died or the transmitter fell off. The Duckling_Movement_Dataset.csv only includes movements from up to 40 days after hatch. Locations from ducklings older than 40 days were included in the habitat analysis dataset. Users are advised to read the rest of the metadata record carefully for additional details. We tracked radio-tagged Mallard and Gadwall ducklings approximately daily to obtain locations. Duckling locations were obtained using a truck equipped with dual 4-element Yagi antennas (Advanced Telemetry Systems, Inc, Isanti, Minnesota, USA) and a null-peak system (AVM Instrument Company, Auburn, California, USA). When a duckling was detected, a truck-mounted electronic compass (Revolution, True North Technologies, Hudson, Massachusetts, USA) and a hand-held GPS unit (Garmin Ltd., Olathe, Kansas) were used to obtain 3 bearings and GPS locations that were input into triangulation software (Location of a Signal, 4.0.3.8, Ecological Software Solutions, Urnäsch, Switzerland) to calculate Universal Transverse Mercator (UTM) coordinates for each duckling. Carley R. Schacter Sarah H. Peterson Mark P. Herzog C. Alex Hartman Michael L. Casazza Joshua T. Ackerman 2021 publication San Francisco Estuary and Watershed Science volume 19, issue 3 Davis, California eScholarship https://doi.org/10.15447/sfews.2021v19iss3art5 Digital and/or Hardcopy 2016 2019 ground condition Schacter et al. 2021 This source describes the method that was used to generate digitized wetland boundaries in Suisun Marsh for use by radio-tagged ducklings. The same wetland layers that were generated for this Source Input were used to determine how much water was available within 0.5 km and 1.0 km of Mallard and Gadwall nests, respectively, and to calculate the straight line distance between each nest and the closest wetland. Sarah H. Peterson Joshua T. Ackerman C. Alex Hartman Andrew C. Greenawalt Michael L. Casazza Mark P. Herzog 2024 publication Ornithological Applications volume 126, issue 3 n/a Oxford Academic https://doi.org/10.1093/ornithapp/duae017 Digital and/or Hardcopy 2016 2019 ground condition Peterson et al. 2024 This source describes the radio-tagging method and tracking of the ducklings in the present data release. This source also describes how salinity concentrations were determined for each wetland based on the data described in Schacter et al. (2021). Laura Askim Erik Fintel Brian Kreb Kristin Quigley 2022 publication https://nrm.dfg.ca.gov/FileHandler.ashx?DocumentID=218048 https://data.ca.gov/dataset/vegetation-suisun-marsh-2018-ds29631 Digital and/or Hardcopy 2018 ground condition Askim et al. 2022 This data source describes the vegetation polygon layer, digitized from a lidar (light detection and ranging) flight over Suisun Marsh in June 2018, that was used to generate several of the wetland habitat attributes. Copernicus 2024 Maps Data https://dataspace.copernicus.eu/ https://doi.org/10.5066/F76W992G Raster 2016 2019 ground condition Copernicus Multispectral satellite imagery was used to digitize flooded wetland boundaries, as described in Schacter 2021. Point Blue Conservation Science Unknown Maps Data https://data.pointblue.org/apps/autowater/ Raster 2016 2019 ground condition Point Blue Open water raster imagery was used to digitize flooded wetland boundaries, as described in Schacter 2021. To examine the influence of habitat variables in explaining habitat selection of managed wetlands by dabbling duck broods in a brackish marsh (Suisun Marsh, California, USA), we tracked radio-tagged ducklings (Mallard and Gadwall) approximately daily to obtain locations. We recorded the pulse rates of transmitters to aid in determining whether ducklings were alive or dead each day. We obtained duckling locations using a truck equipped with dual 4-element Yagi antennas (Advanced Telemetry Systems, Inc, Isanti, Minnesota, USA) and a null-peak system (AVM Instrument Company, Auburn, California, USA). When a duckling was detected, we used a truck-mounted electronic compass (Revolution, True North Technologies, Hudson, Massachusetts, USA) and a hand-held GPS unit (Garmin Ltd., Olathe, Kansas) to obtain 3 bearings and GPS locations that we input into triangulation software (Location of a Signal, 4.0.3.8, Ecological Software Solutions, Urnäsch, Switzerland) to calculate Universal Transverse Mercator (UTM) coordinates for each duckling. We used a single daily location from each brood (2 transmitters were deployed in each brood) as the used locations in a habitat selection analysis. For each duckling brood, we drew a 2.4 km buffer around the nest and used that buffer to generate a set of 400 available wetland locations for the habitat selection analysis. To quantify duckling movements from the nest relative to age, year, and species, we calculated the straight-line distance between every brood location and the nest where that brood hatched. For more specific details on the deployment of radio transmitters or the tracking protocol, please refer to the associated journal publications listed in the larger work section and the source inputs section. To identify the flooded wetlands that were available for duck broods and which wetlands were used, we digitized flooded wetland boundaries using multispectral satellite imagery and a GIS layer of shallow water transportation ditches (canals) for duck broods on Grizzly Island at three times during the duck breeding season (April, May, and July) from 2016 to 2019 (12 water maps), as described in Schacter et al. (2021). Because the water maps were created using high-resolution satellite imagery, each water map represents an actual date and those dates vary slightly among years (2016 to 2019) for each of the three time periods (April, May, and July). For each brood, every used and available location was associated with a unique wetland identifier and the corresponding wetland habitat characteristics from the digitized water map (Schacter et al. 2021) that most closely aligned with when that brood hatched. The wetland attributes included: 1) wetland type, 2) Dominant wetland vegetation category, 3) The percent of the wetland that contained vegetation cover , 4) Surface water salinity concentration, and 5) Flooded wetland area. More specifically, for each year (2016–2019), we assigned every flooded wetland unit to one of five wetland types (permanent, semi-permanent, and three types of seasonal wetlands) based on the timing of flooding and the duration that water was present in the wetland. We used a vegetation polygon layer, digitized from a lidar (light detection and ranging) flight over Suisun Marsh in June 2018 (Askim et al. 2022), to assign each wetland unit to one of six wetland vegetation categories for the entire study period, based on the dominant vegetation within the outer boundary of the flooded area in each wetland. We calculated the percent of the area within the overall flooded wetland unit boundary that contained emergent vegetation by taking the sum of all individual polygons within the wetland unit’s boundary that were not defined as open water or barren. We assigned a salinity concentration to the water in each wetland unit separately for April, May, and July during 2016–2019, based on the protocols and salinity concentrations that were reported in Schacter et al. (2021) and Peterson et al. (2024). We calculated the flooded area of each managed wetland unit (km2) for each time period (April, May, and July) and year (2016–2019) combination. Flooded wetland area along with the wetland habitat characteristics described above were then associated with the corresponding unique wetland units for all 12 water maps. For more specific details on the wetland habitat variables please refer to the associated journal publications listed in the larger work section and the source inputs section. All analyses were conducted using the program R. Schacter et al. 2021 Peterson et al. 2024 Askim et al. 2022 Copernicus Point Blue Unknown Duckling_Habitat_Selection_Dataset.csv Comma Separated Value (CSV) file containing data. Producer Defined Species Duckling species Producer Defined Mallard Anas platyrhynchos Producer defined Gadwall Mareca strepera Producer defined Used Defines whether the location was one used by the brood (1) or was available to the brood (0) within 2.4 km of the nest where the brood hatched Producer Defined 1 A location in a flooded wetland that was used by the radio-tracked brood Producer defined 0 A location in a flooded wetland that was available to the radio-tracked brood Producer defined Weight For statistical analyses, each data point that represented a real duckling location in a wetland (used location column equals 1) was given a weight of 1 and each available location (used location column equals 0) was given a weight of 1000 Producer Defined 1 The weight given to a used location Producer defined 1000 The weight given to an available location Producer defined Brood ID Unique identifier for each radio-tracked brood Producer Defined The unique combination of the study year and nest identification of the radio-tracked brood. Wetland ID Unique identifier for each wetland unit Producer Defined NA Wetland ID was only extracted for used locations and not extracted for the available locations Producer defined Unique alphanumeric identifier for each wetland unit Wetland Type Every flooded wetland unit for each year was assigned to one of five wetland types based on the timing of flooding and the duration that water was present in the wetland Producer Defined Winter and spring seasonal Winter and spring-flooded seasonal wetlands remained flooded from the fall (September or October) through mid to late May but typically dried out during June and were entirely dry by August 1 Producer defined Semi permanent Semi-permanent wetlands were typically flooded from September or October and retained water until at least August 1 Producer defined Winter seasonal Winter-flooded seasonal wetlands remained flooded from the fall (September or October) through early spring (March or April) but were typically dry by the end of April Producer defined Permanent Permanent wetlands were flooded year-round Producer defined Summer seasonal Summer-flooded seasonal wetlands were typically flooded during spring or summer instead of during the fall and generally retained water through August Producer defined Dominant Veg Category Each unique wetland unit was assigned to one of six wetland vegetation categories for the entire study period, based on the dominant vegetation within the outer boundary of the flooded area in each wetland Producer Defined Pickleweed Dominated by Salicornia species Producer defined Multiple veg types If the total area of vegetated polygons was more than 30% of the maximum flooded wetland area but none of the vegetation categories comprised more than 30% of the total wetland area, we assigned the vegetation category as having multiple vegetation types Producer defined Sparsely vegetated If the total area of vegetated polygons was less than 30% of the maximum wetland area, then we assigned the category to be sparsely vegetated Producer defined Phragmites Dominated by Phragmites australis Producer defined Bulrush/cattail Dominated by Shoenoplectus species (e.g., California bulrush, hardstem bulrush, Bolboschoenus maritimus (alkali bulrush) or Typha species (cattail) Producer defined Saltgrass Dominated by saltgrass (Distichlis spicata) and/or alkali heath (Frankenia salina) Producer defined Percent Veg of Unit The percent of the area within the overall flooded wetland unit boundary that contained emergent vegetation was calculated by taking the sum of all individual polygons within the wetland unit’s boundary (not the flooded wetland boundary for a given year and time period) that were not defined as open water or barren Producer Defined 0.0 100.0 percentage 0.1 Salinity Concentration Salinity concentration of the surface water in each wetland unit for each year and time period Producer Defined 0.16 64.61 parts per thousand 0.01 Wetland Area The flooded area of each wetland unit for each year and time period combination Producer Defined 0.000009 0.842469 square kilometers 0.000001 Year Study year Producer Defined 2016 2019 year 1 Water Map Time Period The water map (from Schacter et al. 2021) that was used to determine the flooded wetland area, based on which water map was closest to the hatch date of the radio-tracked brood Producer Defined April Wetland attributes come from the April water map Producer defined May Wetland attributes come from the May water map Producer defined July Wetland attributes come from the July water map Producer defined Duckling_Movement_Dataset.csv Comma Separated Value (CSV) file containing data. Producer Defined Species Duckling species Producer Defined Mallard Anas platyrhynchos Producer defined Gadwall Mareca strepera Producer defined Year Study year Producer Defined 2016 2019 year 1 Species Year Unique combination of species and study year Producer Defined Mallard_2016 Mallard tracked in 2016 Producer defined Gadwall_2016 Gadwall tracked in 2016 Producer defined Gadwall_2017 Gadwall tracked in 2017 Producer defined Mallard_2017 Mallard tracked in 2017 Producer defined Gadwall_2018 Gadwall tracked in 2018 Producer defined Mallard_2018 Mallard tracked in 2018 Producer defined Gadwall_2019 Gadwall tracked in 2019 Producer defined Mallard_2019 Mallard tracked in 2019 Producer defined Brood ID Unique identifier for each radio-tracked brood Producer Defined The unique combination of the study year and nest identification of the radio-tracked brood. Duckling Age The age of the brood (0 is the day the brood hatched) Producer Defined 0 40 days 1 Distance from Nest The distance between each brood location in a wetland and the nest where the brood hatched Producer Defined 21 8683 meters 1 GS ScienceBase U.S. Geological Survey mailing address

Denver Federal Center, Building 810, Mail Stop 302

Denver CO 80225 United States 1-888-275-8747 sciencebase@usgs.gov Unless otherwise stated, all data, metadata and related materials are considered to satisfy the quality standards relative to the purpose for which the data were collected. Although these data and associated metadata have been reviewed for accuracy and completeness and approved for release by the U.S. Geological Survey (USGS), no warranty expressed or implied is made regarding the display or utility of the data for other purposes, nor on all computer systems, nor shall the act of distribution constitute any such warranty. CSV https://doi.org/10.5066/P13NYGAI None 20260227 Sarah H. Peterson U.S. Geological Survey, Western Ecological Research Center Biologist mailing and physical

800 Business Park Drive, Suite D

Dixon CA 95620 USA Unknown sepeterson@usgs.gov FGDC Biological Data Profile of the Content Standard for Digital Geospatial Metadata FGDC-STD-001.1-1999