This dataset contains data collected in the field for the field component of the 2018-20 UMRR Science in support of Restoration Forest Canopy Gap Dynamics study titled "Forest canopy gap dynamics: quantifying forest gaps and understanding gap – level forest regeneration in Upper Mississippi River floodplain forests."

These data are a component of a floodplain forest canopy gap dynamics study initiated in 2019 and funded through the US Army Corps of Engineers Upper Mississippi River Restoration, Science Supporting Restoration program. The study included two components: a geospatial component to utilize lidar to identify and map canopy gaps across multiple navigation pools (8, 9, 13, 21, 24, 26 (through Maple Island just south of Lock and Dam 26), and the lower 32 miles of the Illinois River from its confluence with the Mississippi River to Kampsville, IL) within the Upper Mississippi River floodplain and a field component to characterize vegetation in a small subset of the remotely sensed gaps. This layer provides lidar derived polygons the subset of gaps that were selected for field surveys in 2020. Gaps were selected based on a randomly stratified sample of the lidar derived gaps, with stratification occurring based on gap size (small (0.0406-0.1012 ha), medium (0.1013-0.3035 ha), large (0.3036-0.8093 ha)) and annual days of flooding within the gap (low (0.1-20 days), moderate (20.1 - 40 days), high (40.1 - 100 days)). The objective of the field study was to determine whether gaps over varying sizes and flood regimes showed variation in vegetation dynamics within the gaps. The geospatial data and scripts used for initial gap identification are available at: https://doi.org/10.5066/P9BLTSTZ. Floodplain inundation rasters that were used to derive the annual inundation rates are available at: https://doi.org/10.5066/F7VD6XRT.

These data are a component of a floodplain forest canopy gap dynamics study initiated in 2019 and funded through the US Army Corps of Engineers Upper Mississippi River Restoration, Science Supporting Restoration program. The study included two components: a geospatial component to utilize lidar to identify and map canopy gaps across multiple navigation pools (8, 9, 13, 21, 24, 26 (through Maple Island just south of Lock and Dam 26), and the lower 32 miles of the Illinois River from its confluence with the Mississippi River to Kampsville, IL) within the Upper Mississippi River floodplain and a field component to characterize vegetation in a small subset of the remotely sensed gaps. This layer provides lidar derived polygons the subset of gaps that were selected for field surveys in 2020. Gaps were selected based on a randomly stratified sample of the lidar derived gaps, with stratification occurring based on gap size (small (0.0406-0.1012 ha), medium (0.1013-0.3035 ha), large (0.3036-0.8093 ha)) and annual days of flooding within the gap (low (0.1-20 days), moderate (20.1 - 40 days), high (40.1 - 100 days)). The objective of the field study was to determine whether gaps over varying sizes and flood regimes showed variation in vegetation dynamics within the gaps. The geospatial data and scripts used for initial gap identification are available at: https://doi.org/10.5066/P9BLTSTZ. Floodplain inundation rasters that were used to derive the annual inundation rates are available at: https://doi.org/10.5066/F7VD6XRT.

This dataset contains data collected in the field for the field component of the 2018-20 UMRR Science in support of Restoration Forest Canopy Gap Dynamics study titled "Forest canopy gap dynamics: quantifying forest gaps and understanding gap – level forest regeneration in Upper Mississippi River floodplain forests."

These data are a component of a floodplain forest canopy gap dynamics study initiated in 2019 and funded through the US Army Corps of Engineers Upper Mississippi River Restoration, Science Supporting Restoration program. The study included two components: a geospatial component to utilize lidar to identify and map canopy gaps across multiple navigation pools (8, 9, 13, 21, 24, 26 (through Maple Island just south of Lock and Dam 26), and the lower 32 miles of the Illinois River from its confluence with the Mississippi River to Kampsville, IL) within the Upper Mississippi River floodplain and a field component to characterize vegetation in a small subset of the remotely sensed gaps. This layer provides field-collected centroids for the subset of gaps that were selected for field surveys in 2020. Gaps were selected based on a randomly stratified sample of the LiDAR derived gaps, with stratification occurring based on gap size (small (0.0406-0.1012 ha), medium (0.1013-0.3035 ha), large (0.3036-0.8093 ha)) and annual days of flooding within the gap (low (0.1-20 days), moderate (20.1 - 40 days), high (40.1 - 100 days)). The objective of the field study was to determine whether gaps over varying sizes and flood regimes showed variation in vegetation dynamics within the gaps. The geospatial data and scripts used for initial gap identification are available at: https://doi.org/10.5066/P9BLTSTZ. Floodplain inundation rasters that were used to derive the annual inundation rates are available at: https://doi.org/10.5066/F7VD6XRT.

These data are a component of a floodplain forest canopy gap dynamics study initiated in 2019 and funded through the US Army Corps of Engineers Upper Mississippi River Restoration, Science Supporting Restoration program. The study included two components: a geospatial component to utilize lidar to identify and map canopy gaps across multiple navigation pools (8, 9, 13, 21, 24, 26 (through Maple Island just south of Lock and Dam 26), and the lower 32 miles of the Illinois River from its confluence with the Mississippi River to Kampsville, IL) within the Upper Mississippi River floodplain and a field component to characterize vegetation in a small subset of the remotely sensed gaps. This layer provides field-collected centroids for the subset of gaps that were selected for field surveys in 2020. Gaps were selected based on a randomly stratified sample of the LiDAR derived gaps, with stratification occurring based on gap size (small (0.0406-0.1012 ha), medium (0.1013-0.3035 ha), large (0.3036-0.8093 ha)) and annual days of flooding within the gap (low (0.1-20 days), moderate (20.1 - 40 days), high (40.1 - 100 days)). The objective of the field study was to determine whether gaps over varying sizes and flood regimes showed variation in vegetation dynamics within the gaps. The geospatial data and scripts used for initial gap identification are available at: https://doi.org/10.5066/P9BLTSTZ. Floodplain inundation rasters that were used to derive the annual inundation rates are available at: https://doi.org/10.5066/F7VD6XRT.

Throughout the Upper Mississippi River System (UMRS), floods in 1993 and 2019 were record-setting events in terms of flood stage and duration of flooding. Floodplain tree species are adapted to survive moderate frequencies, intensities, and durations of inundation. These events generated interest in quantifying levels of tree mortality and presented an opportunity to study how differences in flood attributes may affect tree survivorship and forest community dynamics post-flood. In 1995, 547 plots randomly located within forested land cover were established within seven reaches on the Upper Mississippi River (Pool 4, Pool 8, Pool 13, Pool 17, Pool 22, Pool 26, Open River) and one reach of the Illinois River (La Grange Pool). In 2021, 39-46 sites (n = 342) from each of the eight reaches were navigated to using GPS coordinates from the original study and resurveyed using the same field sampling protocol used in 1995. In many cases (n = 287), the original posts and tree tags installed in 1995 could not be found and thus plots in 2021 were re-established by navigated as close to the original GPS coordinates as possible and referencing the 1995 data. In some of these cases (n = 24), plots were inaccessible (e.g., in a water body) due to landform changes that had occurred in the intervening years between field sample collections, and in these instances, plots were relocated to a nearby site with similar forest composition, elevation, and river geomorphology. Plots were re-established in order to maintain a consistent number of plots per reach for potential future monitoring, however, those that were moved were noted along with original 1995 coordinates in the provided ‘Plot_2021.csv’ file and can be included or removed in subsequent analyses depending on research questions and goals. Plots were sampled using a fixed radius sampling methodology with a radius of 10 meters, and data collected included: overstory species composition, diameters, and health status; canopy cover; seedling counts and composition; cover classes for different forest height strata; ground cover; and herbaceous cover. Seedlings, ground cover, and herbaceous cover were measured within ten randomly located 0.5 m x 0.5 m subplots.

Due to a gap in information on regeneration of hard-mast trees in the floodplain, interpretation was completed of existing 8“/pixel aerial imagery delineating a detailed hardwood forest spatial dataset within the floodplain forest of the Fish and Wildlife Upper Mississippi National Wildlife & Fish Refuge, McGregor District. The detailed map of bottomland hardwoods will allow for the identification of existing habitat and assess bottomland hardwood tree patterns to enable management of these areas.

Due to a gap in information on regeneration of hard-mast trees in the floodplain, interpretation was completed of existing 8“/pixel aerial imagery delineating a detailed hardwood forest spatial dataset within the floodplain forest of the Fish and Wildlife Upper Mississippi National Wildlife & Fish Refuge, McGregor District. The detailed map of bottomland hardwoods will allow for the identification of existing habitat and assess bottomland hardwood tree patterns to enable management of these areas.

We conducted this study in Upper Mississippi River Pools 4 and 8 to document the occurrence and cover of Phalaris arundinacea and other forest floor plants in UMR floodplain forests and relate measures of abundance of Phalaris, and community composition and diversity to estimates of flood duration which can exhibit high spatial heterogeneity. We used modeled inundation duration which allowed us to sample at randomly selected forest “places” across a large spatial scale in forest areas that had a range of tree canopy cover and basal area. We also evaluated changes in plant composition, Phalaris and wood nettle cover and silver maple seedling persistence between years relative to modeled inundation duration. From these repeated samples we also asked whether Phalaris and wood nettle cover the 1st year influenced plant composition the following year. Forest floor vegetation was quantified as percent cover within 5, 4m radius plots: one in the center and 4 others centered 25m from the center plot in the 4 cardinal directions. Percent cover was recorded to the nearest 5% for cover over 5%, to nearest 1 percent below 5%. We also noted species that were rare (1-5 individuals with less than 1%) as 0.1% cover, or numerous with insignificant cover (<1% cover), as 0.5% cover. Summing cover over all species in a plot could yield a number greater than 100, or greater than the total cover estimated for that plot if plant canopies overlapped. The species abbreviations (codes) follow the USDA Plants website conventions (https://plants.sc.egov.usda.gov/home).