FIGURE 3 - available via license: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Content may be subject to copyright.
(a) Heat Map of Areal Density (ha/km 2 ) of Putative Geographically Isolated Wetlands (GIWs) across the Conterminous United States (U.S.). Map was generated using ArcGIS point density algorithms. (b) Heat map of count density (number of GIW polygons per km 2 ) of putative GIWs across the conterminous U.S. Map was generated using ArcGIS point density algorithms.
Source publication
Geographically isolated wetlands (GIWs) are wetlands completely surrounded by uplands. While common throughout the United States (U.S.), there have heretofore been no nationally available, spatially explicit estimates of GIW extent, complicating efforts to understand the myriad biogeochemical, hydrological, and habitat functions of GIWs and hamperi...
Similar publications
The current chaotic state of Indian cities in terms of lack of effective public infrastructure systems implies a failure of the practice of urban planning. One of the reasons for this failure is often a compartmentalised approach to urban planning. An integrated approach, wherein urban development and public infrastructure systems are analysed and...
The climate of the Arctic is warming rapidly and this is causing major changes to the cycling of carbon and the distribution of permafrost in this region. Tundra lakes are key components of the Arctic climate system because they represent a source of methane to the atmosphere. In this paper, we aim to analyze the geometry of the patterns formed by...
Commercial sea routes joining Europe with other cultures are vivid examples of cultural interaction. In this work, we present a serious game which aims to provide better insight and understanding of seaborne trade mechanisms and seafaring practices in the eastern Mediterranean during the Classical and Hellenistic periods. The game incorporates prob...
Climate change has highlighted the need to promote renewable energies. The offshore wind industry is growing exponentially because of some political strategies supported by various organizations, such as the European Union. The implementation of these strategies is commonly associated with significant investments, public acceptance, or achieving be...
Influenza is an acute respiratory infectious disease to which the population is generally susceptible and has an important impact on public health. Time series analysis and geospatial analysis were used to analyze the spatial and temporal distribution characteristics of influenza epidemic and its influencing factors in 103 counties in Hubei Provinc...
Citations
... These estimates are critical to understand the potential effects this decision has on the nation's water quality, in addition to motivating state and local governments to bolster existing protections where federal rulings fall short. Here, we expand upon the previous estimates of Lane and D'Amico (2016) to determine to what extent GIWs across the entire US may fall outside federal or state protections. ...
... Several data preprocessing steps were required before comparing NWI and NHD features, building on similar methods as Lane and D'Amico (2016). First, NWI classes that were not considered relevant for estimating GIWs were removed. ...
... GIWs represent 28-55% of all wetlands in Florida and South Carolina (10-35% of their wetland area), and in the northeast, Delaware and Maryland have some of the highest representation of GIWs (26-60% of their wetlands, 12-50% of their wetland area). Notably, our estimates of the area of GIWs are far larger than the 16 million acres estimated previously by Lane and D'Amico (2016). This discrepancy may be due to differences in distance thresholds used, as well as nearly a individual wetland, with columns for the wetland attribute, acreage of the wetland, latitude and longitude (WGS 1984) of the wetland centroid, distance of the wetland in meters to the nearest NHD feature, the state abbreviation, and wetland type. ...
For decades, federal protections were extended to wetlands adjacent to “waters of the US” by the Clean Water Act. In its Sackett v. EPA ruling, however, the US Supreme Court redefined the meaning of “adjacent,” eliminating protections to wetlands without a continuous surface connection to these waters (i.e., geographically isolated wetlands, GIWs). Yet it remains unclear how this continuous surface test will work in reality, where ecological connectivity often extends beyond physical connectivity. Here, we calculate the number of US wetlands that could be considered geographically isolated depending upon the distance threshold used to define isolation (ranging from 1 m to 100 m from the nearest hydrological feature). Overall, we estimate that 27–45% of wetlands, at minimum, could be considered geographically isolated using this range of distance thresholds. Over 3 million wetlands are within 1–100 m of the nearest hydrological feature, making them most vulnerable to losing prior protections from the Clean Water Act. The Midwest and Northeast have the largest share of potential GIWs within this range. Freshwater emergent wetlands and forested/shrub wetlands make up the majority of these vulnerable wetlands, though this varies by state. Roughly 47% of these wetlands are located in states without state-level protections for GIWs. Our analysis highlights the heterogeneity of risk to wetlands across the country and the scale of the uncertainty imposed by the updated Sackett definition. State-level protections that are robust to changes in federal protections are urgently needed to secure the country’s wetlands from further pollution and destruction.
... We first characterized recent wetland loss based on published information in the PPR of Canada and the United States. Compared to the U.S. portion of the PPR, prairie pothole wetlands in Canada have no federal, and relatively limited provincial, legislative protection, but generally exist in a landscape with similar agricultural pressures (Scarth, 1998;Lane and D'Amico, 2016). Recent PPR wetland loss rates in the United States are approximately 0.096 % annually with a mean size of lost basins equal to ca. 0.3 ha (Dahl, 2014). ...
... Relative to this baseline, we created four wetland loss scenarios, one based on the historical (1997-2009Dahl, 2014) loss rate in the United States (0.096 %/yr), and three based on the historical (2001Watmough et al., 2017) loss rate in Canada (0.260 %/yr). Wetland loss scenarios were calculated in SAS (SAS Institute, Cary, North Carolina) using all NWI-PPR basin polygon areas (with 'geographically isolated' basins identified by Lane and D'Amico, 2016; provided by C. Lane, March 23, 2020) exported as an attribute database from ArcGIS (Environmental Systems Research Institute, Redlands, CA). We implemented wetland loss scenarios by first generating size-class bins matching the log-normal size class distribution (mean, median, maximum) of lost basins observed in the Canadian PPR (Watmough et al., 2017). ...
... We implemented wetland loss scenarios by first generating size-class bins matching the log-normal size class distribution (mean, median, maximum) of lost basins observed in the Canadian PPR (Watmough et al., 2017). We then binned all NWI-PPR wetland polygon areas less than or equal to the maximum size lost and defined as 'geographically isolated' by Lane and D'Amico (2016) into the 20 log-normal size-class bins. We used PROC SURVEYSELECT in SAS to randomly select basins from the binned subsample, while matching the mean and median wetland sizes, until a target percent area loss (relative to baseline area) had been achieved. ...
... In this effort, we extracted h crit from the inflection point of water level recession rates as a function of stage to quantify surface connectivity which was indirectly inferred in the past based on the proximity to adjacent downstream water bodies due to lack of direct characterization method for surface connectivity Lane & D'Amico, 2016). The identified spill thresholds yield values that comport with previous studies ( Figure S3 in Supporting Information S1); in short, we are relatively confident in their use for inferring the timing, duration, and importance of surface connectivity. ...
Depressional wetlands influence the functions of wetlandscapes by storing and releasing water, providing critical habitat, amplifying carbon and nutrient cycling, and influencing microclimate. Despite persistent subsurface connectivity, depressional wetlands are surrounded by uplands so only sporadically connect via surface pathways. However, the frequency, duration, and relative importance of surface connectivity in depressional wetlands remains poorly understood, limiting quantification of their landscape functions. Using multiple years of stage variation in 67 depressional wetlands across four contrasting wetlandscapes, we observed wetland spill elevation is exceeded 10%–40% of the time, with substantial variation within and between wetlandscapes. Moreover, surface connectivity increased water loss rates by 200%–350% on a depth basis and 350%–850% on a volumetric basis compared with subsurface water loss rates. This temporally disproportionate water export suggests that short‐lived surface connectivity is crucial for aggregate landscape export of water‐borne materials and numerous hydrologic and habitat services. Contrasting water loss rates above and below spill thresholds create homeostatic feedback that stabilizes water levels near the thresholds. We explored geomorphic, climatic, and vegetative influences on hydrologic connectivity, quantified as nighttime recession rates below the spill threshold for groundwater connectivity and percent time above thresholds for surface connectivity. Groundwater connectivity was consistently greater in deeper wetlands and wetlandscape identity was the primary factor explaining variation in surface and subsurface connectivity. Our results highlight the critical role of surface connectivity in coastal plain wetlands, illustrate the heterogeneity of those wetland functions within and across wetlandscapes, and provide hydrologic benchmarks for evaluating restoration of aggregate landscape functions.
... This needed inquiry is particularly critical for headwater wetlands that serve as the headward extent of flow networks (Enviromental Protection Agency 2015; Lane et al. 2018). Given the diffuse and distributed nature of upper catchments, headwater wetlands are numerous and comprise a large proportion of total wetland area in the United States (Lane and D'Amico 2016). They are commonly directly connected to or proximal to headwater streams, which comprise the majority of stream length in the United States (Nadeau and Rains 2007). ...
... Furthermore, and although there is accumulating evidence that they play important roles in maintaining the chemical, physical, and biological integrity of downstream waters (Evenson et al. 2018a; Thorslund et al. 2018), general rules about if, how, and the degrees to which they do so remain lacking (Enviromental Protection Agency 2015; Lane et al. 2018). Additionally, these features are difficult to identify, catalog, and map due to their small size, abundance, and density (Meyer et al. 2007;Lang et al. 2012;Lane and D'Amico 2016), as well as tendency towards seasonal or intermittent flow, making them particularly vulnerable to loss (Van Meter and Basu 2015; Creed et al. 2017). ...
Many headwater wetlands are integrated into flowpath networks and can serve as sources of streamflow for downgradient waters. We demonstrate this with five years of data in vernal pool, swale, and headwater stream complexes in the Central Valley, California. Long-term United States Geological Survey data suggest that the mean flow duration from the smallest watersheds in this region, including those with vernal pool, swale, and headwater stream complexes, is ~ 85 days per year. Our data concur, indicating that the annual days of flow per year from our vernal pool, swale, and headwater stream complexes ranges from ~ 20–200, but is ~ 85 when annual precipitation is 100% of normal. Peak stages are evident first in vernal pools which then propagate sequentially downstream through swales, headwater streams, and to the Sacramento River at celerities of ~ 1-1.5 m/s, consistent with expected flood wave velocities. Geospatial analyses show that these vernal pool, swale, and headwater stream features cover > 4% of the study area. Our results suggest these systems can be significant sources of streamflow, and therefore play an important role in maintaining the chemical, physical, and biological integrity of downstream waters, which has important implications for the definition of waters of the United States subject to regulation under the Clean Water Act.
... Change in surface water extent can occur when areas temporarily store surface water following snowmelt or large precipitation events, or show change concentrated at the edges as waterbodies expand and contract over days, weeks, months, or years (Donchyts et al., 2016;Parra et al., 2021). Further, the amount of surface water storage varies greatly across different ecoregions (Poff et al., 2006;Lane and D'Amico, 2016) and the response of that surface storage to changes in precipitation and snowmelt can also be highly variable . Continual changes in surface water extent necessitate reliable, accurate, and high spatial and temporal resolution surface water products. ...
Frequent observations of surface water at fine spatial scales will provide critical data to support the management of aquatic habitat, flood risk and water quality. Sentinel-1 and Sentinel-2 satellites can provide such observations, but algorithms are still needed that perform well across diverse climate and vegetation conditions. We developed surface inundation algorithms for Sentinel-1 and Sentinel-2, respectively, at 12 sites across the conterminous United States (CONUS), covering a total of >536,000 km2 and representing diverse hydrologic and vegetation landscapes. Each scene in the 5-year (2017-2021) time series was classified into open water, vegetated water, and non-water at 20 m resolution using variables from Sentinel-1 and Sentinel-2, as well as variables derived from topographic and weather datasets. The Sentinel-1 algorithm was developed distinct from the Sentinel-2 model to explore if and where the two time series could potentially be integrated into a single high-frequency time series. Within each model, open water and vegetated water (vegetated palustrine, lacustrine, and riverine wetlands) classes were mapped. The models were validated using imagery from WorldView and PlanetScope. Classification accuracy for open water was high across the 5-year period, with an omission and commission error of only 3.1% and 0.9% for the Sentinel-1 algorithm and 3.1% and 0.5% for the Sentinel-2 algorithm, respectively. Vegetated water accuracy was lower, as expected given that the class represents mixed pixels. The Sentinel-2 algorithm showed higher accuracy (10.7% omission and 7.9% commission error) relative to the Sentinel-1 algorithm (28.4% omission and 16.0% commission error). Patterns over time in the proportion of area mapped as open or vegetated water by the Sentinel-1 and Sentinel-2 algorithms were charted and correlated for a subset of all 12 sites. Our results showed that the Sentinel-1 and Sentinel-2 algorithm open water time series can be integrated at all 12 sites to improve the temporal resolution, but sensor-specific differences, such as sensitivity to vegetation structure versus pixel color, complicate the data integration for mixed-pixel, vegetated water. The methods developed here provide inundation at 5-day (Sentinel-2 algorithm) and 12-day (Sentinel-1 algorithm) time steps to improve our understanding of the short- and long-term response of surface water to climate and land use drivers in different ecoregions.
... GIWs span many wetland types and hydrogeomorphic settings (e.g. prairie potholes in Midwestern US and Central Canada, vernal pools of New England and eastern Canada, Delmarva and Carolina Bays, playas of southwestern US and Mexico,), and have traditionally received less protection due to their apparent disconnectivity from downstream navigable waters [20,21]. They are also generally smaller than most riparian wetlands and contain water for only a part of the year [12]. ...
Wetlands protect downstream waters by filtering excess nitrogen (N) generated from agricultural and urban activities. Small ephemeral wetlands, also known as geographically isolated wetlands (GIWs), are hotspots of N retention but have received fewer legal protections due to their apparent isolation from jurisdictional waters. Here, we hypothesize that the isolation of the GIWs make them more efficient N filters, especially when considering transient hydrologic dynamics. We use a reduced complexity model with thirty years of remotely sensed monthly wetland inundation levels in 3,700 GIWs across eight wetlandscapes in the US to show how consideration of transient hydrologic dynamics can increase N retention estimates by up to 130%, with greater retention magnification for the smaller wetlands. This effect is more pronounced in semi-arid systems such as the prairies in North Dakota, where transient assumptions lead to 1.8 times more retention, compared to humid landscapes like the North Carolina Pocosins where transient assumptions only lead to 1.4 times more retention. Our results highlight how GIWs have an outsized role in retaining nutrients, and this service is enhanced due to their hydrologic disconnectivity which must be protected to maintain the integrity of downstream waters.
... Non-floodplain wetlands (NFWs) are a common type of wetlands, surrounded by uplands outside of floodplains and riparian areas (Lane et al., 2018). These waters are numerous, widely distributed, and tightly linked to socio-economic development, but their typically small size (even down to 0.01 ha; Lane and D'Amico, 2016) and shallow nature (water depth < 6 m; Ramsar Convention, 1971) leave them frequently unmonitored and unmapped. Thus, due to their minimal physical stature, NFWs remain poorly protected and vulnerable to degradation and destruction caused by both hydro-climatic and anthropogenic drivers (e.g., agricultural intensification, urban expansion, eutrophication, salinization, and invasive species) (Van Meter and Basu, 2015;Creed et al., 2017;Golden et al., 2019). ...
... Specifically, for the papers that did not specify "non-floodplain" in the abstract and conclusions, a visual inspection of the figures was conducted on a distance criterion of 10-meters away from rivers, coasts, and floodplains. This distance criterion was used to identify putative GIWs from the U.S. Fish and Wildlife Service National Wetlands Inventory, relative to the U.S. Geological Survey National Hydrography Dataset, according to Lane and D'Amico (2016). We used the above literature set to map the proportion of NFWs being studied across different continents, after identifying specific and common wetland types and their research frequency. ...
... On this basis, mutual aid with local adaption considerations (i.e., taking local conditions into considerations), especially on the dominant hydro-climatic characteristics and landscape processes, e.g., the freeze-thaw processes for polar wetlands (Rains, 2011) and irrigation practices for multipond systems (Chen et al., 2020b), is preferred, when developing novel technologies, such as hyperspectral remote sensing (Wu et al., 2019), in detecting NFWs' presence and physical characteristics. Together, these could be more reliable than the distance criterion currently in use (Lane and D'Amico, 2016), whether it is 10 meters or longer. At continental and global scales, reassembling exemplary and sophisticated datasets like the U.S. National Wetlands Inventory and Pan-European High-Resolution Layers, while integrating observations of various research focal areas (e.g., the six data categories of the integrated monitoring framework proposed by Chen et al., 2019), with regular automatic or semi-automatic updates (e.g., via Google Earth Engine Data Catalog), is recommended to reflect the impact of human activities in disparate regions and boost watershed and climate change sciences (Berrang et al., 2015;Golden et al., 2021). ...
Non-floodplain wetlands (NFWs) are important but vulnerable inland freshwater systems that are receiving increased attention and protection worldwide. However, a lack of consistent terminology, incohesive research objectives, and inherent heterogeneity in existing knowledge hinder cross-regional information sharing and global collaboration. To address this challenge and facilitate future management decisions, we synthesized recent work to understand the state of NFW science and explore new opportunities for research and sustainable NFW use globally. Results from our synthesis show that although NFWs have been widely studied across all continents, regional biases exist in the literature. We hypothesize these biases in the literature stem from terminology rather than real geographical bias around existence and functionality. To confirm this observation, we explored a set of geographically representative NFW regions around the world and characteristics of research focal areas. We conclude that there is more that unites NFW research and management efforts than we might otherwise appreciate. Furthermore, opportunities for cross-regional information sharing and global collaboration exist, but a unified terminology will be needed, as will a focus on wetland functionality. Based on these findings, we discuss four pathways that aid in better collaboration, including improved cohesion in classification and terminology, and unified approaches to modeling and simulation. In turn, legislative objectives must be informed by science to drive conservation and management priorities. Finally, an educational pathway serves to integrate the measures and to promote new technologies that aid in our collective understanding of NFWs. Our resulting framework from NFW synthesis serves to encourage interdisciplinary collaboration and sustainable use and conservation of wetland systems globally.
... For instance, in the conterminous United States (CONUS), conservative estimates indicate that headwaters constitute over 79% of the freshwater river length and drain approximately 70% of the land area (Colvin et al., 2019). Similarly, while over 50% of historical wetlands have been lost, remaining mapped wetlands are still estimated to cover 44.6 million ha or ~ 5.5% of the CONUS (Dahl, 2011), with high densities in areas such as the Prairie Pothole Region, the Mississippi Alluvial Valley, and the Atlantic and Gulf Coastal Plain ecoregions (Tiner, 2003;Keddy et al., 2009;Lane and D'Amico, 2016). ...
... In addition, the spectral signature (i.e., the variation of light reflectance as a function of wavelength) of surface water overlaps with that of shadows created by trees, stumps, and other structures (Huang C. Headwater and wetland datasets can be spatially and temporally inconsistent. The combination of uncertainties associated with image resolution, imaging dates (e.g., season, day), update intervals, cartographic conventions/interpretations, classification methods, and continued stream and wetland losses can lead to an uneven patchwork of mapped streams (Fig. 4b;Colvin et al., 2019) and wetlands (Tiner et al., 2015;Lane and D'Amico, 2016). For example, while NWI has focused more recent updates on areas of greater wetland densities (coasts and the Great Lakes), the NWI base map dates range from the 1970s to the present, with 51% of the CONUS mapped in the 1980's and 35% mapped since 2000 (as of September 2020). ...
Headwater streams and inland wetlands provide essential functions that support healthy watersheds and downstream waters. However, scientists and aquatic resource managers lack a comprehensive synthesis of national and state stream and wetland geospatial datasets and emerging technologies that can further improve these data. We conducted a review of existing United States (US) federal and state stream and wetland geospatial datasets, focusing on their spatial extent, permanence classifications, and current limitations. We also examined recent peer-reviewed literature for emerging methods that can potentially improve the estimation, representation, and integration of stream and wetland datasets. We found that federal and state datasets rely heavily on the US Geological Survey's National Hydrography Dataset for stream extent and duration information. Only eleven states (22%) had additional stream extent information and seven states (14%) provided additional duration information. Likewise, federal and state wetland datasets primarily use the US Fish and Wildlife Service's National Wetlands Inventory (NWI) Geospatial Dataset, with only two states using non-NWI datasets. Our synthesis revealed that LiDAR-based technologies hold promise for advancing stream and wetland mapping at limited spatial extents. While machine learning techniques may help to scale-up these LiDAR-derived estimates, challenges related to preprocessing and data workflows remain. High-resolution commercial imagery, supported by public imagery and cloud computing, may further aid characterization of the spatial and temporal dynamics of streams and wetlands, especially using multi-platform and multi-temporal machine learning approaches. Models integrating both stream and wetland dynamics are limited, and field-based efforts must remain a key component in developing improved headwater stream and wetland datasets. Continued financial and partnership support of existing databases is also needed to enhance mapping and inform water resources research and policy decisions.
... Low-relief, wetland-dominated landscapes are found across the United States (e.g., West Coast Vernal Pools, Midwest Prairie Potholes, Mid-Atlantic Delmarva Bays, and Southeast Carolina Bays) (Lane & D'Amico, 2016;Tiner, 2003). These wetlands are often called geographically isolated wetlands , nonfloodplain wetlands (Golden et al., 2019), or more recently, headwater wetlands (Hondula et al., 2021). ...
Hydrologic controls on carbon processing and export are a critical feature of wetland ecosystems. Hydrologic response to climate variability has important implications for carbon‐climate feedbacks, aquatic metabolism, and water quality. Little is known about how hydrologic processes along the terrestrial‐aquatic interface in low‐relief, depressional wetland catchments influence carbon dynamics, particularly regarding soil‐derived dissolved organic matter (DOM) transport and transformation. To understand the role of different soil horizons as potential sources of DOM to wetland systems, we measured water‐soluble organic matter (WSOM) concentration and composition in soils collected from upland to wetland transects at four Delmarva Bay wetlands in the eastern United States. Spectral metrics indicated that WSOM in shallow organic horizons had increased aromaticity, higher molecular weight, and plant‐like signatures. In contrast, WSOM from deeper, mineral horizons had lower aromaticity, lower molecular weights, and microbial‐like signatures. Organic soil horizons had the highest concentrations of WSOM, and WSOM decreased with increasing soil depth. WSOM concentrations also decreased from the upland to the wetland, suggesting that continuous soil saturation reduces WSOM concentrations. Despite wetland soils having lower WSOM, these horizons are thicker and continuously hydrologically connected to wetland surface and groundwater, leading to wetland soils representing the largest potential source of soil‐derived DOM to the Delmarva Bay wetland system. Knowledge of which soil horizons are most biogeochemically significant for DOM transport in wetland ecosystems will become increasingly important as climate change is expected to alter hydrologic regimes of wetland soils and their resulting carbon contributions from the landscape.
... The term vulnerable waters emerged due to their susceptibility to degradation or destruction because of the insufficiency of their mapped extent and limited regulatory protection (see Creed and others 2017). Yet vulnerable waters are often abundant watershed components within natural landscapes (Freeman and others 2007; Lane and D'Amico 2016; Allen and others 2018; Hafen and others 2020; Fesenmyer and others 2021; Messager and others 2021), with estimates suggesting they comprise up to 89% of longitudinal stream extent worldwide (Allen and others 2018) and greater than 16% of inland wetlands in the conterminuous USA (Lane and D'Amico 2016); no global estimates exist for non-floodplain wetlands. ...
... In contrast to headwater streams, no global data are yet available on the potential extent of nonfloodplain wetlands, representing a significant data gap in the effective management of these vulnerable waters (see, though, Borja and others 2020). Analyses in spatially data-rich areas such as the conterminous USA suggest that non-floodplain wetlands comprise approximately 16-23% of existing total freshwater-wetland areal extent (Lane and D'Amico 2016; this study, see Supplemental Material and Figure 3B). However, global wetland losses to date have been substantive; the USA alone has lost 50% of wetlands since European settlement, with some states having lost more than 90% (Dahl 1990). ...
Watershed resilience is the ability of a watershed to maintain its characteristic system state while concurrently resisting, adapting to, and reorganizing after hydrological (for example, drought, flooding) or biogeochemical (for example, excessive nutrient) disturbances. Vulnerable waters include non-floodplain wetlands and headwater streams, abundant watershed components representing the most distal extent of the freshwater aquatic network. Vulnerable waters are hydrologically dynamic and biogeochemically reactive aquatic systems, storing, processing, and releasing water and entrained (that is, dissolved and particulate) materials along expanding and contracting aquatic networks. The hydrological and biogeochemical functions emerging from these processes affect the magnitude, frequency, timing, duration, storage, and rate of change of material and energy fluxes among watershed components and to downstream waters, thereby maintaining watershed states and imparting watershed resilience. We present here a conceptual framework for understanding how vulnerable waters confer watershed resilience. We demonstrate how individual and cumulative vulnerable-water modifications (for example, reduced extent, altered connectivity) affect watershed-scale hydrological and biogeochemical disturbance response and recovery, which decreases watershed resilience and can trigger transitions across thresholds to alternative watershed states (for example, states conducive to increased flood frequency or nutrient concentrations). We subsequently describe how resilient watersheds require spatial heterogeneity and temporal variability in hydrological and biogeochemical interactions between terrestrial systems and down-gradient waters, which necessitates attention to the conservation and restoration of vulnerable waters and their downstream connectivity gradients. To conclude, we provide actionable principles for resilient watersheds and articulate research needs to further watershed resilience science and vulnerable-water management.
