Introduction

Emerald ash borer, Agrilus planipennis (Coleoptera: Buprestidae), is the most destructive exotic forest insect in North American history (Herms and McCullough 2014; McCullough 2020). As of this writing, emerald ash borer (EAB) has reached 36 states in the U.S. and five Canadian provinces, leaving hundreds of millions of dead ash trees (McCullough 2020; Sun et al. 2024). EAB is an ecological specialist whose larvae and adults feed and congregate almost exclusively on ash species (Fraxinus spp., Oleaceae). In its native east-Asian range, EAB is typically a low-density pest of ash, while ash populations outside the insect’s native range that lack coevolved defense mechanisms are particularly sensitive to infestation and mortality (Haack et al. 2015). Following detection of an established EAB population in Forest Grove, Oregon in June 2022, Oregon ash (Fraxinus latifolia) became the most recent North American ash species to be observed infested in its native range, resulting from an unidentified dispersal event of EAB to northwestern Oregon. Widespread infestation of Oregon ash is imminent: the sole publication examining potential impacts to Oregon ash found it is highly susceptible to EAB (Wagner and Todd 2015). Additionally, current field surveys of EAB in northwest Oregon indicate that EAB is well-established in the discovery locale and that EAB readily infests and damages Oregon ash, in situ (Maze, personal observation). Given the high mortality rates of other evolutionarily naive Fraxinus species following EAB establishment and the cessation of new ash regeneration in similar North American systems (Klooster et al. 2012; Ward et al. 2021), this suggests a high probability for functional extinction of Oregon ash in its native range.

Oregon ash is native to the West Coast, ranging from the southern coast of British Columbia through western Washington and Oregon, and ultimately extending to southern California, where it is mostly introgressed with native velvet ash (F. velutina) (Baldwin et al. 2012). In the Pacific Northwest, Oregon ash typically grows in moist, bottomland habitats reaching heights of 16–25 m as a late-successional climax species in Oregon’s Willamette Valley and Washington’s Puget Trough (Arno and Hammerly 1977). Oregon ash also becomes a co-dominant to dominant overstory species in early to mid-successional habitat types such as riparian corridors along tributaries of the Willamette River, Oregon and lower Columbia River where it provides shade, bank stabilization, and filtration of pollutants and excess nutrients—ecological services that maintain stream conditions favorable to salmonids (salmon and trout) and other native aquatic species (Prive 2017). Oregon ash is unique among local overstory tree species in its ability to tolerate soils inundated with water for extended periods (Frenkel and Heinitz 1987; Prive 2017). The lack of native replacement tree species filling this prospective functional void suggests exotic species are likely to fill gaps (Eschtruth et al. 2006; Hoven et al. 2017) and provide inferior levels of stream shading and other ecosystem services such as bank stabilization, wildlife habitat, inputs of large wood, and native invertebrate forage currently provided by Oregon ash (Gandhi and Herms 2010, Davis et al. 2017, Prive 2017, Kreutzweiser et al. 2018, Engelken et al. 2020). Oregon ash mortality and subsequent canopy loss induced by EAB is expected to lead to significant physical and chemical changes directly impacting water quality and altering native plant and animal community composition and successional trajectories (Bartemucci et al. 2006; Prive 2017; Klooster et al. 2018). Research on the direct and indirect impacts of EAB to North American ash populations have, up to now, focused on both urban forests and ex-urban native Fraxinus populations as far west as Colorado, with no known published research on potential impacts involving endemic Oregon ash.

A primary concern following EAB-induced Oregon ash canopy loss is the expected increase in local, incoming solar radiation and thereby stream temperatures (Malcolm et al. 2004). Water temperatures above 16 °C are particularly problematic for salmonids due to reduced dissolved oxygen levels (Bjornn and Reiser 1991). Macroinvertebrate prey species that juvenile salmon depend upon also suffer reduced fitness when streams warm (Allan 2004). Additionally, salmonid disease outbreaks and mistimed migrations are positively correlated with increasing water temperature, leading to further declines in salmonid fitness and population size (Bjornn and Reiser 1991).

Stream temperatures in the Pacific Northwest are frequently above the state water quality standards designed to protect sensitive aquatic species, including Endangered Species Act (ESA)-listed salmonids (salmon and steelhead trout). Oregon’s Department of Environmental Quality (DEQ) addresses this problem by develo** pollution reduction plans designed to reduce thermal inputs to impaired waterbodies. A key driver of increased stream temperatures in the Willamette Basin is loss of streamside shading and concomitant solar loading (DEQ 2006). System potential shade—the maximum proportion of solar radiation that is attenuated or scattered before reaching the waterbody—is achieved when the riparian plant community has reached its mature, undisturbed condition in which vegetation heights are at, or near, their expected potential, resulting in the maximum effective shade for the stream. For streams in the lower Willamette Basin where our study is situated, DEQ identifies restoration and protection of riparian vegetation as the primary methods for increasing stream shading and mitigating increased stream temperatures, and bases the nonpoint source load allocations (i.e., the allowable amount of a pollutant, in this case heat) on achieving system potential shade conditions.

Currently, no models or data exist for predicting landscape level impacts following widespread mortality of Oregon ash, which we hypothesize will lead to increased solar loading of local waterways bearing ESA-listed salmonid species. Here, we integrate field survey data estimating canopy contribution of Oregon ash with LiDAR-based canopy cover data and use these data to parameterize models of solar loading to investigate impacts to salmonid-bearing waterbodies in the Portland, Oregon metropolitan area before and after EAB-induced functional extinction of Oregon ash.

Methods

Study sites

The present study involves two salmonid-bearing waterways in the metropolitan area of Portland, Oregon that currently support healthy populations of Oregon ash: (1) the Columbia Slough is a slow-moving, tidally-influenced, and heavily leveed water body running parallel to the Columbia River and draining 132 km2 while flowing 33 km westward from Fairview Lake in Troutdale, Oregon to the confluence of the Willamette and Columbia rivers (Columbia Slough Watershed Council 2023), through mostly industrial and natural areas. (2) Johnson Creek begins in an agricultural region east of Portland, receiving tributaries and draining 138 km2 while flowing westward for 42 km through agricultural, natural, suburban, urban, and industrial habitats before emptying into the Willamette River near Milwaukie, Oregon (Lee and Snyder 2009). Both waterbodies serve as off-channel refuge along the Willamette and Columbia rivers to migrating salmonids. Even though both waterbodies may seasonably be warmer than what is optimal for salmonids and have been identified by the State of Oregon as impaired due to high water temperatures (DEQ 2022), they are still cooler than the Willamette River. Given Portland’s location at the confluence of the Willamette and Columbia rivers, all salmonids that spawn within the Willamette basin and Columbia basin upstream of Portland, pass through or adjacent to Portland. Accordingly, the Columbia Slough and Johnson Creek provide valuable habitats for all migrating salmonids, not only those that spawn or rear in these two waterbodies. Each of these waterways can be approximately bisected into a western reach that is dominated by urban landscape features and an eastern reach dominated by exurban and rural landscape features (Fig. 1). As such, we separately analyzed these west and east reaches.

Fig. 1
figure 1

Solid black lines highlight the western reaches and the dashed black lines highlight the eastern reaches of the modeled waterbodies, Columbia Slough and Johnson Creek, located in the Portland, Oregon metropolitan area. Waterbodies were modeled for solar loading and Oregon ash canopy loss using LiDAR-based canopy cover data along with field survey data estimating canopy contribution of Oregon ash to parameterize models of solar loading impacts to salmonid-bearing waterbodies before and after EAB-induced functional extinction of Oregon ash. Portland city limits are shown in dashed brown lines

Full size image

To randomize geographic sampling, we used ArcGIS 10.7.1 software (ESRI 2019) to subdivide our study’s waterbodies into 30 m lengths and assign a unique ID to each segment. These IDs were assigned random numbers using the ‘random()’ function in Microsoft Excel, and sequentially sorted by random number. The first 5% of rows for each waterbody reach were selected as sample sites. We wanted to evenly split our sample sites between north and south banks as both waterbodies have a primarily east–west orientation and so used the same randomization procedure to evenly assign sample sites to each of these banks. Finally, these sampling sites were exported back to ArcGIS, and geoprocessing functions allowed us to create 30 m × 30 m quadrats adjacent to these waterbody reaches/segments (quadrat design is further explained in the next section and illustrated in Fig. 2).

Fig. 2
figure 2

Illustration showing) a 30 m × 30 m quadrat and 30 m × 10 m belt transects wherein field surveys were conducted to characterize the canopy contribution of Oregon ash; b the LiDAR-derived tree canopy and height data represented as textured green. The GIS features that were used to model solar loading included c modeling nodes (solar loading quantification points; diamonds) located every 25 m along the center of the waterbody and d land cover sampling points spaced every 3 m in a star pattern around each model node to provide land cover inputs (vegetation height, canopy density, human structure type) used to calculate the proportion of the incoming solar loading that is attenuated by these variables; e the 10 m × 10 m riparian grid (dashed lines) used to represent Oregon ash in the solar loading model; and f randomly selected grid cells (black cross hatching) representing areas with canopy loss in the "Ash Loss" modeling scenario

Full size image

Field surveys to estimate canopy contribution of oregon ash

Field surveys served two objectives: (1) to quantify the proportional contribution of Oregon ash to the riparian canopy to forecast the impacts of canopy loss, and (2) provide precise counts of Oregon ash while characterizing the distribution of all co-dominant and dominant riparian shrubs and trees. Our survey focused on 30 m-wide riparian zones directly above ordinary high-water of both north and south banks along the full length of the waterbodies—a total potential survey area of 2.05 km2 and 2.59 km2 for the Columbia Slough and Johnson Creek, respectively. We subsampled the total riparian area by surveying the randomly assigned 30 m × 30 m quadrats described above (Fig. 2). We further subdivided each quadrat into three 10 m × 30 m belt transects (“transects”) running parallel to the waterbody to provide greater resolution for two reasons: (1) vegetation growing near the water’s edge typically provides more shade than plants growing further from the waterbody, and (2) fine scale habitat characteristics such as soil nutrient and moisture content can vary close to streams and may influence species distributions (Giller and Malmqvist 1998). Thus, these transects are our sampling units. We surveyed 348 transects from 116 quadrats along the Columbia Slough, and 429 transects from 143 quadrats along Johnson Creek, representing 5.3% and 5.2% of the total defined riparian areas, respectively.

Prior to implementing field surveys, we conducted a desktop analysis using recent aerial photographs (Oregon Metro 2020) to visually identify sampling quadrats where Oregon ash would be excluded from field surveys due to structural or other land-use factors such as parking lots and roads (34 quadrats identified for this category). These sites nonetheless provided canopy cover data for the model, included as 0% canopy.

Field surveys were then conducted when leaves were present, between July and October 2020 (and prior to detection of EAB in the Pacific Northwest). Vegetation was functionally categorized within each transect into overstory (tallest tree cohort within transect), understory (trees > 2.0 m but distinctly below overstory species), and shrub (< 2.0 m). When the same species occupied two or more functional layers it was separately recorded within each category. Beyond these general procedures, field survey methods varied among sites depending on the conditional context involving Oregon ash, as follows:

Condition 1: No Oregon ash

When no Oregon ash was present within a transect, we noted its absence and conducted a rapid vegetation assessment to inform future mitigation and restoration efforts by recording up to five of the most abundant species within overstory, understory, and shrub functional layers.

Condition 2: overstory exclusively Oregon ash

When Oregon ash was the only overstory species within a transect, we recorded the number of Oregon ash boles, and measured their height ranges within each transect using a laser rangefinder. As in Condition 1, we documented up to five co-dominant species within the understory and shrub layers.

Condition 3: co-dominant overstory

When Oregon ash was co-dominant with other species in the overstory, we recorded the number of Oregon ash boles, measured their height ranges, and estimated percent canopy contribution as the average of two independent visual estimates. Up to five co-dominant species within the understory and shrub layers were recorded.

Canopy composition analysis

We used ArcGIS and first-return LiDAR data collected in 2019 to generate a riparian canopy model, which quantified canopy heights, distribution, and the total canopy cover within each transect. The transect-specific canopy cover results were combined using ArcGIS with the associated field surveys to estimate the total canopy area, as well as the proportion composed of Oregon ash. Using R Statistical Software (v4.2.2; R Core Team 2022) we investigated how the proportional contribution of Oregon ash varied according to proximity to each waterbody (0–10 m, 10–20 m, 20–30 m), in addition to waterbody side (north vs. south bank). Also, we were interested in whether the surrounding land-use affected ash canopy contribution (western urbanized sections versus eastern suburban/ex-urban sections).

Effective shade, solar loading models

To assess the potential impacts of riparian canopy loss along the Columbia Slough and Johnson Creek, we modeled solar loading and quantified the effective shade using remote sensing data. Effective shade is the proportion of solar radiation that is attenuated or scattered before reaching the stream by the cumulative biotic and abiotic landscape features (Fig. 3). To calculate effective shade, we modeled the incoming solar load and the solar load blocked by these landscape features using the shade module of Heat Source version 26 (Michie et al. 2021), a publicly available computer model used and maintained by Oregon DEQ (Boyd and Kasper 2003). We modeled solar loading and calculated the effective shade for both waterbodies from July 1 to August 31 at 15-min timesteps. We summarized these results as the July–August mean at each modeling node: the summertime period when stream temperatures in the Portland area tend to be greatest.

Fig. 3
figure 3

Illustration of the modeling approach for solar loading and effective shade calculation. S1 is the incoming solar loading based on the day of the year, position of the sun, and elevation of the site. S2 represents the solar load that reaches the surface of the stream (modeling node, Fig. 2c) accounting for any attenuation (dashed lines) due to riparian or topographic shade

Full size image

The model relies on GIS inputs to characterize the surrounding topography and land cover features (including vegetation canopy cover) that affect the amount of solar radiation reaching the stream surface. For modeling effective shade of the Columbia Slough and Johnson Creek, these GIS inputs were compiled using TTools version 9.0 (Michie 2021), a GIS tool that assembles stream channel and land cover geospatial data for input into the Heat Source model.

For both waterbodies we modeled the solar loading every 25 m along the center of the channel (Fig. 2c). For each modeling node, we sampled the surrounding topography and land cover within 75 m of the center of the waterbody, in 3 m increments (Fig. 2d). Model parameters are described in detail in Table 1.

Table 1 Inputs used to model solar loading, including the datasets used to characterize land cover surrounding the modeled streams and the Heat Source modeling parameters. See the References section for links to complete GIS metadata
Full size table

The GIS vector layers in Table 1 were converted to raster datasets. These raster layers were then combined to create unique land cover codes that represent both the features on the landscape (e.g., canopy type, impervious areas, open water, wetlands, etc.) as well as the height above the ground surface of any features (e.g., trees or buildings) at that location. Separate land cover layers were created for the three modeling scenarios described below. The raster layers representing land cover codes had a pixel resolution of 0.91 m.

Modeling scenarios

We modeled three scenarios as part of this effort to compare conditions on the landscape and concomitant effects on solar loading:

  1. (1)

    Current Conditions: an assessment of riparian shade as represented in 2019 LiDAR data, before the functional extinction of Oregon ash due to EAB.

  2. (2)

    System Potential: a hypothetical future condition representing a maximum of potential riparian shade. In this scenario, all existing canopy vegetation from 2019 is assumed to have reached full-grown heights and all areas with no tree canopy are assumed to have full-grown trees. This scenario represents the maximum possible effective shade for each waterbody in the absence of human impacts. It illustrates what is possible in terms of riparian shade and serves as the restoration target adopted by state regulators.

  3. (3)

    Ash Loss: a model of riparian shade based on 2019 conditions, but assuming complete removal of Oregon ash from the landscape due to EAB.

To simulate the impact of the complete loss of Oregon ash, we modified the “Current Conditions” scenario described above by removing portions of the existing riparian canopy based on the ash abundance estimated by the field surveys. In ArcGIS we subdivided the three 10 m × 30 m belt transects every 10 m to create areas that are approximately 10 m × 10 m (Fig. 2e). The 10 m × 10 m areas were developed to approximate the canopy extent of a fully grown Oregon ash tree. We selected 10 m × 10 m as a reasonable estimation of the canopy size of a single Oregon ash tree based on observations during our field surveys and as informed by plant nursery professionals (Plantas nativa, personal communication). We then randomly selected a subset of these 10 m × 10 m areas that currently have tree canopy present for each model run of the “Ash Loss” scenario (Fig. 2f). We selected a random number of these 10 m × 10 m areas based on the relative proportion of Oregon ash abundance observed in the field surveys. Abundances of Oregon ash varied by waterbody, stream bank (north vs. south), distance from the channel (0–10 m, 10–20 m, or 20–30 m), and location along the waterbody (east vs. west). The selected areas were modeled as having no trees to simulate the loss of ash trees. Simulating this canopy loss was done by modifying the model input of the raster land cover codes to model these areas as having a height of zero meters. We generated and modeled 100 iterations of the “Ash Loss” scenario to evaluate the loss of riparian canopy by EAB.

Model calibration

Riparian vegetation attenuates less than 100% of light; to account for this the Heat Source model includes a canopy density parameter that represents the proportion of incoming solar radiation that is blocked by vegetation. The canopy density values used in the initial model runs were based on the vegetation characteristics described in the Lower Willamette temperature Total Maximum Daily Load (TMDL; DEQ, 2006) and included a 75% density value for deciduous vegetation and an 80% density value for coniferous vegetation.

We refined the canopy density parameter using riparian canopy cover measurements recorded as part of the City of Portland’s watershed monitoring program (Portland Area Watershed Monitoring and Assessment Program; City of Portland 2023a) and the Bureau of Environmental Service’s Restoration Monitoring Program (City of Portland 2023b). Densiometer readings were collected and converted to canopy cover using the methods of Lemmon (1957). We found that the field-measured canopy cover values were typically substantially higher than the TMDL canopy density values; particularly in riparian areas dominated by conifers. For the purposes of this assessment, we used the 25th percentile canopy cover values to better reflect measured field conditions. As a result, both density values were increased, with the density value for deciduous canopy of 78% and a density value for coniferous canopy of 90%.

Results

The field surveys revealed spatial variation in the abundance of Oregon ash along the two surveyed waterbodies, both longitudinally (along the waterway edge) and laterally (distance from the waterway edge). Within the riparian area of the Columbia Slough, we found that there was a greater proportional abundance of Oregon ash along the western portion of the waterbody with the 0–10 m transect closest to the channel having the highest proportion of Oregon ash (Fig. 4). Conversely, we found the riparian canopy along the eastern portion had a considerably lower proportional abundance of Oregon ash. The 0–10 m transects closest to the channel had 3–4 times less canopy composed of Oregon ash when compared to the western portion of the Columbia Slough.

Fig. 4
figure 4

Mean proportion of the riparian canopy composed of Oregon ash along the Columbia Slough and Johnson Creek based on field survey observations

Full size image

We observed a similar pattern along Johnson Creek: a greater abundance of Oregon ash in the 0–10 m transects closest to the stream channel. In contrast to the Columbia Slough, we found that Oregon ash composed a greater proportion of the riparian canopy along the mostly rural, eastern portion of Johnson Creek (Fig. 4).

The results of our solar loading modeling highlight how riparian shade varies along our two modeled waterbodies. The “Current Conditions” scenario illustrates that neither of the two waterbodies are fully shaded; that is, the modeled effective shade under the “Current Conditions” scenario is less than that of the “System Potential” scenario (Fig. 5). Additionally, the “System Potential” scenario demonstrates the differences in the amount of solar loading that can be attenuated by riparian vegetation along a waterbody. The western portion of the Columbia Slough is relatively wide (mean channel width of 50 m)—even with a mature riparian canopy along both stream banks (“System Potential”), the greatest possible effective shade would be less than 25% (Fig. 5A). The eastern portion of the Columbia Slough is narrower (mean channel with of 16 m), allowing riparian vegetation to attenuate a greater proportion of the incoming solar load (more than 75% in some areas; Fig. 5A). Conversely, we found substantially lower variability in the maximum potential effective shade along Johnson Creek (mean channel width of 7 m along both east and west segments), allowing for more shading of the stream channel (Fig. 5B). From the mouth to river kilometer 40, the effective shade ranged from approximately 75–90% in the “System Potential” scenario (Fig. 5B).

Fig. 5
figure 5

Results of the three modeling scenarios summarized along the length of each waterbody (500 m rolling averages) for A Columbia Slough and B Johnson Creek. The three scenarios include: (1) System Potential is a hypothetical scenario which represents the amount of solar loading attenuated (effective shade) by fully mature riparian vegetation, (2) Current Conditions which represents effective shade under 2019 canopy conditions, and (3) Ash Loss scenario which shows the range of effective shade conditions assuming complete loss of the existing Oregon ash tree canopy

Full size image

Our models indicate that EAB-induced loss of Oregon ash in the riparian canopy would increase solar loading to the Columbia Slough and Johnson Creek (Table 2) with the magnitude of increase based on the current extent of ash in the canopy, which in turn correlates to land use, as well as waterbody characteristics. In the Columbia Slough, the western, more urban and wider portion would experience a 1.8% increase in summer solar loading, while the narrower eastern portion of the Columbia Slough would see a slightly greater increase of 2.7% (Table 2).

Table 2 Modeled percent increase in solar loading compared to current (2019) conditions due to the loss of Oregon ash from EAB across the 100 model runs
Full size table

The eastern, exurban, and rural portion of Johnson Creek would see the greatest increase in solar loading during the summer with a mean increase of 23.7% (Table 2; Fig. 5A). A smaller increase in solar loading of 13.1% would be expected along the western, urban portion of Johnson Creek (Table 2).

Discussion

Our results suggest the decrease in canopy cover in riparian corridors and the resultant increase in solar loading due to the EAB-induced loss of Oregon ash will have significant impacts to water quality and aquatic habitat of the Columbia Slough and Johnson Creek. Given the abundance and distribution of Oregon ash and its susceptibility to EAB, it is expected that low elevation riparian and wetland habitat and water quality in the Willamette Valley and lower Columbia River will rapidly degrade following establishment of EAB. Ash mortality will result in increased water temperatures (Broadmeadow et al. 2011, Engleken and McCullough 2020), affecting an array of aquatic organisms at multiple trophic levels. EAB in the Pacific Northwest, west of the Cascade Mountain range, presents a forest pest worst-case scenario: the loss of a dominant and largely functionally irreplaceable riparian and wetland tree species that provides critical habitat for both ESA-listed and non-listed species and water quality ecosystem services.

Our findings highlight that the water quality impacts of Oregon ash loss will depend not only the abundance of Oregon ash along the waterbody, but also on the characteristics of the waterbody and, unsurprisingly, the location of the ash canopy. The Columbia Slough illustrates this dynamic: we observed the greatest abundance of Oregon ash along the western portion of the Columbia Slough—over half of the tree canopy on the north bank is composed of Oregon ash—however, even with this riparian canopy, this portion of the waterbody is challenging to shade due to its relatively wide channel. Despite the abundance of Oregon ash along this part of the Columbia Slough, the increase in solar loading due to the loss of riparian canopy would be relatively small (1.8%; Fig. 5A). Conversely, we observed a lower abundance of Oregon ash along the narrower eastern portion of the Columbia Slough, particularly in the transects closest to the channel. The potential loss of riparian canopy along this portion of the Columbia Slough would result in a somewhat greater increase in solar loading (2.7%, Fig. 5A).

Our field surveys indicate that the riparian canopy along Johnson Creek is composed of less Oregon ash than the Columbia Slough; however, the anticipated impacts of canopy loss due to EAB is substantially greater (see shaded areas of Fig. 5B, which represent loss of effective shade). Additionally, our models indicate that the loss of riparian shade would not be observed equally along Johnson Creek. Both waterbodies have an east–west orientation, but Johnson Creek has a narrower channel, allowing for more shading of the stream channel compared to the Columbia Slough. Johnson Creek highlights the importance of where the riparian canopy is composed of Oregon ash: proximity to the stream channel and stream side. Trees closer to the channel and trees on the south side of the waterbody are better able to attenuate solar loading. We see a smaller increase in solar loading along the western portion of Johnson Creek due to the lower abundance of Oregon ash on the southern bank. Despite the overall lower abundance of Oregon ash, given the size of the waterbody and the location of the existing Oregon ash canopy, the loss of this riparian canopy would result in a greater increase in solar loading to Johnson Creek compared to the Columbia Slough (Fig. 5).

While we focused our study on riparian corridors, Oregon ash provides a similar shading function in many forested wetlands in the Willamette Valley (although many of these forested wetlands lack surface water during the driest and warmest times of the year). The magnitude of ecological impacts in both habitats will largely be determined by what plant associations establish post-invasion (Van Driesch and Reardon 2014). Oregon ash is unique, regionally, as an early-seral to climax species and in its ability to form pure stands and maintain a forested state in wetlands and floodplains as the dominant species at all stages of stand development (Prive 2017); although with reduced germination on scoured, more mineral soils and moderate shade tolerance, Oregon ash is often a late-seral species (Fierke and Kauffman 2005; Prive 2017). In the past 90 + years, however, local, successional dynamics among tree species are less clear in wetlands and riparian areas that now rarely experience high-intensity flood events, including the bulk of the Willamette River basin and Puget Trough that were hydrologically modified due to large-scale water-control efforts beginning in the 1930s (Griffith 1983). On wet sites with less scour and with a higher organic soil component, Oregon ash is often a colonizing tree species, and is one of few tree species, locally, that can develop into a closed canopy forest in these conditions (Prive 2017). Other local tree species, such as Oregon white oak (Quercus garryana), black cottonwood (Populus trichocarpa), and alders (Alnus rubra and A. rhombifolia) are shade intolerant, and currently unlikely to persist in later seral stages in some settings (Prive 2017); although the future competitive absence of Oregon ash on the landscape makes their future relative abundance less certain.

The loss of Oregon ash in riparian corridors and wetlands where it is the dominant or co-dominant plant species may have other hydrologic impacts, as well. In other North American wetlands where ash is the dominant or co-dominant tree species, such as black ash (F. nigra) wetlands in the upper Midwest (MacFarlane and Meyer 2005), research suggests that large-scale dieback caused by EAB would result in higher water tables and a conversion to graminoid to shrub dominated systems (Pavik et al. 2012; Slesak et al. 2014; Looney et al. 2015) with significant changes to food webs and habitat structure (Youngquist et al. 2017) and potential for altered nitrogen cycling (Toczydlowski et al. 2020). In the wetlands and floodplains in the Willamette Valley, especially those with the longest inundation periods and highest water tables, we expect to see a broad disruption of successional dynamics and, at many sites, a conversion to open, shrub dominated systems (Pavik et al. 2012) or to wetlands invaded by exotic reed canary grass (Phalaris arundinacea) (Fierke and Kauffman 2005; Prive 2017), along with occasional slough sedge-dominated (Carex obnupta) wetlands, especially where Oregon white oak canopy remains.

Although the modeled increases in solar loading may seem modest, it is important to note that both waterbodies in our study regularly exceed Oregon’s water quality criteria for temperature throughout the summer (City of Portland 2019). Current daily maximum water temperatures of both waterbodies are often warmer than what is optimal or tolerated by ESA-listed salmonids. Elevated water temperatures have been found to decrease the growth rates of rearing juvenile salmonids, while also resulting in an increase in predation vulnerability when compared to juveniles reared in cooler water (Marine and Cech 2004). Adult salmonids are also impacted by warm water temperatures, with increased rates of prespawn mortality (Bowerman et al. 2017) and slower migration rates (Goniea et al. 2006). Migrating adult salmonids have been found to rely more extensively on refuges of cold water, such as colder tributaries, when migrating through warm waterbodies (Goniea et al. 2006). Both the Columbia Slough and Johnson Creek are essential salmonid habitat and as noted above, cooler than the Willamette River during the summer (DEQ 2020), providing valuable refuges to migrating salmonids. Any increase in solar loading to these waterbodies would contribute to increased water temperatures that are less hospitable to these endangered and threatened aquatic species. These increases in water temperature will have implications for both federal and state regulators and local municipalities and other entities seeking regulatory compliance; putting the latter at increased risk of non-compliance, in addition to threatening ESA-listed salmonid populations.

Although the results of our study in and adjacent to Portland, Oregon suggest significant local impacts to water quality from EAB-induced loss of Oregon ash in riparian habitat, our study systems are relatively lacking Oregon ash compared to many other, typically more hydric, low-elevation forested wetlands and tributaries of the Willamette River and lower Columbia River where Oregon ash is often the sole tree species. While compensatory growth of other tree species in our study waterways is expected to eventually mitigate some EAB-induced increase in solar loading (Costilow et al. 2016), this compensatory growth is expected to be uneven given the distribution of Oregon ash, recruitment of exotic species in mortality gaps (Eschtruth et al. 2006; Prive 2017), and habitat requirements of other native tree species that may reduce their ability to fill niches left open by Oregon ash. These factors suggest the loss of riparian shade, post-EAB establishment, will be significantly greater in many areas and across basin-scale watersheds than our results indicate. The loss of Oregon ash canopy throughout the greater Willamette River basin and elsewhere upstream of Portland will result in an increase in solar loading to local waterways, ultimately increasing the temperature of mainstem waterways, downstream.

Since the discovery of the first known, established West Coast EAB population in Forest Grove, Oregon, in June of 2022 (Oregon Invasive Species Council 2022), local governments and organizations, as well as state and federal agencies and departments, have sought to predict the impacts of EAB and mitigate impacts on the West Coast. While initial risk assessments focused on potential impacts to urban street trees and property values, we continue to emphasize that efforts must be directed toward characterization of the risk and consequences to water quality and habitat caused by establishment of EAB in western Oregon and Washington. Surveys and early detection must be prioritized so that nascent EAB population growth is slowed, and the rate and severity of impacts and the associated increased likelihood of water quality-related regulatory non-compliance, are reduced. The lessons learned from impacts of EAB in natural areas east of the Continental Divide should be applied in both natural and rural areas in the Pacific Northwest. Efforts should be made by local agencies, watershed councils, and soil and water conservation districts to augment restoration and natural area enhancement planting palettes with alternate species best approximating mature Oregon ash structure, however, as noted above, options are limited. Following EAB establishment and Oregon ash mortality, it is unlikely that any native tree species will successfully replace Oregon ash, via plantings or passive recruitment, in the most seasonally saturated, low-elevation soils in the Pacific Northwest (Prive 2017).