The Relationship Between Urbanization and Stream Degradation

Stream Degradation and King County' Salmon Population

King County in Washington State is home to some of the most significant spawning beds in the nation for several major species of salmon, such as the endangered steelhead and Chinook (or King) species. As a result of Washington's urbanization over the preceding years and decades, the purity and integrity of King County's streams and rivers have been degraded and the waterways contaminated (Morley, Karr, 2002). Currently, the Department of Natural Resources' Water and Land Resources Division is working on numerous projects to reclaim and maintain the streams and rivers of the County (Salmon and Trout Topics: Recovery, 2016). King County's own Wildlife Program Publication offers an indication of how important it is to maintain clean streams and waters for the salmon population. This study aims to answer the three-fold question: Has stream degradation affected salmon habitats in King County Washington state (specifically the habitats of king and steelhead salmon), and if so, how great has it affected the habitat; and what is the correlation, if any, between the rise of urbanization in Washington and the depletion of the salmon population? The time frame under consideration is 1970 to 2010.

The purpose of this study is to identify solutions that can be implemented to help reduce the effect of stream degradation. Stream degradation is the result of a number of circumstances: pollution, erosion (from storm water run-off of impervious surfaces, to agricultural impacts and waste water treatment facilities), temperature changes in streams/rivers, dams, shore armoring, and other forms of environmental destruction. Urbanization plays a large role in the advancement of Washington State in economic and social terms -- but its effect on the salmon population is one that still needs to be better understood (source). Thus, the potential value of this study may appeal to urban developers, urban planners, harvesters, and fish and wildlife agencies in other areas of the world that are impacted by the need to adopt better and greater conservation efforts in order to minimize the impact and/or risk of urban development on the surrounding natural habitat.

By answering the research question, this study expects to draw correlation between the rise of urban development in Washington and the degradation of streams in King County. It is hypothesized that population growth in the area positively correlates with the decline of king salmon as well as steelhead (both are protected species, according to the West Coast Region of the National Oceanic and Atmospheric Administration), with the alteration of stream flows and levels serving as the main culprit for the decline in these species' population (West Coast Salmon & Steelhead Listings, 2015).

The first part of the research question will be answered by performing a qualitative meta-analysis of the available and relevant literature on stream degradation and salmon habitats/populations in King County. The second part of the research question will be answered through the meta-analysis of available quantitative statistical data on the area's aquatic life habitats from studies that have utilized descriptive statistics with an interval measurement level (Greasly, 2008, p. 7). To answer the third part of the research question, the study will perform bivariate correlation analysis using variables gathered from the available relevant census data.

Literature Review

The literature for this review was collected using a variety of keyword searches in online databases. Relevant literature was culled from search results by assessing abstracts for information pertinent to the parameters of the study.

Robinson, Newell and Marzluff (2005) have shown that King County has been heavily impacted by urbanization and the expansion of the human population in terms of development of the landscape and natural habitats, putting this process in conflict with the natural processes of the watershed regions and the natural habitats that are supported by them. The study is significant in that it shows correlation between urbanization and stream degradation and is representative of the body of literature as a whole identified over the course of this review.

Vanderhoof, Stolnack, Rauscher, and Higgins (2011) and Jensen (2012) show that growth of the human population in recent years has been mainly confined to urban areas in an attempt to contain development and implement "smart growth" procedures that minimize the impact of the urbanization on rural areas and regions close in proximity to them. Their studies of these policies indicate that some sense of the impact of urbanization in King County has been ascertained by policy makers concerned about the natural habitat and settings of its fish and wildlife populations.

Morley and Karr (2002) measured the biological integrity of urban streams in the Puget Sound Basin in their study of the impact of urban growth on the waterways in King County and found that biological integrity diminished as land-cover urbanization increased within certain areas of the Sound.

In terms of effecting strategies for the conservation of streams and rivers in King County, a number of studies and policy papers illustrate the avenues available to the County.

Booth, Hartley and Jackson (2002) note in their study of the relationship between forest cover, impervious-surface area and the mitigation of stormwater impacts in King County that there are a number of factors that play in part in the degradation of aquatic systems in the area and that "costly structure retrofits of urbanized watersheds" can actually add to the problem of degradation rather than solve it: for example, such retrofittings "cannot restore the predevelopment flow regime or habitat conditions" that existed prior to urbanization; the "widespread conversion of forest to pasture or grass in rural areas" actually leads to more degradation of aquatic systems "even when watershed imperviousness remains low" (Booth, Hartley, Jackson, 2002, p. 835). The study indicates that the same aims cannot be achieved in both developed and developing areas; those that are already developed must be addressed as-is and not with the intention of reverting the landscape to some prior condition (which is virtually impossible despite the not-for-lack-of-trying strategies adopted by the County), while in developing areas there still remains the possibility of integrating risk-mitigating strategies in order to save waterways.

The County's own publication Living with Salmon in King County describes a number of actions that "can enhance streams for salmon" including disposing of toxic substances by calling the Health Department Household Hazards for support, reducing the amount of lawn chemicals/fertilizers applied to one's property, reducing one's lawn area in an effort to preserve natural vegetation (i.e., to landscape with native plants), to fence in waterways so livestock cannot trample them or pollute them (and "stir up sediment, covering the gravels in which fish spawn and suffocating their eggs") (p. 2). The publication describes the spawning cycles of the two types of salmon focused on in this study, the Chinook and the steelhead: the king salmon "are the largest of the Pacific salmon and reach up to 80 lbs" and they spawn in stream and rivers that are large and fast flowing; upon hatching from their eggs, king salmon migrate to the Pacific where they stay for a number of years before returning to the source of their origin for spawning (Austin, Stenberg, 2012, p. 3). Steelhead, the most popular salmon in King County, are smaller, averaging at 10-20 lbs; they spawn at least once in their lifetime and may spawn multiple times; unlike king salmon, steelhead spawn in shallow streams where the rate of flow is only moderate. Steelhead spawn from December to March; king from July to October. Given that the "survival rate of eggs is low," the seasons in which spawning is conducted is important within the overall context of urbanization -- for the activities of individuals that use and/or impact water or waterways, fertilizers and/or chemicals, will inevitably alter the nature of the habitat where the eggs must live: "With current management practices and habitat losses, the odds become stacked against many of our salmon stocks. Out of an estimated 1,000 stocks of native salmon in the Pacific Northwest, 106 are already extinct and another 314 are at risk" (Austin, Stenberg, 2012, p. 4).

Luchetti et al. (2014) have also assessed the land use effects and regulatory effectiveness of King County policy on streams within the rural watershed regions of the area. Their study examines the extent to which regulatory practices associated with development in King Country have impacted the conservation of streams. Their findings are consistent with those of Booth et al. (2002) and Morley and Karr (2002) in that contamination levels are a going concern and remain problematic for fish and wildlife in the region. Whether regulatory efforts are having a positive impact on the region, however, remains to be seen, as the researchers note that too few years have passed since the implementation of regulations designed to conserve streams and rivers in the County; to better assess the overall impact more time must pass in order to allow for a better and broader gathering of data.

The study by Booth and Jackson (1997) finds that the rise of urbanization in King County has negatively impacted the "form and function of the downstream aquatic system" and introduced obstacles that are very unlikely to be corrected (p. 1). The researchers assert, moreover, that "a better understanding of the critical processes that lead to degradation" is needed in order to adequately address the issue (p. 1). What is meant by "better understanding," however, is "a more precise, process-based" assessment "of how altered landscapes produced degraded stream channels" (p. 18). Such an assessment may in fact be unnecessary, as the results are already evident and beyond repair, according to the researchers. What the study suggests is that it may be too late for some species of aquatic life as the impact of urbanization is more fully felt by observers of the fish and wildlife populace.

Kerwin (2016) shows in his study of the King Country region that a number of variables have been assessed in recent decades, including rainfall, temperature of waterways, and the footprint of upland development. Of these, the "temperature assessment completed in 1996" offers an indication of the extent to which water temperatures have altered: assessments still being conducted have noted "exceedences of the water temperature criterion" on seven rivers and creeks within the County (with variances among these seven ranging from Camp Creek exceeding its temperature criterion for only one day to Clearwater exceeding its criterion for more than half a year and Greenwater exceeding by more than two-thirds of the year). Thus, throughout the region of King County, there is no consistent pattern even within the variables identified as impactful on salmon populations.

DeGasperi et al. (2009) issue the same warning following the conclusion of their study on the relationship between hydrologic alteration and biological impairment in the Puget Lowland. Their findings show that urbanization alters the course, flow and nature of streams and rivers in areas connected the spread of urban development and that without a more accurate assessment of the intricate nature of the causation behind stream degradation, preventing the inevitable decline of aquatic life in these regions is all but impossible. This study along with the others described in this review indicates that the common theme among these assessments is the correlation between urbanization and stream degradation. Understanding the precise impact, however, requires further research. Indeed, going back to Klein's (1979) assessment of the impact of urbanization on stream degradation, one sees that this is a consistent theme that has been observed by researchers for more than 30+ years.

Methodology

Meta-analysis for the first portion of the research question was conducted by performing a series of searches in databases for literature pertaining to the issue of stream degradation and Chinook/steelhead salmon populations from 1970 to 2010. In all, 24 studies were surveyed and the method of selecting themes was based on identifying common characteristics reported.

The first part of the question identified correlation themes. These were typically stated at the outset of the study or issued within the conclusion. Correlation identification required identifying the correlation keyword with the other keyword being urbanization or some variation thereon. Stream degradation was described variously by studies and so some leeway was given to extracting evidence based on variations of the degradation definition (based on benthic analysis, biological life, temperature analysis, flow analysis, contamination, pollution, run-off, alteration, hydrology, water quality, stream channel complexity, land cover, etc.). Each description was accepted as a variable in degradation and counted as a relevant source.

For the second part, data was gathered and analyzed by examining studies that were interval-based in their assessment of aquatic habitat; these totaled 6 and their data was categorized according to the subject under analysis (flows, temperatures, stream width, aquatic life) and the length of duration of the study.

For the third part, bivariate correlation analysis was conducted using SPSS technology to identify correlation trends between human population and stream degradation awareness.

Findings

Every study reviewed as part of the meta-analysis showed correlation between urbanization and stream degradation. More than half of the studies could point to the nature of this correlation being linked to the physical alteration of stream/river flows and depth levels, while a similar proportion identified the causal link between the correlating variables in terms of pollution and disturbance of natural habitats. Kerwin (2016), for example, identifies "water quality, peak flows and bedload movement" as variables "adversely compromised due to upland development" (p. 86). The majority of the studies indicate that the physiological aspect of streams and rivers has changed over the course of time and that this change begins to be observed by academics beginning in the 1970s, with both king and steelhead salmon populations being placed on the threatened list of species, according to King County censuses.

A major common theme to all of them was that deforestation and the "conversion of fringe rural resource lands to other uses (primarily suburban development and transportation)" (DeGasperi, 2009, p. 531). Among the other ommon themes found in the literature were correlation between aquatic integrity, biological health as determined by the existence of benthic macroinvertebrate, and the land cover of urbanized areas along with areas marginally affected by urbanized land cover. Numerous regions with King County were analyzed over more than three decades within the course of these studies and the conclusion of each was virtually the same. Another common theme among them all is the call for further research to better understand the nature of the relationship between urbanization and stream degradation; however, by the 21st century these calls are appearing to be more skeptical of any real or effective solution being implemented to "bring back" the aquatic life once supported by these aquatic systems.

No clear census data for the area's salmon population has been provided by any reliable source; an assessment of salmon populations, particularly king and steelhead, is mainly dependent upon the observation methods of local inhabitants, such as the Muckleshoot Indian Tribe, which has been instrumental in delivering information about the local salmon population over the preceding decades (Kerwin, 2016). However, hard data regarding the actual numbers of fish is difficult to verify. Thus, assessing the change in habitat over time using interval levels of measurement became the target approach to the question. The existent literature has revealed that not all streams and rivers in the county have been impacted proportionally. In describing the change in aquatic life habitat, most studies concentrate on flow, temperature, stream/river width, depth, and soil disturbance. What descriptive statistics are available indicate that the area's king and steelhead populations saw their lowest levels in terms of population in the 1970s, though in certain regions, such as White River, the spring Chinook population has been growing and has seen increases in run sizes, according to observations gathered by the Muckleshoot Indian Tribe (1996). Yet, as Kerwin (2016) notes, "since 1967, run sizes of fall Chinook, coho, pink, chum and winter steelhead have been highly variable" and "there is no reliable historical source of information on salmonid species abundance in the Puyallup River basin of record" (p. 15). Moreover, Healey (1991) shows that young king salmon can live for days to twelve months in the freshwater rivers of the Puget Sound before they finally migrate to the estuaries of the Pacific. Thus, maintaining a clear sense of salmon levels over a consistent span of time and space is made all the more difficult considering there is no definitive way to measure for the Chinook population, in particular, as its cycles are not strictly fixed.

Meta-analysis does indicate that upland development has negatively impacted the aquatic habitats of the waterways in King County but as DeGasperi et al. (2009) conclude, it is uncertain exactly how this impact is effecting the life cycles of the salmon population without a more intensive study of the various regions, which are exceedingly diverse. Therefore, meta-analysis has revealed that stream degradation has been evident since the 1970s and aquatic habitats have changed, but the impact on the life of the king and steelhead salmon can only be deduced by the observational methods of the local population and these are indeterminate.

The most extensive study was performed by Luchetti et al. (2014) and in the limited timeframe of their analysis (2010-2012), variables ranged in alteration: land cover composition did not change. Hydrolic condition indices varied little and insignificantly, with only certain creeks showing any substantial variation. Historically, these variations are more significant over longer periods of time. The main finding here was that reforestation projects and policies implemented by King County to conserve the natural habitats in the region have been effective in at least providing some stability in the region regarding change in habitat.

The human census population data has shown nearly 100% growth since 1970 and over 100% growth since 1960. The 1970 population of King County stood at 1.156 million people. That decade saw the area grow by 100,000 persons. By 1990, the population stood at 1.507 million people, with another 230,000 persons being added that decade and another 200,000 the following decade to push 2010 total population to 1.931 million people. Growth has consistently climbed in a double-digit percentage rate over the past four decades, more than doubling the population size of the county by a million persons. If one goes back to 1900, the census data for King County shows that the county has grown by 20-fold over the past century, with 100,000 to 200,000 people being added to the census every ten years (Census 2000 PHC-T-4: Ranking Tables for Counties: 1990 and 2000; State and County Quickfacts; Population of Counties by Decennial Census: 1900 to 1990; U.S. Decennial Census).

Correlation with stream degradation studies produced the following results: Stream degradation reached its highest recorded point in the 1970s following an explosion in population size from the previous decade and a ten-fold explosion of growth over the previous five decades. At the same time, the public consciousness regarding the natural habitats of aquatic life was only beginning to develop and show signs of awareness, as it is in this decade that the main studies are performed identifying correlation between urbanization and stream degradation and that the local Indian tribe's data regarding salmon population is shown to be at its lowest.

With the implementation of conservation policies in King County throughout the following years, notably in the 1990s with the focus being on restoring habitats, reforesting and mitigating risks of contamination through educational awareness programs, correlation has not been shown to be as clear. The bivariate analysis of human population growth and stream degradation (with stream health measured by B-IBI statistics gathered from the relevant literature) indicated steady correlation where available; however, the B-IBI measurements over the past four decades were found to be to sparse, disconnected and unrepresentative of the streams and rivers of the county as a whole to be genuinely indicative of any practical or real correlation. Too many gaps existed in the data regarding years assessed and space/region identified.

Booth et al. (2003) provided the best example of how such a correlation study could be conducted for future research; their study utilized information gathered from 46 sites in King County over a two-year period (1997 to 1999) in which stream degradation was characterized by benthic invertebrate samplings at each site along with hydrologic analyses at a third of the sites; thus, a tri-variate method was employed of stream survey, landscape survey (human activity measured by land coverage -- road density, impervious area), and observation/qualitative values of locals. However, even here it is seen that quantitative statistics are only useful to a point and that qualitative observational methodology is relied upon to "ground" the study, so to speak, in an overall air of concern. The question this raises, however, is whether or not a research bias is permeating the quality of research conducted on this subject, with all data extrapolated geared towards finding a correlation between urbanization and stream degradation; so far no compelling evidence has been put forward that salmon populations are actually being impacted by either variable, as it ultimately comes down to a quasi-qualitative assessment and the application of theoretical understanding of spawning cycle, stream conditions believed to be needed for life to continue, and the awareness of population growth in the region. These factors will be discussed more fully in the following section, Conclusion and Recommendations.

Overall, even though studies dating back to the 1970s show correlation between urbanization and stream degradation in King County, the findings are inconclusive regarding correlation between urbanization and salmon depletion, which is only observable via qualitative methods of assessment. There remains no concrete or verifiable quantitative evidence of precise salmon population numbers. Even the "State of Salmon in Watersheds 2014 Report" published by Washington state and available on the state's .gov site indicates that its optimistic reading on salmon level increases (as a result of conservation strategies implemented in recent years) in some parts of the state and its pessimistic reading (showed by a decreasing levels of steelhead and chinook in the Puget sound) is "a nonstatistical evaluation of adult abundance trends for wild fish and is based on data provided by the Washington Department of Fish and Wildlife, tribes, and regional recovery organizations" -- in other words, it is all based on perception (Are Wild Salmon Increasing in Numbers?, 2014).