Minimizing Unintended Consequences of Water Resources Decisions

While engineering solutions are meant to help, they often inadvertently harm, as the following examples from water resources engineering illustrate.

In the 19th century, many US cities decided to combine storm and sanitary sewage. Flush toilets were being adopted, and rather than build separate collection systems for the new waste stream, it was easier to route it into the existing storm sewers. Thus were born combined sewers, which led to severe outbreaks of cholera and typhoid in downstream cities (Tarr et al. 1984).

In the 20th century, large dams were constructed to control floods, store water to outlast multiyear droughts, generate emissions-free hydropower, and create recreation areas. These are all positive contributions to community well-being. The same dams, unfortunately, have also disrupted aquatic habitats, stalled the natural flow of sediment, and incised downstream channels (Kondolf et al. 2014). Passage of the US Clean Water Act in 1972, despite the measurable improvements in water quality nationwide, inadvertently made it very difficult to pass stored sediment downstream to maintain reservoir storage and restore the downstream river to a more natural state (George et al. 2017).

Low-head dams have been built across rivers to divert water for irrigation or power. Shallow but extending from bank to bank—and often unmapped and unregulated—low-head dams have become fatal hazards for recreationalists because of the submerged hydraulic jumps that occur downstream. About 40 deaths occur each year at low-head dams in the United States, according to the Association of State Dam Safety Officials.

In western states in particular, water conservation campaigns have reduced household water consumption as intended (DeOreo et al. 2016). But the unintended consequences have included higher wastewater salinity (Schwabe et al. 2020), higher peak demands, and more impervious area in place of landscapes. In agriculture, more efficient irrigation with sprinklers and drip systems has reduced return flows, which otherwise support many downstream needs like baseflow, riparian vegetation, and water rights for other users (R. B. Sowby, N. C. Hansen, E. Dicataldo, “The Dual Legacy of US Irrigation from 1985 to 2015: Gains in Efficiency and Losses in Return Flow,” submitted, J. Water Resour. Plann. Manage., ASCE, Reston, Virginia).

Unintended consequences continue even in the 21st century. Seeking to resolve a financial crisis, a US upper Midwest city decided to switch to a lower-cost water source. The change coincided with elevated levels of lead that were detected at customer taps and prompted a multiyear public health crisis (Masten et al. 2016; Pieper et al. 2017). The city’s name has now become synonymous with drinking water catastrophes: “We don’t want another Flint.”

Do unintended consequences diminish the progress we have made toward solving our water problems? Have our past efforts been mistaken, or should we adjust our perspective?

Even while looking forward for planning and backward for data, decisions about water are necessarily made in the present. Water problems are complex, involving communities, infrastructure, the environment, economics, schedules, and politics. As a society, our needs and conditions change over time; the factors that motivated a past project may not be clear, or even relevant, today.

If our conditions change, so should our knowledge. We cannot expect our own plans, let alone those of our forebears, to work out perfectly. It is the very nature of humans to learn by error, by experience, by incremental improvement. Scottish poet Robert Burns wrote, “The best laid schemes of mice and men often go awry,” and Isaac Newton popularized the saying, “If I have seen farther it is by standing on the shoulders of giants.” Such is the humble perspective we might adopt when tempted to spurn the water resources decisions of the past. To use a Bible phrase fitting for this journal’s audience, we drink from wells we did not dig.

In mapping out a new approach for 21st-century environmental engineering, the National Academies (2019) concluded that engineers need to create “systems and infrastructure that allow people and ecosystems to thrive” together. This does not mean that engineering should abandon the past, just that engineering should learn from unintended consequences and “build on its strengths while positioning itself to keep pace with the scope and scale of society’s needs.”

We have no influence over the past, but we can change how we approach water resources decisions in the future. It is clear that the so-called traditional engineering approaches of yesterday—those that emphasize low costs, maximize single objectives (Farmer et al. 2015), and require only linear thinking—will no longer do. Instead, we must learn new methods that consider, and even leverage, the high complexity of today’s water resources problems.

While we cannot anticipate every unintended consequence, we can expand the search for potential ones and deliberately assess risks. Suckling et al. (2021) proposed classifying unintended consequences along two dimensions—whether they are knowable and whether they are avoidable. This simple but powerful exercise can lead to new insights and solutions. The same study recommended five general strategies to minimize unintended consequences and bring them into the knowable-and-avoidable space; briefly, they are preliminary assessment, iteration, boundary definition, consensus, and active learning.

Another strategy is to insist on interdisciplinary teams. We should not be surprised when projects completed exclusively by engineers focus exclusively on engineering objectives. Already there has been a trend toward more professionally diverse design teams that include city planners, ecologists, social scientists, economists, and public health officials, to name just a few. Water resources research, too, has become more interdisciplinary (Grigg 2021; Sowby and Grigg 2022). However, the success of such teams “requires more than placing experts from different disciplines into a room and adding interdisciplinary verbiage to a proposal … Interdisciplinary collaborations require meaningful interactions and genuine integration across disciplines” (National Academies 2019). This is one strength of the Envision sustainable infrastructure framework (ISI 2021), where collaboration is built into a project from the beginning. In no way does an interdisciplinary team diminish the role of engineers. On the contrary, it validates their contributions with those of other experts.

Open-ended discussions with interdisciplinary teams about potential project impacts can bring to light what otherwise might be passed over. Teams might discuss the environmental consequences typically addressed in an environmental impact statement (EIS), even if one is not required for a project. They might also consider the eight broader outcomes suggested by the Accreditation Board for Engineering and Technology (ABET) for inclusion in civil engineering undergraduate programs. The five categories of the Envision framework (or even all 64 of its credits) are another way to guide such conversations. The three lists are found in Table 1.

One strength of this journal has been numerical modeling, which is well established and well suited to the complexity of modern water problems. But as Sowby and Walski (2021) and Pianosi et al. (2020) summarized, certain modeling techniques, like system optimization, are underutilized among practitioners. The potential for numerical models to find unintended consequences is one more reason for engineers outside of academia to learn to use them. Project teams may gain further insight by performing sensitivity analyses of key variables in such models. Responses in water resources systems are usually nonlinear, and perturbing variables to their extremes could reveal unanticipated consequences. Modeling in this way can help project teams address a large portfolio of concerns.

Even while we humbly and carefully plan to minimize new problems, we must still deal with the old ones. Our existing water resources systems and policies may not change quickly or comprehensively enough to stop, let alone reverse, the negative unintended consequences they have created.

In the meantime, we should do the best we can with what we have. Stated more formally, we should seek optimal use of existing infrastructure and resources within the framework of existing regulations. Additional research on this topic, particularly positive examples in the form of case studies from practitioners (Sowby and Walski 2021), will help the industry progress.

Two recent examples stand out. Where one unintended consequence of dams is the accumulation of sediment, and US policies seem to discourage sediment discharge, Shelley et al. (2022) described several cases where sediment was inexpensively, safely, and legally discharged downstream with great success. Such cases demonstrate how we can work with existing water resources systems and still solve some of their unintended problems.

Another example helps reverse the historic design practice of sizing culvert barrels just large enough for required upstream water depths. This practice, while economical, produced thousands of culverts with water velocities and downstream scour holes that prevent the upstream passage of fish and other aquatic organisms (Hotchkiss and Frei 2007). This outdated approach was changed drastically in the state of Washington in 2018 when the Supreme Court upheld a decision that requires the Washington DOT to replace more than 1,000 restrictive culverts (Graves 2021). Besides opening up many miles of stream habitat for endangered species of salmon, future maintenance costs will be reduced due to slower velocities of water exiting the replaced culvert barrels.

Water resources engineering has come a long way, and society enjoys many benefits this field has enabled. Society also struggles with many negative consequences the field has unintentionally produced. Minimizing further unintended consequences (Table 2) will require us to learn new skills and approach projects in new ways. A humble perspective about the past—a perspective that enables us to learn from our collective mistakes—represents one step we may take. We can also expand our search for potential unintended consequences in the future by using interdisciplinary teams and methods of analysis that capture the complexity of water resources decisions. In the present, we may seek creative opportunities to work within existing infrastructure, resources, and regulations to mitigate some of the unintended consequences and provide positive examples for others to follow.

No data, models, or code were generated or used during the study.