
use peer critique and a stress test to find where my solution and simulation fall short, then refine both by prioritizing criteria and making trade-offs.
Phase 4: Build
Do Now. Team Malik’s 50% plan stops removing urchins after year 5, when there are 0.5 per m². With 10% growth a year and no removal, about how many years until urchins pass 2 per m² again?
Solve 0.5 × 1.1n > 2. About how many years?
Gallery critique. Teams post version 1: the problem statement, matrix, simulation output and trade-off notes. Visitors use three sticky-note colors, I like…, I wonder…, What if…, and every note must point to a criterion, a number or an assumption.
Stress test. Each team tests its solution against its criteria under a harder scenario (sample scenarios): kelp coast, a second marine heat wave in year 6 that cuts kelp growth to 5 points a year; estuary, a drought that raises salinity 15 ppt; forest, a dry, windy year. Then the team asks: does my simulation give enough information to judge the solution? What negative consequences might outweigh the benefits?
Worked example 1: a hidden weakness. Team Malik’s sticky notes said: “I wonder what happens after divers stop.” Running the model on showed urchins recovering past 2 per m² in about 15 years, so the kelp would be lost again. A second problem: removing 50% of urchins a year needs many diver hours in cold water, a safety and cost concern.
Worked example 2: version 2. The team prioritized two criteria, kelp cover above 30% for 20 years and the lowest long-term cost, and made a trade-off: 50% removal for 4 years (20 units), then a maintenance removal of 10% a year (1 unit). With 10% removal and 10% growth, urchins change by a factor of 0.9 × 1.1 = 0.99 a year, so they stay below the threshold. They also revised the simulation, adding a rule for sea stars recovering, and stated a new limitation: the model has no larvae arriving from other reefs. That is refining a solution and revising a simulation based on evidence.
Optimizing means trade-offs. No solution maximizes every criterion. Engineers state which criteria they prioritized, what they gave up, and why the result still meets human needs while reducing harm to biodiversity.
Test, critique and refine.
Version 2 and change log.
Write your change log: two pieces of feedback you used, what you changed in the solution, what you changed in the simulation, and the new test result.
Explain your version 2 trade-offs: which criteria you prioritized, what you gave up, how the solution affects overall environmental stability, and how it still meets human needs.
Two dams built on the Elwha River in Washington State in the early 1900s blocked salmon from most of the river and trapped sediment behind them. After decades of advocacy, including by the Lower Elwha Klallam Tribe, both dams were removed between 2011 and 2014, one of the largest dam removals in history. Before removal, engineers and scientists modeled how the trapped sediment would move downstream and planned how to manage it. Afterward, salmon began returning to upstream reaches, and released sediment rebuilt beaches and the river mouth. The project also showed trade-offs: muddy water for a period, changes for people who used the reservoirs, and long-term monitoring costs.
Think about it: Which parts of the Elwha project match the steps of our design cycle (define, model, test, refine), and what trade-offs did it accept?
Version 2 is tested and justified. Next: the Prove It design brief and pitch to the council.
Exit ticket: what is the most important change from version 1 to version 2, and which evidence caused it?
How confident are you that you can refine a solution and revise a simulation using test results and trade-offs?
Materials: posted version 1 work, sticky notes in three colors, stress-test scenario cards, devices with team simulations.
Case study check: the Elwha and Glines Canyon dams were removed between 2011 and 2014; salmon recolonization and sediment rebuilding of the river mouth are widely reported by the National Park Service and the U.S. Geological Survey.
Listen for: stress tests skipped because the first result looked good; trade-offs stated without what was given up; feedback with no evidence target.
Support / stretch: give a change-log template. Stretch: find the smallest maintenance removal rate that keeps urchins at or below 2 per m² forever (any rate where (1 − f) × 1.1 ≤ 1, so f ≥ about 9.1%).