
run and revise a computer simulation to test how reducing forest fuel density changes fire spread, and build and test a physical prototype for controlling erosion after a fire.
Instruction
Fire belongs here. Most of Missouri’s natural communities evolved with fire. The Missouri Fire Needs Assessment, prepared by The Nature Conservancy and partners, estimates that its fire-adapted ecosystems historically burned every 2 to 20 years, mostly from fires set by people, and that until about a century ago fire was a common way of managing the land. Low fires cleared leaf litter and young trees, kept oak and shortleaf pine woodlands open and sunny, and kept glades and prairies free of cedar. After Missouri began suppressing wildfires in the early 1900s, forest canopies and midstories closed, shading out grasses and wildflowers, and fuel built up. The assessment finds that Missouri should burn at least 2,635,008 acres a year to match its historical fire pattern, but burns only about 125,000 to 150,000, roughly 5 percent of that. The problem is not fire itself but a change in the fire regime.
Not every ecosystem needs the same fire. The assessment gives each ecosystem its own historical interval: one southern Missouri glade burned about every 3 years, while other communities burned closer to every 20. A woodland next to a town has different constraints from a remote national forest. A good solution depends on the place, which is why engineers break a large problem into smaller, specific ones (9-12.ETS1.A.2).
Testing an idea with a simulation (9-12.ETS1.B.2, 9-12.LS4.C.3). The NetLogo Web Fire model fills a grid with trees at a chosen density and sets fire to the left edge. Fire spreads from a burning tree to any neighboring tree. The result is striking: below a threshold density, around 59% in this model, fires usually die out after burning a small fraction of the forest; above it, fires usually sweep across. A small change in density near the threshold makes a large change in the outcome. That suggests a solution to test: reduce density by thinning and prescribed burning. The model is too simple to prove it, and a strong engineer says so. It has no wind, slope, moisture or tree size. Revising the simulation means adding or changing a rule to make it more realistic, then asking whether the conclusion still holds. For example, you could represent thinning or a prescribed burn by lowering density in a strip, a fuel break, and test whether that stops the spread.
Testing an idea with a prototype. After a severe fire, burned slopes lose the plants and litter that held soil. The first heavy rain can wash ash and soil into streams, such as clear Ozark streams, smothering fish eggs and filling reservoirs. Land managers spread straw mulch or lay straw rolls (wattles) across slopes to slow runoff. In today’s lab you will build slopes in trays, apply a treatment, pour a measured “storm” and collect the runoff. In an illustrative test, a bare tray lost 420 g of soil; a tray with straw mulch lost 95 g, a reduction of (420 − 95) ÷ 420 ≈ 77%; a tray with two wattles lost 160 g, a 62% reduction. Mulch performed best in this test, but a full evaluation also weighs cost, how easily each can be applied on steep ground, and whether straw brings in weed seeds (9-12.ETS1.B.1).
📖 Read for this lesson
The reading opens over the lesson; close it to come back to exactly where you were. The checks below draw on it.
Lab: Post-fire erosion-control prototypes
Question. Which low-cost treatment most reduces soil loss from a bare slope during a simulated storm?
Materials
- Three identical plastic trays or cut-open milk cartons with a hole or spout at one end, propped at the same angle (about 20°)
- Dry garden soil or sand-soil mix, straw or dry grass clippings, and string or rubber bands to make small wattles
- A watering can or cup with holes, a 1 L measuring jug, collection containers, coffee filters and a kitchen scale
- Work outdoors or over a tub; wipe spills at once to prevent slips.
- Wear goggles when pouring water onto soil to avoid splashes in the eyes, and wash hands after handling soil.
- Keep water and wet soil away from electrical outlets and scales’ power cords (use battery scales if possible).
Procedure
- Pack each tray with the same mass of soil and prop all three at the same angle. Leave one bare (control).
- Cover the second tray with an even layer of straw of a measured mass; lay two small straw wattles across the third, pressed into the soil.
- Pour 1 L of water from the same height over each tray in the same time (about 30 seconds) and collect all the runoff.
- Filter each runoff sample through a weighed coffee filter, let it dry overnight, and weigh the soil lost.
- Repeat the storm once more on each tray and record totals.
- Calculate the percentage reduction in soil loss for each treatment compared with the control.
Data
| Tray | Treatment | Soil lost, storm 1 (g) | Soil lost, storm 2 (g) | Total (g) | % reduction vs control |
|---|---|---|---|---|---|
| A | Bare (control) | — | |||
| B | Straw mulch | ||||
| C | Wattles |
Analysis
- Which treatment reduced soil loss most? Calculate each percentage reduction and cite your data.
- Evaluate the two treatments against three criteria and constraints: effectiveness, cost of materials per square meter, and ease of applying on a steep, burned slope.
- Identify two ways your tray model is unlike a real burned hillside, and explain how each might change the result.
- Propose one change to your prototype and predict its effect.
If you cannot do this lab in person: Choose one. All three earn full credit and end in the same percentage reductions and the same evaluation against criteria and constraints. Option A — build the trays. The procedure above, outdoors or over a tub. Milk cartons cut open lengthwise, garden soil and dry grass clippings are enough; a kitchen scale and coffee filters do the rest. Option B — the illustrative test in this lesson. No materials and no internet. Two storms were run on each tray, with these soil losses: bare control 245 g then 175 g; straw mulch 58 g then 37 g; wattles 95 g then 65 g. Total each tray, compute each percentage reduction against the control, and check that you get the 77% and 62% quoted in the lesson. Then complete the same evaluation and the same two questions about how the tray model differs from a burned hillside. Option C — the hosted field test. This lesson’s reading hosts a real field measurement: Forest Service catchments after the 2002 Indian Fire near Prescott, Arizona, comparing a bare control against wood-chip, straw-pellet and rice-straw mulches over three real storms, with sediment yields in megagrams per hectare. Use the reading’s worked reductions (93 percent for chips, 80 percent for pellets, and 81 percent for rice straw in the biggest storm) in place of the difference between your own trays’ bare control and treatments, then evaluate straw mulch and wattles against the same three criteria, using the reading’s ground-cover measurements to explain why the pellets’ protection fell from 80 to 42 percent as cover thinned. Cite the reading and its retrieval date.
Formative check
Work through these before moving on. They are not graded — they tell you whether the standard below has landed yet.
Simulation & tools
Find the fire threshold and test a fuel break
Open the Fire model in NetLogo Web. (1) Run it three times each at densities of 50%, 57%, 59%, 61% and 70%, recording the percent burned each time. Graph mean percent burned against density. (2) Describe the threshold you find. (3) Revise the model: in the Code tab, find the setup procedure and add this line at the end of the setup procedure, on its own line just before end, so that the patches near the middle of the world hold no trees (a fuel break): ask patches with [pxcor > -5 and pxcor < 5] [ set pcolor black ]. Press setup again and you should see a bare stripe down the middle. Rerun at 70% density and record the result. Explain in three sentences what your revised model suggests about thinning and fuel breaks, and one thing it leaves out. Choose one. All three earn full credit and answer the same three questions. Option A is the model above. Option B, with paper and a die instead of internet, is the grid game: draw a 20 × 20 grid, roll for every square and shade it as a tree if the roll falls inside your chosen density, then spread fire by hand from the left edge to every shaded square that touches a burning one, and count the percentage burned. Run it three times at each density, and for the revision step leave columns 10 and 11 unshaded as your fuel break. Option C, with nothing at all, is the recorded model output below: constructed teaching numbers showing what the model produces, not measurements. Graph the means against density, describe the threshold and compare the last row with the 70% row.
| Tree density | Percent burned, three runs | Mean |
|---|---|---|
| 50% | 1.2, 0.9, 1.5 | 1.2 |
| 57% | 3.1, 2.4, 9.8 | 5.1 |
| 59% | 21.4, 6.2, 44.1 | 23.9 |
| 61% | 61.0, 74.3, 55.8 | 63.7 |
| 70% | 98.1, 98.6, 97.9 | 98.2 |
| 70% with the fuel break added | 42.8, 39.1, 45.6 | 42.5 |
Open NetLogo Web ↗ · NetLogo Web, Northwestern University CCL · GPL open source
A bare tray lost 420 g of soil and a straw-mulched tray lost 95 g. The mulch reduced soil loss by about %.
In the Fire model, fires at 57% density usually stop early but at 61% usually cross the forest. What does this suggest about thinning a dense forest?
Because the Fire model shows a threshold, it proves that thinning will protect any real Missouri woodland.
How confident are you that you can use and revise a simulation and a prototype test to evaluate a solution for protecting biodiversity?
Practice
Work these on paper or in your notebook, then open Check your answer. Aim for all of Fluency and Application; try at least one Challenge.
Printable version: this unit’s practice workbook (PDF)
Fluency
Build speed and accuracy with the core skill.
- A bare tray lost 420 g of soil and a straw-mulched tray lost 95 g. Find the percentage reduction.
Check your answer
Answer: About 77%(420 − 95) ÷ 420 = 325/420 = 0.774. - A tray with wattles lost 160 g compared with 420 g for bare soil. Find the percentage reduction.
Check your answer
Answer: About 62%260 ÷ 420 = 0.619. - In the NetLogo Fire model, what happens to the percentage of forest burned as density rises from 50% to 70%?
Check your answer
Answer: It stays low at first, then rises sharply around a threshold near 59%, so that at 70% most of the forest burns.Small changes near a threshold produce large changes in outcome.
Application
Use the skill in context. Show your reasoning.
- Straw for one tray costs $2.00 and wattles cost $3.00 (illustrative). Using the soil saved compared with the bare tray (420 g), find the grams of soil saved per dollar for each.
Check your answer
Answer: Straw: about 163 g per dollar; wattles: about 87 g per dollarStraw: 325 ÷ 2 = 162.5. Wattles: 260 ÷ 3 = 86.7. - Name two things the NetLogo Fire model leaves out that matter for real Missouri woodlands, and explain how one could change the model’s conclusion about thinning.
Check your answer
Answer: Wind, slope, fuel moisture, tree size and species, and fire intensity. For example, strong wind can carry embers across gaps, so a thinned forest or fuel break could still burn.A model’s conclusions hold only within its assumptions. - Explain why burning an Ozark glade every few years helps it, but the same schedule may not suit every Missouri ecosystem.
Check your answer
Answer: Glades evolved with fire about every 3 years, which kills young cedars before they can shade out the glade plants. The Missouri Fire Needs Assessment gives each ecosystem its own historical interval, from about 2 to 20 years, so burning a longer-interval community every few years would push it away from the conditions its species are adapted to.The right fire regime depends on the ecosystem.
Challenge
Stretch problems. Expect to think before you write.
- Break the problem “restore an overgrown Ozark pine-oak woodland and protect it from severe wildfire” into three sub-problems, and say which could be tested with the Fire model and which needs a different test.
Check your answer
Answer: Sample: (1) reduce fuel density across the woodland with thinning and prescribed fire (Fire model); (2) prevent fire climbing into crowns through ladder fuels such as low cedar branches (not in the model; needs field data or a more detailed model); (3) limit erosion after any fire (physical prototype test).9-12.ETS1.A.2: break a complex problem into parts that can each be solved and tested.
Review
Keep earlier skills sharp.
- Producers in an ecosystem capture 50,000 kJ of energy. Using the 10% rule, about how much energy reaches the primary consumers?
Check your answer
Answer: ≈ 5,000 kJAbout 10% of the energy at one level is passed to the next; the rest is used or lost as heat.
Fire in Missouri: The Burning Gap, a Measured Mulch Test and the Tray Record
The Nature Conservancy and partners, Missouri Fire Needs Assessment (funded by the USDA Forest Service); Anderson, Juniperus virginiana, USDA Forest Service Fire Effects Information System (2003); Riechers et al., Effects of Three Mulch Treatments on Initial Postfire Erosion in North-Central Arizona, USDA Forest Service Gen. Tech. Rep. PSW-GTR-189 (2008); SLS illustrative tray record; retrieved Nov. 5, 2026 · Public domain (USDA Forest Service works); Missouri Fire Needs Assessment figures cited; SLS illustrative tray data and worked calculations
Why this reading. The lesson’s prototype test asks for percentage reductions and an evaluation against cost, effectiveness and ease of application on a steep burned slope. Three things were missing: the numbers behind Missouri’s changed fire regime, the complete tray record behind the 77 and 62 percent, and a real measurement from a real burned hillside to judge your trays against. All three are here.
1. What changed the fire regime: Missouri’s burning gap
The Missouri Fire Needs Assessment, prepared by The Nature Conservancy and partners with Forest Service funding, reports that most of Missouri’s fire-adapted ecosystems would historically have burned every 2 to 20 years, mostly from fires set by people, and that until about 100 years ago fire was used widely to manage the land. Since wildfire suppression began in the early 1900s, “forest canopies and midstories have closed, preventing light from reaching the forest floor,” and woody plants have taken over former prairies. The shortleaf pine woodlands of southeastern Missouri, once open with a diverse understory, are now crowded with species that shade out native grasses and wildflowers.
| Measure | Acres per year |
|---|---|
| Minimum prescribed fire needed statewide to replicate historical fire (2023 assessment) | 2,635,008 |
| Prairie systems alone burned per year before European settlement | over 2,000,000 |
| Estimated prescribed burning today | 125,000 to 150,000 |
Today’s burning is 125,000 ÷ 2,635,008 ≈ 4.7 percent to 150,000 ÷ 2,635,008 ≈ 5.7 percent of the minimum. One glade record shows the same change locally: fires every 3.2 years from 1630 to 1870, then every 22 years (Forest Service Fire Effects Information System, citing Guyette and McGinnes 1982). The assessment adds that prescribed burns decrease fuel loads, “minimizing the risk of wildfires” in forests and woodlands.
2. A measured field test: mulch on a burned slope
Forest Service researchers tested three mulches after the 2002 Indian Fire near Prescott, Arizona, on four small adjacent burned catchments of 0.34 to 0.48 hectares with slopes from under 10 percent at the bottom to 30–40 percent at the top. One catchment was left bare as a control; one was covered with wood chips from the fire-killed trees; one with compressed straw pellets; one with rice straw spread by hand at 4.5 Mg per hectare (2 tons per acre). A silt fence at the mouth of each caught the sediment, which was collected and weighed after each storm. Yields in megagrams (metric tons) per hectare:
| Treatment | Storm 1 (Jul. 30) | Storm 2 (Aug. 14) | Storm 3 (Sept. 14) |
|---|---|---|---|
| Bare control | 6.4 | 10.8 | 48.4 |
| Wood chips | 0.42 | 0 | more than 15.5 (fence partly failed) |
| Straw pellets | not yet installed | 2.2 | 28.2 |
| Rice straw | not yet installed | not yet installed | 9.1 |
| Storm rainfall (mm) | 9.2 | 9.1 | 61.0 |
| Peak 10-minute intensity (mm/h) | 22.9 | 38.1 | 117.3 |
Ground cover, measured in August and again in late October: the bare control was 58 percent bare ground in August and 38 percent in October as some plants returned; chips covered 81 percent then 58; pellets 62 then 27; straw 75 then 51.
Worked reductions, storm 3 (the only storm all four catchments saw): rice straw (48.4 − 9.1) ÷ 48.4 = 81 percent; pellets (48.4 − 28.2) ÷ 48.4 = 42 percent. In storm 2, pellets cut the yield by (10.8 − 2.2) ÷ 10.8 = 80 percent and chips by 100 percent; in storm 1, chips by (6.4 − 0.42) ÷ 6.4 = 93 percent. The authors’ reading: the big September storm, a ten-year event, moved straw into clumps and exposed bare ground between them, and the pellets had lost cover faster than the straw, which is why their protection fell from 80 to 42 percent. Cover on the ground when the storm hits is what does the work.
3. The tray record (illustrative)
Constructed teaching data, not measurements. Three trays packed with 2.0 kg of dry garden soil, propped at 20 degrees, 1 L poured from 30 cm in 30 seconds, runoff filtered and the filter dried overnight. Two storms each.
| Tray | Treatment | Storm 1 (g) | Storm 2 (g) | Total (g) | Reduction vs control |
|---|---|---|---|---|---|
| A | Bare control | 245 | 175 | 420 | — |
| B | Straw mulch (40 g, even layer) | 58 | 37 | 95 | (420 − 95) ÷ 420 = 77% |
| C | Two straw wattles | 95 | 65 | 160 | (420 − 160) ÷ 420 = 62% |
Every tray lost less in the second storm than the first: the loose surface soil went first, and what remained was packed and wetter. That is one way the trays behave like a real slope, and the Forest Service data show the opposite can happen when a bigger storm arrives.
4. Evaluating against criteria and constraints
| Criterion or constraint | Straw mulch | Wattles |
|---|---|---|
| Effectiveness (evidence above) | 77% in the trays; 81% in the field test at 4.5 Mg/ha, falling where cover clumps | 62% in the trays; barriers slow runoff and trap sediment but leave bare ground between them exposed to raindrops |
| Cost of material per square meter | Straw is cheap; the field rate of 4.5 Mg/ha is 0.45 kg/m2, but hand-spreading over hectares is labor | Fewer materials per m2 covered, but each wattle must be staked and trenched |
| Ease of application on a steep, burned slope | Can be spread by hand or dropped from the air; blows and clumps on steep ground | Heavy; must be placed on contour and pressed into soil by crews on foot |
| Side effects | Straw can carry weed seeds onto a slope that has just lost its plants; the field study used rice straw, and the lesson asks you to weigh the seed risk | Minimal seed risk; can fail if runoff undercuts the roll |
5. Two ways a tray is not a hillside
- Scale and storm: a tray gets 1 L in 30 seconds; the September storm dropped 61 mm at 117 mm/h on hectares of slope, and moved straw that had held through two smaller storms.
- Soil and ash: burned soils can form a water-repellent layer and carry ash that seals the surface; garden soil in a tray does neither, so the tray likely understates early runoff and overstates infiltration. Thin, rocky Ozark soils over bedrock would differ again.
Your turn
- Compute every percentage reduction above and confirm the 77, 62, 81 and 42.
- Rank straw and wattles against the three criteria in the table, then say which criterion the Forest Service data changed your mind about and why.
- Propose one change to the tray prototype that tests the clumping problem the field study found, and predict its effect.
Sources: Missouri Fire Needs Assessment (The Nature Conservancy and partners), home page (thenatureconservancy.github.io); Missouri Fire Needs Assessment, Historical Fire Regimes (thenatureconservancy.github.io); Missouri Fire Needs Assessment, Current Needs (thenatureconservancy.github.io); Anderson, M. D. 2003. Juniperus virginiana. In: Fire Effects Information System, USDA Forest Service (fs.usda.gov); Riechers, G. H., J. L. Beyers, P. R. Robichaud, K. Jennings, E. Kreutz and J. Moll, Effects of Three Mulch Treatments on Initial Postfire Erosion in North-Central Arizona, in USDA Forest Service Gen. Tech. Rep. PSW-GTR-189 (2008), pp. 107-113 (fs.usda.gov)