
use a spreadsheet model to test our plan under different weather patterns, analyze how trait proportions change, and identify the model’s limitations.
Phase 4: Build
Do Now. Our bead model assumed every year is dry. Real weather is not like that. Predict what happens to Plan A if every other year is wet and all plants survive.
In a wet year, almost all plants survive. What happens to the share of tolerant plants?
Breeders, a spreadsheet can run a hundred trials while beads run one. Each row is a generation. Column B holds T, the tolerant share. Two formulas carry the whole model:
In a wet year (sample data), survival is about 95% for both kinds of plant, so T does not change. For the hybrid, the generation 2 row uses the Punnett result, T = 0.75, instead of the formula.
Worked example 1: every year dry. The hybrid’s survival column reads 80%, 68%, 74%, 78%, 79%. It drops below 70% only in generation 2 and then climbs above it to stay, so by generation 5 a dry year leaves at least 70% alive. Plan A reads 45%, 57%, 68%, 74%, 78%: it also passes, but only in generation 4.
Worked example 2: dry, wet, dry, wet, dry. Noah changes one input, the weather pattern. Plan A now reaches only 68% in generation 5, so it fails. The hybrid reaches 74% in generation 5, a narrow pass. Noah writes: “Selection only happens in dry years. More wet years means slower change, so the plan depends on weather we cannot control.”
| Gen | Weather | T (tolerant share) | Survival | Meets 70%? |
|---|---|---|---|---|
| 1 | ||||
| 2 | ||||
| 3 | ||||
| 4 | ||||
| 5 |
Every model has limits. Ours assumes one gene, but real drought tolerance often depends on many genes. It assumes tolerant parents give tolerant seeds, but Dd parents can make dd seeds, so real progress may be slower. It ignores pests, soil, flavor and how dry a dry year is. Because so many causes act at once, no model can predict exactly what a real field will do; it can only show likely trends.
Build the spreadsheet with your team, then try each scenario.
Use the model to plan improvements.
Run your hybrid with three weather patterns (all dry; dry, wet, dry, wet, dry; and one your team chooses). Record generation 5 survival for each and state which patterns meet the criterion.
Name two limitations of the model and explain how each could make the real farm's results better or worse than the model predicts.
Using your data and the criteria, suggest one change to the plan that could make it pass even with wet years.
The model has shown us a weak spot: wet years slow selection. Next class, peer critique and version 2.
Exit ticket: explain, with numbers, how a change in the environment (more wet years) changed how fast the tolerant trait spread.
How confident are you that you can I can identify the limitations of my model and use its data to improve my design.
Materials: laptops or tablets with a spreadsheet app, starter spreadsheet with columns Gen, Weather, T, Survival, Next T; calculators for anyone without a device.
Setup: sample formulas: survival in C2 =IF(B2=”dry”,A2*0.8+(1-A2)*0.3,0.95) with T in column A and weather in column B; next T =IF(B2=”dry”,A2*0.8/C2,A2). Override the hybrid’s generation 2 T with 0.75. Values in this lesson are rounded sample data.
Listen for: “the model is wrong, so it is useless” (all models have limits; they still show trends); “plants become tolerant because they need to.”
Support / stretch: pre-build the formulas so students change only inputs. Stretch: add a random-number column, =RAND()<0.8, to simulate individual plants and compare with the formula.