Count the Kinds: Measuring Schoolyard Biodiversity with Quadrats

Investigation question: Does a regularly mowed lawn or a less-managed patch of the schoolyard hold more plant biodiversity, and how many quadrats does it take to find out?

Scientific purpose

Students measure plant biodiversity in two patches of the schoolyard that people manage differently. Using 50 cm quadrats placed by dice roll, they build a shared key of plant kinds, record the first plant touched at 25 grid points in each quadrat, and calculate species richness and Simpson's index for each patch. Pooling class data on a species-accumulation graph shows whether the sample was large enough. The lab turns biodiversity from a word into a measurement and gives students their own evidence about how one change people make, mowing, affects which plants grow and how evenly they are mixed.

Why this helps students with this chapter

Biodiversity is easy to talk about and hard to measure. Doing it themselves shows students that richness and evenness are different, that where you put a quadrat changes the answer, and that a careful scientist asks whether the sample was big enough before comparing two places. These are the same measurements conservation teams use to decide whether a plan is working, so the lab sets up the evaluation work in Lab 3 and the kākāpō task.

Standards connections

MS-LS2-4 (practice, constructing an argument from empirical evidence that a change to an ecosystem component (mowing) affects plant populations): Students collect their own field data and argue whether mowing changes which plants grow and how evenly they are mixed (q8). The data are real but come from two patches that differ in more than mowing and from a small sample, so the lab builds the argument practice rather than assessing the full Grade 7 PE.

MS-LS2-5 (supports, measuring biodiversity as the evidence used to evaluate a design solution): Richness and a diversity index are the measurements a team would use to judge whether a plan to maintain or restore biodiversity is working. q10 asks students to design such a before-and-after survey for a school pollinator garden, preparing for the evaluation in Lab 3.

Roles (rotate each session)

• Site setter: Rolls the dice, measures out each quadrat position from the baseline, lays the frame, and rerolls if the spot lands on a path, a hazard or an earlier quadrat.
• Point reader: Lowers the pointer, blunt end down, through each of the 25 string intersections and calls the code of the first plant touched, or 'bare'.
• Recorder: Tallies each call in the hits table, keeps the presence list for each quadrat, and checks that every quadrat totals 25 points.
• Kind keeper: Sketches and describes each new plant kind with a hand lens, assigns it the next class code, and checks the group's key against the class chart.

Materials

Estimated cost: 5 to 50 per class (an estimate; Estimate. Cardboard frames from recycled boxes plus string, tape and a pack of skewers cost about 5 dollars per class. Reusable PVC frames for eight groups (eight 2 m lengths of half-inch pipe at about 2 dollars each and 32 elbows at roughly 0.50 to 1 dollar each) add about 35 to 45 dollars and last for years. Dice, hand lenses, calculators and clipboards are typical classroom stock.)

Safety

• Work only in the teacher-checked patches. Poison oak grows at the edges of many California school grounds and causes a severe rash; nettles, thistles and some weeds sting or prick. Do not touch plants you have not been told are safe, and never put plants or fingers in your mouth.
• Bees and wasps visit flowers in less-managed patches, and yellowjackets nest in the ground. Move slowly, do not swat, and step away from any nest. Students allergic to stings carry their prescribed epinephrine auto-injector outdoors, and the teacher knows who they are.
• Skewers have a sharp point: lower them blunt end down, never point them at anyone, and carry them point down. Unsharpened pencils are a safer substitute.
• Do not touch animal droppings, broken glass, needles or other litter; tell the teacher. Do not survey any area that has been sprayed recently.
• On sunny days wear a hat and sunscreen, bring water and take breaks from kneeling in the sun. Watch for sprinkler heads and holes when walking.
• Stay inside the marked patches and within sight of the teacher. Do not pull up or pick plants, and wash hands after the field session.

Teacher preparation and setup

1. A few days ahead, walk the grounds with facilities staff and choose two flat patches near each other, each at least 4 m by 4 m: a regularly mowed lawn and a less-managed patch (an unmowed edge, a weedy margin, a native or pollinator garden, a field corner). Confirm that neither has been sprayed with herbicide, pesticide or fertilizer in the past week, and ask that the lawn not be mowed the day before.
2. Check both patches for hazards and flag or avoid them: poison oak, nettles, thistles and other spiny plants, ant mounds, ground nests of yellowjackets or bees, animal droppings, glass and sprinkler heads. Check the sprinkler schedule.
3. Build the frames and tie the string grid; test one in the lawn to make sure the strings sit just above the plants and the pointer passes cleanly through each intersection.
4. Draw a starter kind key on the board or a class chart with three common lawn plants (for example a fine-bladed grass, a clover with three leaflets, a flat rosette weed with toothed leaves) coded K1 to K3. Groups add new kinds as K4, K5 and so on to the same chart so every group uses the same code for the same plant.
5. Lay a baseline of flagging tape along one edge of each patch and mark it every 50 cm from 0.5 m to 3 m, so a die roll from 1 to 6 gives a position along it.
6. Ask about allergies (stings, grass and pollen). Make sure any student with a prescribed epinephrine auto-injector has it outdoors and that an adult with the group knows how to use it.

Investigation design

Student procedure

1. Session 1, in class (5 minutes): write your prediction in q1 and answer q2. Look at the starter kind key and agree how your group will tell kinds apart (leaf shape, edges, hairs, how the leaves attach, flowers).
2. At the first patch, the Site setter rolls one die for the position along the baseline (the roll times 50 cm) and the second die for the distance into the patch (the roll times 50 cm, measured with the meter stick), then lays the frame with its corner on that spot. If the frame lands on a path, a flagged hazard or an earlier quadrat, roll again; never move the frame to a spot that looks more interesting.
3. Kind keeper and Recorder list every plant kind rooted inside the frame (the presence list). Each new kind gets the next class code, a sketch and a description in the kind key. Use the hand lens and do not pick the plant.
4. Point reader lowers the pointer, blunt end down, straight through each of the 25 string intersections and calls the code of the first plant it touches; if it reaches soil, mulch or dead leaves without touching a living plant, call 'bare'. Recorder tallies every call. A partner watches that the pointer stays vertical.
5. Repeat steps 2 to 4 for quadrats 2 and 3 in the same patch, then sample three quadrats in the second patch. Note the time, the weather and whether the lawn looks freshly mowed.
6. Session 2: total each patch's hits by kind and leave 'bare' out of the plant total. For each kind compute p = hits ÷ plant total, square it, add the squares, and compute Simpson's index D = 1 − (sum of p²).
7. Count each patch's richness: all kinds on your presence lists, whether or not the pointer hit them. Fill the accumulation table by counting how many kinds each quadrat added that earlier quadrats of the same patch had not.
8. Copy your group's totals to the class chart. Using the class data in the order groups sampled, plot total kinds found against number of quadrats sampled for each patch on one graph, one color per patch. Compute the class Simpson's index for each patch from the pooled hits.
9. Answer q3 and q4 on the recording sheet as a group, then the analysis questions individually.

Expected observations

• Mowed lawn quadrats usually show about 2 to 6 kinds, with one or two grasses taking most of the 25 points; Simpson's index often falls roughly between 0.2 and 0.5.
• Less-managed patches usually show more kinds (often 4 to 10 per quadrat) in a more even mix, giving indexes around 0.5 to 0.8, though a mulched bed with a few shrubs can score lower than a weedy lawn.
• Accumulation curves rise steeply over the first few quadrats and then flatten; the less-managed patch usually needs more quadrats before it flattens.
• Bare points are more common in garden beds and trampled areas; leaving them out of the plant total keeps the index about the plants only.
• Group values for the same patch can differ by several kinds and by 0.1 to 0.2 in the index, while the class-pooled values are steadier.

Likely difficulties

• Deciding whether two plants are the same kind: young and old leaves can look different. The Kind keeper compares with the class chart, lumps when unsure, and notes the decision.
• Placing quadrats by eye instead of by the dice, drifting toward interesting plants. The Site setter role and q2 exist to prevent this.
• Counting 'bare' in the plant total, or forgetting to square the proportions. Work patch X from q5 on the board before session 2.
• Several grasses look alike. Treat all fine-bladed grasses as one kind unless the difference is obvious, and record this as a limitation.
• Wet grass soaks knees and recording sheets; use kneeling mats and clipboards with a plastic sleeve, or wait for the lawn to dry.

Structured debrief

1. Write each group's richness and index for both patches on the board beside the class-pooled values. Ask why groups disagree and why the pooled value is more trustworthy: more quadrats, more of the rare kinds found, less chance that one odd quadrat swings the result.
2. Show the accumulation curves. Where a curve flattens, extra quadrats add few new kinds; if it is still climbing, richness is underestimated. Connect to the named-species bar graph: Earth's species list is a curve that has not flattened, which is why scientists keep finding new species.
3. Compare the two measures using patches X and Y from q5: same richness, very different index. Ask when richness and the index could rank your own patches differently, and which question each measure answers.
4. Raise the health question from slide 17: is the less-managed patch healthier? Ask which kinds are native and which are lawn weeds from other continents. Point ahead to Lab 3: measurements like these are how a team would tell whether a plan to restore a habitat is working.

Recording sheet: what students fill in

Lab 1 recording sheet: Count the Kinds

You will sample three quadrats in Patch A (mowed lawn) and three in Patch B (less-managed patch). In each quadrat, list every plant kind you see (the presence list), then record the first plant touched at each of the 25 grid points. A point that touches only soil, mulch or dead leaves is bare. Use the class codes (K1, K2, ...) so every group means the same plant by the same code.

Analysis questions: answer key

Individual evidence of learning

Each student submits their own copy of the hits table, the Simpson's calculation for both patches, the class accumulation graph and answers to q1 through q11. The calculations and graph show measurement skill; q8 is the evidence-based argument about a change to an ecosystem; q10 shows that the student can design the kind of monitoring that would evaluate a biodiversity solution.

Assessment criteria

Accessibility adaptations

• Choose patches beside a hard path so students using wheelchairs or walkers can reach the frame; a longer pointer such as a 60 cm dowel lets a seated student make the point readings.
• Students with grass or pollen allergies can work as Recorder at the patch edge, or build the kind key indoors from photos of each quadrat taken by the group.
• Provide a calculation template with the p, p² and sum columns set up, and allow a calculator or spreadsheet.
• Offer a bilingual word bank for the kind key (leaf, edge, toothed, smooth, hairy, rosette, blade, leaflet, flower).
• Provide large-print recording sheets and a magnifier for students with low vision; the hand lens task can be shared.

Low-cost alternative

Make frames from cardboard and string, use pencils as pointers and a random-number list instead of dice. If no outdoor site is safe or available, run an indoor bead-tray model: two trays of mixed dried beans or beads stand for two patches with known counts, and students sample each with a card punched with 25 holes. They compute richness and Simpson's index from their samples and compare them with the true values, which also shows how sample size affects an estimate.

Model limitations

This activity has not been physically tested by the authors; the ranges in the expected observations are estimates, and a teacher's trial run on the chosen site is the real guide. Plant 'kinds' are look-alike groups rather than identified species, so richness can be undercounted (similar grasses lumped) or overcounted (young and old leaves of one plant split). Point-intercept hits measure cover, not numbers of individual plants, and they miss small or rare kinds that the presence list catches. Six quadrats per group is a small sample. The two patches differ in more than mowing (soil, water, shade, foot traffic, past planting), so the comparison suggests but cannot prove that mowing is the cause. Plants are only one part of biodiversity; insects and soil life are not counted.

Teacher notes

Run the method yourself on the chosen patches a few days ahead; the expected ranges here are estimates, not tested results. Work patch X from q5 on the board before students calculate their own index: it fixes the two most common errors (counting bare points and forgetting to square). For the class accumulation curve, enter quadrats in the order groups finished sampling; a curve built in a different order will look slightly different but should flatten at about the same height. Grade q3, q4 and q6 against the student's own tables rather than against the typical values. If the less-managed patch is a garden bed full of mulch, many bare points are expected and its index may be lower than the lawn's; that is a legitimate result and a good debrief discussion.