Resources

In a field ecology lab, an undergraduate class conducts a standardized, closed-population Schnabel multi-sample mark-recapture census to estimate the population size of grasshoppers (Melanoplus differentialis) in a fenced 1,500 m21{,}500\text{ m}^2 prairie parcel over four consecutive days. On Day 1, 4040 grasshoppers are captured, marked with acrylic enamel on the pronotum, and released (M1=0M_1 = 0, C1=40C_1 = 40, R1=0R_1 = 0). On Day 2, 5050 are captured, of which 55 are marked (C2=50C_2 = 50, R2=5R_2 = 5, M2=40M_2 = 40). The 4545 unmarked individuals are marked and released, bringing total marked individuals at large to M3=85M_3 = 85. On Day 3, 6060 are captured, of which 1212 are marked (C3=60C_3 = 60, R3=12R_3 = 12, M3=85M_3 = 85). The 4848 unmarked individuals are marked and released, bringing total marked individuals at large to M4=133M_4 = 133. On Day 4, 5050 are captured, of which 1818 are marked (C4=50C_4 = 50, R4=18R_4 = 18, M4=133M_4 = 133). (a) Calculate the Schnabel population estimate N^=∑t=1k(Ct⋅Mt)∑t=1kRt\hat{N} = \frac{\sum_{t=1}^k (C_t \cdot M_t)}{\sum_{t=1}^k R_t}. (b) Calculate the population density in grasshoppers/m2\text{grasshoppers/m}^2. (c) If 15%15\% of marked marks wore off between Day 2 and Day 4 due to molting, state whether the calculated N^\hat{N} represents an overestimate or underestimate of the true population, and justify why.

Answer and steps

Final answer

(a) N^=393 grasshoppers\hat{N} = 393\text{ grasshoppers}; (b) 0.262 grasshoppers/m20.262\text{ grasshoppers/m}^2; (c) Overestimate; loss of marks artificially lowers the recapture tally (RtR_t) in the denominator, inflating the estimated population size N^\hat{N}.

Steps

Part (a): Calculating the Schnabel Population Estimate (N^\hat{N})

The Schnabel method is an extension of the Lincoln-Petersen index for closed populations sampled across multiple consecutive intervals (t=1,2,…,kt = 1, 2, \dots, k). The estimator is defined as:

N^=∑t=1k(Ct⋅Mt)∑t=1kRt\hat{N} = \frac{\sum_{t=1}^k (C_t \cdot M_t)}{\sum_{t=1}^k R_t}

where:

  • CtC_t is the total number of individuals caught in sample tt.
  • MtM_t is the cumulative number of previously marked individuals present in the population just prior to taking sample tt.
  • RtR_t is the number of marked individuals recaptured in sample tt.

Tabulate the values for each sampling day:

Day (tt)Total Captured (CtC_t)Cumulative Marked at Large (MtM_t)Recaptured Marked (RtR_t)Product (Ct⋅MtC_t \cdot M_t)
14040000040⋅0=040 \cdot 0 = 0
2505040405550⋅40=2,00050 \cdot 40 = 2{,}000
360608585121260⋅85=5,10060 \cdot 85 = 5{,}100
45050133133181850⋅133=6,65050 \cdot 133 = 6{,}650
  1. Calculate the numerator: ∑t=14(Ct⋅Mt)=0+2,000+5,100+6,650=13,750\sum_{t=1}^4 (C_t \cdot M_t) = 0 + 2{,}000 + 5{,}100 + 6{,}650 = 13{,}750
  2. Calculate the denominator: ∑t=14Rt=0+5+12+18=35\sum_{t=1}^4 R_t = 0 + 5 + 12 + 18 = 35
  3. Compute N^\hat{N}: N^=13,75035≈392.86≈393 grasshoppers\hat{N} = \frac{13{,}750}{35} \approx 392.86 \approx 393\text{ grasshoppers}

Part (b): Calculating the Population Density

Population density represents the number of individuals per unit area:

Density=N^Area\text{Density} = \frac{\hat{N}}{\text{Area}}

Given an enclosure area of 1,500 m21{,}500\text{ m}^2:

Density=392.86 grasshoppers1,500 m2≈0.262 grasshoppers/m2\text{Density} = \frac{392.86\text{ grasshoppers}}{1{,}500\text{ m}^2} \approx 0.262\text{ grasshoppers/m}^2


Part (c): Effect of Mark Loss on N^\hat{N}

  • Result: The calculated N^\hat{N} is an overestimate of the true population.
  • Justification: A core assumption of mark-recapture censuses is that marks are not lost or overlooked. When marks rub or molt off, individuals that were previously captured appear unmarked upon recapture. This artificially reduces the observed number of recaptures (RtR_t). Because the cumulative number of recaptures ∑Rt\sum R_t appears in the denominator of the Schnabel equation (N^∝1∑Rt\hat{N} \propto \frac{1}{\sum R_t}), an artificially low denominator produces a disproportionately large quotient, inflating the estimated population size N^\hat{N}.

Typical errors

OverviewACEpath Teacher's Diagnosis

01 The student misidentifies mark loss as causing an underestimate, erroneously reasoning that having fewer apparent marked individuals means the population appears smaller.

Sample Student Work

Reviewed by ACEpath Teacher
Step-by-Step Review
Walk Through It with Your Teacher
Partially correct
  1. Step 1

    ∑(Ct⋅Mt)=0+2000+5100+6650=13750  ⟹  ∑Rt=0+5+12+18=35  ⟹  N^=1375035=392.86\sum (C_t \cdot M_t) = 0 + 2000 + 5100 + 6650 = 13750 \implies \sum R_t = 0 + 5 + 12 + 18 = 35 \implies \hat{N} = \frac{13750}{35} = 392.86

  2. Step 2

    Density=392.861500  ⟹  Density=0.262 grasshoppers/m2\text{Density} = \frac{392.86}{1500} \implies \text{Density} = 0.262\text{ grasshoppers/m}^2

  3. Step 3

    Underestimate. If marks wear off, fewer marked grasshoppers are seen in the population, making the total population appear smaller than it actually is.

    Correct

    Overestimate. Mark loss causes marked grasshoppers to be recorded as unmarked, which lowers the number of recaptures (RtR_t). Because ∑Rt\sum R_t is in the denominator of N^=∑(Ct⋅Mt)∑Rt\hat{N} = \frac{\sum (C_t \cdot M_t)}{\sum R_t}, a smaller denominator yields an inflated, overestimated population size.

    You reasoned intuitively that fewer visible marks means a smaller population, but mathematically the number of recaptures is in the denominator. When marks are lost, RtR_t decreases, which makes ∑(Ct⋅Mt)∑Rt\frac{\sum (C_t \cdot M_t)}{\sum R_t} larger, producing an overestimate.

WHAT YOU DID WELL
Compliments from Your Teacher

You accurately tabulated the numerator products and summed them to 13,75013{,}750.

You summed the recaptures correctly and found N^≈392.86\hat{N} \approx 392.86.

You correctly calculated the population density by dividing the abundance by 1,500 m21{,}500\text{ m}^2.

THE ERROR
Where Your ACEpath Teacher Spotted the Mistake
Mark-Recapture Assumption Violation Direction

none

WHY IT HAPPENED
ACEpath Teacher's Diagnosis

You connected fewer observed marks directly to a smaller population size rather than tracing how mark loss changes the denominator (RtR_t) of the mark-recapture ratio.

REASONING PATH
Your ACEpath teacher’s suggested approach
Denominator Effect in Lincoln-Petersen and Schnabel Estimators
  • In mark-recapture estimators, N^∝1R\hat{N} \propto \frac{1}{R}. The number of recaptured marked individuals is inversely proportional to the population estimate.

  • Always write out where the affected variable lives in the formula: if an artifact lowers RR, dividing by a smaller number drives N^\hat{N} upward (overestimate).

CORRECTED PATH
ACEpath Teacher's Demonstration
  1. Identify the mathematical role of recaptures: N^=∑(Ct⋅Mt)∑Rt\hat{N} = \frac{\sum (C_t \cdot M_t)}{\sum R_t}.

  2. Analyze the violation: If acrylic marks wear off or molt away, previously marked grasshoppers are misclassified as unmarked.

  3. Evaluate the effect on variables: The observed tally of recaptured marked individuals, ∑Rt\sum R_t, becomes artificially smaller than it should be.

  4. Determine the directional bias: Since ∑Rt\sum R_t is in the denominator, decreasing the denominator increases the quotient N^\hat{N}, leading to an overestimate.

Final answer

Overestimate; mark loss artificially reduces the number of recaptures (RtR_t) in the denominator, inflating the estimated population size N^\hat{N}.

NEXT STEPS
Final Takeaways from Your Teacher
  • Practice checking assumption violations by identifying whether the impacted count is in the numerator or denominator of the estimator.

  • Review how other violations (e.g., trap-shyness, trap-happiness, mortality of marked animals) affect the ratio of marked to unmarked individuals in recaptures.

OverviewACEpath Teacher's Diagnosis

02 The student shifts the indices and pairs each capture count Cₜ with the post-release marked total instead of the marked individuals available at large Mₜ.

Sample Student Work

Reviewed by ACEpath Teacher
Step-by-Step Review
Walk Through It with Your Teacher
Partially correct
  1. Step 1

    Day 1: 40 ⋅\cdot 40 = 1600   ⟹  \implies Day 2: 50 ⋅\cdot 85 = 4250   ⟹  \implies Day 3: 60 ⋅\cdot 133 = 7980   ⟹  \implies Day 4: 50 ⋅\cdot 181 = 9050   ⟹  \implies \sum (C_t ⋅\cdot M_t) = 1600 + 4250 + 7980 + 9050 = 22880

    Correct

    ∑t=14(Ct⋅Mt)=(40⋅0)+(50⋅40)+(60⋅85)+(50⋅133)=13,750\sum_{t=1}^4 (C_t \cdot M_t) = (40 \cdot 0) + (50 \cdot 40) + (60 \cdot 85) + (50 \cdot 133) = 13{,}750

    You multiplied each catch count CtC_t by the cumulative marked total after the day's newly marked grasshoppers were released. In the Schnabel index, MtM_t is the number of marked individuals at large in the population before sample tt is collected (so M1=0M_1 = 0, M2=40M_2 = 40, M3=85M_3 = 85, and M4=133M_4 = 133).

  2. Step 2

    \sum R_t = 5 + 12 + 18 = 35   ⟹  \implies N^\hat{N} = 2288035\frac{22880}{35} = 653.71

    Correct

    N^=13,75035≈392.86≈393 grasshoppers\hat{N} = \frac{13{,}750}{35} \approx 392.86 \approx 393\text{ grasshoppers}

    You correctly computed ∑Rt=35\sum R_t = 35, but substituting your inflated numerator from Step 1 caused your calculated population estimate N^\hat{N} to be too high.

  3. Step 3

    Density\text{Density} = 653.711500\frac{653.71}{1500}   ⟹  \implies Density\text{Density} = 0.436 grasshoppers/m\text{ grasshoppers/m}^2

    Correct

    Density=392.861,500 m2≈0.262 grasshoppers/m2\text{Density} = \frac{392.86}{1{,}500\text{ m}^2} \approx 0.262\text{ grasshoppers/m}^2

    You used the correct density formula, but the result was affected by the upstream error in N^\hat{N}.

  4. Step 4

    Overestimate. Losing marks lowers the number of recaptures R_t in the denominator, which makes the calculated N^\hat{N} larger than the true value.

WHAT YOU DID WELL
Compliments from Your Teacher

You correctly calculated the cumulative sum of recaptured individuals ∑Rt=35\sum R_t = 35.

You applied the correct definition for population density by dividing total population by parcel area.

You correctly reasoned that mark loss reduces the denominator RtR_t and creates an overestimate of population size.

THE ERROR
Where Your ACEpath Teacher Spotted the Mistake
Index / Timing Misalignment in Multi-Sample Mark-Recapture

No, this is a conceptual timing confusion rather than a simple arithmetic slip: you systematically used the post-sampling marked count instead of the pre-sampling marked count at large for each day.

WHY IT HAPPENED
ACEpath Teacher's Diagnosis

Confusing the cumulative marked pool prior to sampling (MtM_t) with the cumulative marked pool after release (Mt+1M_{t+1}). On Day 1, no marked individuals exist in the population before capture begins, so M1=0M_1 = 0.

REASONING PATH
Your ACEpath teacher’s suggested approach
Timing of $M_t$
  • MtM_t is the number of marked individuals at large just before drawing sample tt.

  • Always set M1=0M_1 = 0 for the initial capture session. Each subsequent MtM_t equals the total marked individuals released prior to day tt.

Numerator Product Alignment
  • Every product term is Ct⋅MtC_t \cdot M_t, representing sample size multiplied by existing marked pool size.

  • Construct a four-column table (tt, CtC_t, MtM_t, RtR_t) before calculating the sums to prevent shifting row indices.

CORRECTED PATH
ACEpath Teacher's Demonstration
  1. Recompute each daily product using pre-sample marked counts: Day 1 is 40⋅0=040 \cdot 0 = 0, Day 2 is 50⋅40=2,00050 \cdot 40 = 2{,}000, Day 3 is 60⋅85=5,10060 \cdot 85 = 5{,}100, and Day 4 is 50⋅133=6,65050 \cdot 133 = 6{,}650.

  2. Sum the correct products: ∑(Ct⋅Mt)=0+2,000+5,100+6,650=13,750\sum (C_t \cdot M_t) = 0 + 2{,}000 + 5{,}100 + 6{,}650 = 13{,}750.

  3. Recalculate the estimate: N^=13,75035≈392.86≈393 grasshoppers\hat{N} = \frac{13{,}750}{35} \approx 392.86 \approx 393\text{ grasshoppers}.

  4. Recalculate population density: 392.861,500 m2≈0.262 grasshoppers/m2\frac{392.86}{1{,}500\text{ m}^2} \approx 0.262\text{ grasshoppers/m}^2.

Final answer

(a) N^≈393 grasshoppers\hat{N} \approx 393\text{ grasshoppers}; (b) 0.262 grasshoppers/m20.262\text{ grasshoppers/m}^2; (c) Overestimate (justification: lower RtR_t in denominator inflates N^\hat{N}).

NEXT STEPS
Final Takeaways from Your Teacher
  • Set up an explicit tracking table with columns for tt, CtC_t, MtM_t, RtR_t, and Ct⋅MtC_t \cdot M_t whenever solving multi-sample mark-recapture problems.

  • Double-check that M1=0M_1 = 0 in closed Schnabel censuses so the first product C1⋅M1C_1 \cdot M_1 is always zero.

  • Practice identifying whether biological violations (e.g., mark loss, trap-shyness, trap-happiness) affect the numerator or denominator of the estimator.

OverviewACEpath Teacher's Diagnosis

03 The student inverts the density formula by dividing parcel area by estimated population rather than dividing population by area.

Sample Student Work

Reviewed by ACEpath Teacher
Step-by-Step Review
Walk Through It with Your Teacher
Partially correct
  1. Step 1

    \sum (C_t ⋅\cdot M_t) = (40 ⋅\cdot 0) + (50 ⋅\cdot 40) + (60 ⋅\cdot 85) + (50 ⋅\cdot 133) = 13750   ⟹  \implies \sum R_t = 0 + 5 + 12 + 18 = 35   ⟹  \implies N^\hat{N} = 1375035\frac{13750}{35} = 392.86

  2. Step 2

    Density\text{Density} = 1500392.86\frac{1500}{392.86}   ⟹  \implies Density\text{Density} = 3.82 grasshoppers/m\text{ grasshoppers/m}^2

    Correct

    Density\text{Density} = N^Area\frac{\hat{N}}{\text{Area}} = 392.86 grasshoppers1,500 m2\frac{392.86\text{ grasshoppers}}{1{,}500\text{ m}^2} ≈\approx 0.262 grasshoppers/m\text{ grasshoppers/m}^2

    You inverted the ratio by placing the area (1,500 m21{,}500\text{ m}^2) in the numerator and the population (392.86392.86) in the denominator. Population density represents individuals per unit area, so you must divide the population size by the parcel area.

  3. Step 3

    Overestimate. Mark loss leads to fewer observed recaptures \text{Overestimate. Mark loss leads to fewer observed recaptures } R_t, and dividing by a smaller denominator produces a larger population estimate \text{, and dividing by a smaller denominator produces a larger population estimate } N^\hat{N}.

WHAT YOU DID WELL
Compliments from Your Teacher

You correctly tabulated and summed the cross-products and recaptures to find the Schnabel population estimate.

You demonstrated a clear understanding of mark-recapture assumptions by explaining how mark loss reduces the denominator and inflates the final population estimate.

THE ERROR
Where Your ACEpath Teacher Spotted the Mistake
Inverted density ratio

no

WHY IT HAPPENED
ACEpath Teacher's Diagnosis

You confused population density (individuals per unit area, NArea\frac{N}{\text{Area}}) with mean area available per individual (area per individual, AreaN\frac{\text{Area}}{N}), leading to an inverted quotient with swapped units.

REASONING PATH
Your ACEpath teacher’s suggested approach
Match the fraction to the desired unit
  • When computing a density expressed in individuals/m2\text{individuals/m}^2, the count of individuals must be in the numerator and square meters in the denominator: Density=PopulationArea\text{Density} = \frac{\text{Population}}{\text{Area}}.

  • Check your unit label against your fraction: if your unit is grasshoppers/m2\text{grasshoppers}/\text{m}^2, ensure your calculation is grasshoppersm2\frac{\text{grasshoppers}}{\text{m}^2}.

CORRECTED PATH
ACEpath Teacher's Demonstration
  1. State the definition of population density: Density=N^Area\text{Density} = \frac{\hat{N}}{\text{Area}}.

  2. Substitute N^≈392.86 grasshoppers\hat{N} \approx 392.86\text{ grasshoppers} and Area=1,500 m2\text{Area} = 1{,}500\text{ m}^2.

  3. Evaluate the quotient: 392.861500≈0.262 grasshoppers/m2\frac{392.86}{1500} \approx 0.262\text{ grasshoppers/m}^2.

Final answer

(a) N^≈393 grasshoppers\hat{N} \approx 393\text{ grasshoppers}; (b) 0.262 grasshoppers/m20.262\text{ grasshoppers/m}^2; (c) Overestimate, because mark loss artificially reduces RtR_t in the denominator.

NEXT STEPS
Final Takeaways from Your Teacher
  • Always write down the units alongside your numbers when substituting into a rate or density formula (e.g., 392.86 grasshoppers1500 m2\frac{392.86\text{ grasshoppers}}{1500\text{ m}^2}) to guarantee that the numerator and denominator match the required unit.

Lock it in

Similar question

01

An ecologist marks 8080 beetles in a forest plot. In a second sample of 100100 beetles, 2020 are marked. If tags on 55 marked beetles fell off before the second sample and went uncounted, calculate the estimated population size with the observed data versus the true estimated population size.

Answer

Observed estimate: N^obs=M⋅CRobs=80⋅10020=400 beetles\hat{N}_{\text{obs}} = \frac{M \cdot C}{R_{\text{obs}}} = \frac{80 \cdot 100}{20} = 400\text{ beetles}. True estimate (if marks had stayed on, Rtrue=20+5=25R_{\text{true}} = 20 + 5 = 25): N^true=80⋅10025=320 beetles\hat{N}_{\text{true}} = \frac{80 \cdot 100}{25} = 320\text{ beetles}. The mark loss caused an overestimate of 8080 beetles (25%25\% inflation).

02

A researcher uses the Schnabel method across 33 sampling events for small rodents. By Day 3, predatory owls preferentially prey upon marked rodents due to their bright fluorescent tags, reducing their numbers in the wild. Predict whether the resulting N^\hat{N} will be an overestimate or underestimate, and explain your reasoning.

Answer

Overestimate. Differential predation on marked individuals reduces the pool of marked animals available for recapture, leading to fewer observed recaptures (RtR_t). Because recaptures appear in the denominator of N^=∑(Ct⋅Mt)∑Rt\hat{N} = \frac{\sum (C_t \cdot M_t)}{\sum R_t}, a deflated denominator yields an inflated, overestimated population size.

03

During a multi-sample mark-recapture survey of pond turtles, unmarked immigration occurs between sampling days, introducing new unmarked adults into the closed study area. Explain how this immigration affects the proportion of marked animals caught in subsequent samples and whether N^\hat{N} overestimates or underestimates the initial population size N0N_0.

Answer

Overestimate. Influx of unmarked individuals 'dilutes' the marked proportion in the population, lowering the probability of recapturing a marked individual (lowering RtR_t). With a lower RtR_t in the denominator, the calculated N^\hat{N} is larger than the true initial population N0N_0.

Loading workspace…