📝 Modeling with differential equations applications (38 MCQs)

📖 Explanation: The uninhibited model $\frac{dy}{dt} = ky$ assumes unlimited resources, leading to exponential growth without bound. When a population stabilizes at a carrying capacity $L$, environmental limitations such as food, space, or competition become significant. The correct model must incorporate a term that reduces growth as $y$ approaches $L$, such as the logistic factor $(1 - y/L)$. Using the uninhibited model ignores these density-dependent factors, making it invalid near equilibrium.

📖 Explanation: In mixing problems, the rate out depends on the instantaneous concentration $y(t)/V(t)$. If inflow and outflow rates differ, volume $V(t)$ changes over time. Using a constant denominator like 100 assumes steady-state volume, which is only valid when inflow equals outflow. This error leads to incorrect concentration calculations and wrong long-term behavior predictions. Students must recognize that $V(t) = V_0 + (r_{in} - r_{out})t$ and substitute accordingly in the outflow term.

📖 Explanation: Newton’s Law involves the temperature difference $T - T_e$, not absolute temperature. The correct setup is $\ln((70-20)/(90-20)) = k(10)$, giving $\ln(50/70) = 10k$. Using raw temperatures ignores the ambient baseline and yields an incorrect decay constant. This misconception arises from misremembering the law’s structure. The differential equation models how the gap between object and environment shrinks exponentially, so all calculations must reference $T - T_e$ to maintain physical consistency.

📖 Explanation: The decay constant $k$ satisfies $0.8 = e^{-5k}$, so $k = -\ln(0.8)/5$. Half-life is $T = \ln 2 / k = 5\ln 2 / (-\ln 0.8)$. The student incorrectly used 0.8 directly in the denominator, treating the remaining fraction as the exponent argument rather than solving for $k$ first. This reveals confusion between the exponential model’s parameters and observable quantities. Proper analysis requires isolating $k$ from the given data before computing derived metrics like half-life.

📖 Explanation: The growth rate $\frac{dy}{dt} = ky(1 - y/L)$ is a quadratic in $y$, peaking at $y = L/2$. This inflection point marks the transition from accelerating to decelerating growth. In epidemiology, this corresponds to peak infection rate, crucial for healthcare planning. Misidentifying this maximum leads to poor resource allocation. The symmetry of the logistic curve around $L/2$ ensures equal time spent below and above this threshold under ideal conditions, making it a critical benchmark for intervention timing.

📖 Explanation: Separating variables requires dividing by $y^2$, which assumes $y \neq 0$. However, $y=0$ is itself a valid solution satisfying both the ODE and $y(0)=0$. By algebraically manipulating the equation, the student inadvertently discarded this equilibrium solution. This highlights a critical pitfall: always check for constant solutions before separating variables. In modeling contexts, such equilibria often represent extinction or steady states with physical significance that non-equilibrium solutions cannot capture.

📖 Explanation: Horizontal slopes occur when $\frac{dy}{dx}=0$, requiring numerator zero at $y=2$. Vertical slopes imply undefined derivative, occurring when denominator is zero at $x=0$. Only $\frac{dy}{dx} = \frac{y-2}{x}$ satisfies both: numerator vanishes at $y=2$, denominator at $x=0$. Other options fail one condition. Interpreting slope fields requires linking geometric features to algebraic structure. This skill is essential for validating models when analytical solutions are unavailable, allowing researchers to infer system behavior directly from directional patterns.

📖 Explanation: Euler’s global error is theoretically $O(\Delta x)$, but each step introduces round-off error proportional to machine precision. As $\Delta x$ decreases, more steps amplify cumulative round-off, eventually offsetting truncation error reduction. This creates an optimal step size beyond which accuracy degrades. Students often assume smaller steps always improve results, neglecting numerical stability. Understanding this trade-off is vital for reliable computational modeling, especially in stiff systems where error dynamics are complex.

📖 Explanation: Half-life $T = \ln 2 / k$ is inversely proportional to $k$. Doubling $k$ halves $T$, meaning drug clears faster. Clinically, this necessitates more frequent dosing to maintain therapeutic levels. Misunderstanding this inverse relationship could lead to subtherapeutic dosing in fast metabolizers or toxicity in slow ones. Pharmacokinetic modeling must account for individual $k$ variations through personalized medicine approaches, emphasizing that population averages mask critical inter-patient differences affecting treatment efficacy.

📖 Explanation: Terminal velocity occurs when drag balances weight: $cv_\tau = mg \Rightarrow v_\tau = mg/c$. Since $c$ depends on shape and fluid properties (not mass), doubling $m$ doubles $v_\tau$. Heavier objects fall faster under linear drag because gravity scales with mass while drag does not. This contrasts with vacuum free-fall where all masses accelerate equally. Recognizing parameter dependencies prevents erroneous assumptions about scale invariance in resistive media, crucial for engineering design involving varying payloads.

📖 Explanation: Separation requires form $h(y)dy = g(x)dx$. Here, $x+y$ couples variables inseparably; no algebraic manipulation isolates them. Attempting integration treats $x$ as constant in $\int dy/(x+y)$, violating multivariable calculus rules. This common error stems from misidentifying equation type. Correct methods include substitution ($u=x+y$) or integrating factors for linear equations. Diagnosing non-separability early avoids wasted effort and reinforces classification skills essential for selecting appropriate solution techniques in modeling.

📖 Explanation: From $t = -\ln(y/y_0)/k$, error propagation gives $\Delta t \approx |\partial t / \partial (y/y_0)| \cdot \Delta(y/y_0) = \frac{1}{k(y/y_0)} \cdot \Delta(y/y_0)$. For small errors, relative age error $\Delta t / t \approx \Delta(y/y_0) / [(y/y_0) \ln(y_0/y)]$. Near typical values, this amplifies measurement uncertainty significantly. A 2% activity error can cause >100-year age discrepancy. This underscores carbon dating’s precision limits and why multiple samples are needed, illustrating how model sensitivity affects archaeological conclusions.

📖 Explanation: Closed orbits indicate periodic oscillations without damping or growth, characteristic of conservative systems like Lotka-Volterra. This arises from specific nonlinear coupling preserving a first integral (analogous to energy). Real ecosystems usually have dissipation, making true cycles rare; observed quasi-cycles often reflect external forcing or stochasticity. Recognizing orbit topology helps distinguish idealized models from reality. Assuming closed orbits imply stability is misleading—they’re neutrally stable, easily disrupted by perturbations, highlighting the fragility of such predictions in ecological forecasting.

📖 Explanation: At $t=0$, $y=0$ for pure water. Correct solution satisfies $y(0)=0$: $c_{in}V(1-1)=0$. Student’s version gives $y(0)=c_{in}V$, implying initial salt content equals steady state. This suggests they solved homogeneous equation $\frac{dy}{dt} = -\frac{r}{V}y$ instead of nonhomogeneous $\frac{dy}{dt} = rc_{in} - \frac{r}{V}y$. Missing particular solution ignores input source. Initial conditions anchor transient response; neglecting them yields physically impossible states, emphasizing the need to verify boundary behavior in derived formulas.

📖 Explanation: New rate k' = 1.1k, new doubling time T' = \ln 2 / (1.1k) = T / 1.1. Percentage decrease is (T - T')/T \times 100\% = (1 - 1/1.1) \times 100\% \approx 9.09\%. This nonlinear response arises because $T \propto 1/k$. Students often assume proportional changes transfer directly, but reciprocal relationships invert percentages. Understanding this elasticity is crucial in finance (rule of 70), epidemiology (epidemic doubling), and demography, where small rate shifts dramatically alter timelines despite intuitive expectations of linearity.

📖 Explanation: Euler’s error expansion includes $O(\Delta x)$ and higher-order terms. For larger $\Delta x$, $O(\Delta x)$ dominates, so halving $\Delta x$ nearly halves error. As $\Delta x$ shrinks, $O(\Delta x^2)$ terms contribute relatively more, slowing convergence toward asymptotic $O(\Delta x)$ regime. Additionally, round-off may play minor role here. This illustrates that theoretical error bounds describe limiting behavior, not finite-step performance. Practitioners must empirically validate convergence rates rather than assume ideal scaling, especially near computational thresholds.

📖 Explanation: Logistic model allows $y>L$ as a mathematical state representing transient overshoot, common in real systems with time lags (e.g., reproduction delays). While basic logistic assumes instantaneous regulation, extended models incorporate delays causing damped oscillations above $L$. Dismissing $y>L$ ignores ecological hysteresis and resilience concepts. The model’s validity isn’t binary; it approximates reality within context. Teaching should emphasize that $L$ is an attractor, not a hard barrier, and deviations inform understanding of system memory and recovery dynamics.

📖 Explanation: For cylinder, $\frac{dh}{dt} = -c\sqrt{h}$. Separating: $\int h^{-1/2} dh = -c \int dt \Rightarrow 2\sqrt{h} = -ct + C$. Time to drain from $h_0$ is proportional to $\sqrt{h_0}$. Thus, time from $H/4$ is $\sqrt{(H/4)/H} \times 10 = (1/2) \times 10 = 5$ min. This square-root dependence arises from velocity $\propto \sqrt{h}$. Students often assume linear scaling with height, missing the physics-driven nonlinearity. Recognizing functional forms in applied laws prevents erroneous extrapolation in engineering design involving fluid discharge.

📖 Explanation: Origin symmetry means if $(x,y)$ has slope $m$, then $(-x,-y)$ also has slope $m$. Thus $f(-x,-y) = f(x,y)$. Option B would give rotational symmetry of direction vectors but not slope values. Options C and D relate to axis symmetries. Interpreting geometric field properties algebraically tests deep understanding of how ODE structure manifests visually. This skill aids in verifying numerical simulations and identifying conserved quantities or invariant manifolds in dynamical systems without solving explicitly.

📖 Explanation: Instantaneous absorption assumes entire dose enters bloodstream at $t=0$, making $y(0)=D$ and peak immediate. Real absorption takes time, creating a rise phase before elimination dominates. This oversimplification overestimates early concentrations and misses $C_{max}$ timing, critical for toxicity assessment. More realistic models use compartmental absorption terms. Recognizing when simplifications break down ensures appropriate model fidelity for safety-critical applications, balancing tractability against physiological accuracy in medical dosing protocols.

📖 Explanation: Integrating factor need not include integration constant because multiplying ODE by any valid $\mu$ makes left side exact derivative. Constants in $\mu$ would appear multiplicatively on both sides and cancel during integration. For example, $\mu = Ce^{\int P dx}$ gives $C \frac{d}{dx}(e^{\int P} y) = C Q e^{\int P}$, and $C$ divides out. This flexibility simplifies computation. Misconception that $\mu$ requires constant stems from confusing it with general solution integration. Clarifying this avoids unnecessary complexity in solving linear ODEs.

📖 Explanation: At $y=L/2$, $\frac{dy}{dt} = k(L/2)(L/2) = kL^2/4$. Reducing $k$ to $k/2$ immediately halves the rate since expression is linear in $k$. No quadratic effect because $y$ hasn’t changed yet. This highlights direct parameter control in mitigation strategies. Students might confuse rate change with cumulative cases or think effects lag. Understanding instantaneous vs. integrated impacts informs real-time policy decisions during outbreaks, where timely $k$ reduction (via distancing) directly curbs transmission intensity at critical inflection points.

📖 Explanation: Solution is $I(t) = \frac{V}{R}(1 - e^{-Rt/L})$. Time constant $\tau = L/R$ governs exponential approach speed, independent of $V$. Amplitude scales with $V$, but rise time doesn’t. Confusing amplitude with timescale is common. In engineering, this means larger voltages don’t speed up circuit response; only component values do. Recognizing parameter roles prevents design errors in timing-sensitive applications like filters or motor controls, where settling time specifications constrain $L$ and $R$ choices regardless of operating voltage.

📖 Explanation: Correct formula: $y/y_0 = e^{-kt} \Rightarrow t = -\ln(y/y_0)/k$. Since $y/y_0 = 0.25 < 1$, $\ln(0.25) < 0$, so negative divided by positive $k$ gives positive $t$. Student likely wrote $t = \ln(0.25)/k$ without negation, yielding negative result. This reflects careless handling of logarithmic signs in decay models. Emphasizing that $-\ln(fraction) = \ln(1/fraction)$ prevents sign errors. Such mistakes undermine dating reliability, stressing need for dimensional and sign checks in scientific computation.

📖 Explanation: Maximum sustainable yield occurs at $y=L/2=500$, where natural growth peaks at $0.1*500*(0.5)=25$. Harvest exceeding this causes decline to extinction. Thus $h_{max}=25$. Students often pick $kL=100$ (uninhibited max) or $kL/4$ miscalculation. This optimization principle underlies fisheries and wildlife management. Exceeding MSY collapses stocks irreversibly. Deriving this requires finding max of growth function, not memorizing formulas. It exemplifies how calculus informs conservation policy through quantitative sustainability thresholds.

📖 Explanation: $y=y^2$ is convex (y&#039;&#039;=2y^3>0). Euler uses tangent at left endpoint, lying below curve for convex functions, thus underestimating true increase. Wait—actually for y&#039;=y^2, Euler *overestimates* because slope increases rapidly; tangent at $x_n$ has lower slope than average over interval? Correction: For y&#039;=y^2, y&#039;&#039;=2yy&#039;=2y^3>0, so function is convex upward. Tangent line lies *below* curve, so Euler *underestimates*. But blowup suggests overestimate. Re-evaluate: Exact solution blows up at $x=1$. Euler with fixed $\Delta x$ reaches infinity in finite steps because discrete iteration $y_{n+1}=y_n + y_n^2 \Delta x$ grows super-exponentially near singularity. Continuous solution integrates smoothly to pole; discrete map jumps across it. This illustrates numerical methods’ failure near singularities, requiring adaptive stepping or transformation.

📖 Explanation: With changing volume, steady-state concentration differs from constant-volume case. If $r_i > r_o$, $V \to \infty$, concentration $\to 0$ even with continuous solute input. Constant-$V$ model wrongly predicts nonzero limit. Qualitative misrepresentation includes whether system dilutes indefinitely or concentrates. Recognizing volume dynamics alters asymptotic analysis fundamentally. This affects environmental remediation designs where variable flow regimes determine pollutant persistence. Models must capture volumetric evolution to predict correct fate, not just short-term transients.

📖 Explanation: Rewriting as $\frac{d(y-x)}{dx} = y-x -1$ shows deviation $u=y-x$ satisfies u&#039;=u-1, with equilibrium $u=1 \Rightarrow y=x+1$. In cooling contexts, this could represent forced convection balance where object temp tracks ambient plus offset. Asymptotes reveal dominant balances in limits. Identifying them helps validate models against expected physical regimes. Here, linear asymptote suggests non-standard cooling; typical Newtonian cooling has horizontal asymptote. Thus, this model describes different physics, emphasizing need to match asymptotic behavior to system characteristics.

📖 Explanation: Since $T_{1/2} = \ln 2 / k$ and $\tau = 1/k$, then $\tau = T_{1/2} / \ln 2 \approx 1.44 T_{1/2}$. Mean lifetime appears naturally in integrals like $\int_0^\infty t e^{-kt} dt = 1/k^2$, and in quantum mechanics as inverse decay width. Half-life is intuitive for discrete halving; $\tau$ simplifies continuous mathematics. Choosing between them depends on context: communication favors $T_{1/2}$; calculation favors $\tau$. Understanding both prevents unit conversion errors in interdisciplinary work involving radioactivity or particle physics.

📖 Explanation: Lag phase reflects physiological adaptation before division begins. Constant $k$ fails here. Making $k(t)$ ramp up from zero (e.g., sigmoidal) captures delay without altering asymptotic behavior. Allee effect addresses low-density mating issues, not lag. Gompertz modifies growth shape but not initial delay explicitly. Logistic adds carrying capacity but still assumes immediate growth. Time-varying parameters offer flexible phenomenological fitting. This exemplifies adapting models to biological phases, recognizing that single-mechanism ODEs often need temporal modulation to match empirical trajectories accurately.

📖 Explanation: Under quadratic drag, $v_\tau \propto \sqrt{m}$; under linear, $v_\tau \propto m$. Quadratic dependence dampens mass sensitivity: 4× mass → 2× speed vs. 4× speed. Real skydivers experience quadratic drag, so weight variations have moderate impact on terminal velocity, enhancing safety across body types. Linear model would exaggerate risks for heavier individuals. Recognizing drag regime informs equipment design and training protocols. This comparison shows how correct physical modeling prevents over-engineering or underestimation of hazards in human-scale aerodynamics.

📖 Explanation: Steep slopes mean rapid change; fixed $\Delta x$ causes large local truncation error. Adaptive step sizing reduces $\Delta x$ where $|f|$ or $|\partial f/\partial y|$ is large, maintaining accuracy. Uniform stepping wastes effort in flat regions and fails in steep ones. Slope fields visually guide mesh refinement. This heuristic bridges geometric intuition and algorithmic implementation, essential for robust simulation of multiscale phenomena like shock waves or biochemical switches where dynamics vary orders of magnitude spatially or temporally.

📖 Explanation: Higher past $^{14}C$ means organism started with more $^{14}C$ than assumed standard. Measured remaining fraction appears larger relative to expected initial, suggesting less decay occurred → younger calculated age. Wait: if initial was higher, same remaining amount implies *more* decay needed to reach it, so calculated age would be *older*. Yes: $t = -\ln(R/R_0)/k$. If true R_0&#039; > R_0 (assumed), then true R/R_0&#039; < R/R_0, so \ln(R/R_0&#039;) < \ln(R/R_0) (more negative), so $t_{calc} = -\ln(R/R_0)/k > t_{true}$. Thus, uncalibrated date is too old. Calibration curves correct for historical fluctuations. Ignoring production variability introduces systematic bias, underscoring need for dendrochronological cross-validation in precise dating.

📖 Explanation: Discrete logistic map exhibits period-doubling and chaos for $k\Delta t > 2$, while continuous version never does. Large $\Delta t$ discretization introduces artificial nonlinearity. This shows numerical schemes aren’t neutral; they embed their own dynamics. Choosing $\Delta t$ isn’t just accuracy issue—it can qualitatively alter predicted behavior. Validating simulations requires checking convergence to continuous limit and distinguishing physical chaos from numerical artifacts. This cautions against blind trust in computational outputs without theoretical grounding.

📖 Explanation: System acts as low-pass filter. Phase lag $\phi = \arctan(\omega \tau)$ where $\tau = V/r$ is residence time. Larger $V$ or smaller $r$ increases lag. Frequency alone insufficient; system parameters set response characteristics. Initial conditions affect transient, not steady-state phase. Amplitude affects magnitude, not timing. Understanding frequency response enables design of buffers to smooth fluctuations in chemical processing or wastewater treatment. This connects ODE theory to signal processing concepts, showing how physical systems transform inputs based on intrinsic timescales.

📖 Explanation: Exact solution is quadratic: $y = \frac{m}{2}x^2 + bx + C$. Euler gives $y_{n+1} = y_n + (mx_n + b)\Delta x$, which is piecewise linear. This matches exact solution only if $m=0$ (constant slope). For $m \neq 0$, Euler accumulates error because it misses curvature. Linearity of RHS doesn’t imply linearity of solution. Misconception confuses equation form with solution geometry. Recognizing when numerical methods are exact (e.g., for autonomous linear ODEs with specific schemes) builds deeper appreciation of algorithm-solution compatibility.

📖 Explanation: Dimensional analysis: $[y]=people$, $[t]=time$, $[k]=1/(people·time)$, $[L]=people$. Integral $[y·t] = people·time$. Combination $L^2/k$ has units $people^2 / (1/(people·time)) = people^3·time$ — wrong. Try $L/k$: $people / (1/(people·time)) = people^2·time$ — still wrong. Actually, $\int y dt$ has units people·time. $L \cdot (1/kL) = 1/k$ has units people·time. So $\int y dt \propto 1/k$. But total cases should scale with $L$. Reconsider: In logistic, area under curve relates to $L \times$ characteristic time. Characteristic time $\sim 1/(kL)$, so area $\sim L/(kL) = 1/k$. Independent of $L$? Counterintuitive. Actually, rescaling $z=y/L$ gives $\int y dt = L \int z dt$, and $dz/dt = kL z(1-z)$, so time scales as $1/(kL)$, thus $\int z dt \sim 1/(kL)$, so total $\sim L \cdot 1/(kL) = 1/k$. Indeed, total person-time infected is independent of $L$ in this model! But question asks for estimation method. Dimensional analysis guides form; exact value requires integration. Option C is dimensionally inconsistent. Best answer is that dimensional analysis suggests proportionality to $1/k$, but among choices, none perfect. Given options, C is intended despite flaw, testing awareness that parameters combine nontrivially. In practice, $\int y dt = \frac{L}{k} \ln\left(\frac{L}{y_0} - 1\right)$ for $y_0<L$, confirming $L/k$ scaling. So A is actually correct dimensionally and structurally. Revise: A is right. Explanation focuses on scaling.

📖 Explanation: ODE becomes $\frac{dT}{dt} = k(T - T_0 - \alpha t)$. Rewrite as $\frac{dT}{dt} - kT = -k(T_0 + \alpha t)$, linear nonhomogeneous with time-dependent forcing. Integrating factor $e^{-kt}$ still applies; RHS integrates via parts. Quasi-steady assumes $dT/dt \approx 0$, invalid during transients. Second-order unnecessary. Small $\alpha$ may justify perturbation, but exact solution exists. Adapting standard methods to time-varying parameters extends applicability without abandoning analytical tools. This demonstrates flexibility in technique application, crucial for real-world scenarios where boundary conditions evolve, such as climate-controlled environments or seasonal thermal cycles.