CoChat
Sign Up
Cited by 40Evacuation and Crowd DynamicsFacility Location and Emergency ManagementEarthquake and Tsunami Effects

The DOI matches its Crossref record; 16 of the 16 references it cites were found in a registry or an index; none of the 5 sources its claims cite is checked for support yet.

Sensitivity of tsunami evacuation modeling to direction and land cover assumptions

Open access
Schmidtlein, Mathew C., Wood, Nathan J.·2015·Applied Geography
DOI 10.1016/j.apgeog.2014.11.014
Open this paper in CoChat — free accountOpen PDF

Abstract

Although anisotropic least-cost-distance (LCD) modeling is becoming a common tool for estimating pedestrian-evacuation travel times out of tsunami hazard zones, there has been insufficient attention paid to understanding model sensitivity behind the estimates. To support tsunami risk-reduction planning, we explore two aspects of LCD modeling as it applies to pedestrian evacuations and use the coastal community of Seward, Alaska, as our case study. First, we explore the sensitivity of modeling to the direction of movement by comparing standard safety-to-hazard evacuation times to hazard-to-safety evacuation times for a sample of 3985 points in Seward's tsunami-hazard zone. Safety-to-hazard evacuation times slightly overestimated hazard-to-safety evacuation times but the strong relationship to the hazard-to-safety evacuation times, slightly conservative bias, and shorter processing times of the safety-to-hazard approach make it the preferred approach. Second, we explore how variations in land cover speed conservation values (SCVs) influence model performance using a Monte Carlo approach with one thousand sets of land cover SCVs. The LCD model was relatively robust to changes in land cover SCVs with the magnitude of local model sensitivity greatest in areas with higher evacuation times or with wetland or shore land cover types, where model results may slightly underestimate travel times. This study demonstrates that emergency managers should be concerned not only with populations in locations with evacuation times greater than wave arrival times, but also with populations with evacuation times lower than but close to expected wave arrival times, particularly if they are required to cross wetlands or beaches.

Overview

AI snapshot

Written from the paper’s full text.

TL;DR

This study tests how pedestrian tsunami-evacuation travel times in Seward, Alaska, respond to reversing the least-cost-distance search direction and to varying land-cover speed conservation values (SCVs). It finds the reversed safety-to-hazard approach closely approximates hazard-to-safety times with a slightly conservative bias, and that the model is relatively robust to land-cover SCV changes, though uncertainty and underestimation are greater for higher evacuation times and wetland/shore crossings.

Key findings

  • For 3,985 sampled points, safety-to-hazard baseline times were strongly related to hazard-to-safety normal-direction times (R² = 0.976, p < 0.0001), with a median total difference of 0.53 min and median proportional difference of 0.22. Differences were ≥ 0 for 98.9% of points; the 1.1% negative differences all had very short evacuation times (≤ 1.2 min) and normal-direction times at most 0.01 s greater than baseline.
  • The baseline safety-to-hazard model slightly overestimated evacuation times, which the authors call a conservative bias preferred for risk reduction, and it requires far less processing than running a separate model for every hazard-zone cell.
  • Across 1,000 Monte Carlo land-cover SCV sets, the model was relatively robust: median mean difference from baseline was 0.10 min (~6 s), and median absolute deviation was 0.05 min (~3 s).
  • Local sensitivity was greatest in areas with higher evacuation times or wetland/shore land cover. Negative bias dominated almost all of the study area, with largest biases about 35 s; wetlands/shore made the strongest contribution to uncertainty magnitude, negative bias, and loss of precision. Base evacuation times correlated with uncertainty magnitude (Spearman r_s = 0.68) and bias (r_s = −0.63), and more weakly with precision (r_s = −0.39).

Methods

  • Case study: Seward, Alaska, using a baseline anisotropic least-cost-distance (LCD) pedestrian-evacuation model in ESRI ArcMap 10.1 following Wood and Schmidtlein (2012). Inputs included a 2009 5-ft LiDAR DEM, 2005 1-m IKONOS-derived land cover, the maximum tsunami-hazard zone from Suleimani et al. (2010), and a 1.1 m/s slow-walking speed; land-cover SCVs were based on Soule and Goldman (1972) terrain energy coefficients.
  • Direction sensitivity: a regular sample of 3,985 non-zero evacuation-time points was drawn; a separate anisotropic model was run for each point with the point as origin and normal Tobler slope SCVs, then least-cost paths to safe zones were derived and compared with baseline safety-to-hazard times.
  • Land-cover SCV sensitivity: Monte Carlo simulation generated 1,000 sets of SCVs using truncated normal distributions, with SD 0.10 for wetlands, shore, and heavy brush and larger SDs for light brush and grass/dirt/gravel due to truncation near 1; 1,000 evacuation surfaces were created and assessed with overall accuracy/precision metrics, local uncertainty measures, and spatial simultaneous autoregressive error models using 8 nearest neighbors.

Limitations

  • The paper states a continuing limitation is the lack of an empirical basis for land-cover SCVs; the Soule and Goldman (1972) energy coefficients were developed in different land-cover areas, so analysts must arbitrarily assign coefficients to land-cover types.
  • The paper states that without known evacuation times, it assumes the base model SCVs are approximately correct; low sensitivity to SCV changes does not prove evacuation times are correct.
  • The paper reports normality and heteroscedasticity problems in the spatial regression diagnostics but continues because these do not affect slope coefficient estimation and results were highly significant; it also notes that small negative biases may have limited practical significance in Seward’s constrained hazard zone but could produce more substantial underestimates in a more extensive tsunami evacuation zone.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1016/j.apgeog.2014.11.014

Cites

16 of 16 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

0 of 5 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    Proportional differences between baseline and normal-direction evacuation times are greatest when baseline model evacuation times are shortest.

    “and that the proportional differences are greatest when baseline model evacuation times are shortest (Fig. 4c).”
    • Also stated in: § Sensitivity of LCD model to path direction · plain
      “the greatest proportional differences are in areas where evacuation times are the lowest.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: The regular sample of 3985 points with non-zero evacuation times is representative of Seward's tsunami-hazard zone.
  • Findingown data

    Safety-to-hazard evacuation times slightly overestimated hazard-to-safety evacuation times in Seward.

    “Safety-to-hazard evacuation times slightly overestimated hazard-to-safety evacuation times”
    • Also stated in: § Sensitivity of LCD model to path direction · plain
      “In almost every case, the baseline evacuation model also tends to slightly overestimate the standard-direction model evacuation times.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The regular sample of 3985 points with non-zero evacuation times is representative of Seward's tsunami-hazard zone.
  • Findingown data

    Total differences between baseline and normal-direction evacuation times increase as baseline model evacuation times increase.

    “Scatterplots showing comparisons of baseline model evacuation times and normal direction model suggest that total differences between the baseline and normal direction model evacuation times increase as baseline model evacuation times increase (Fig. 4a, b)”
    • Also stated in: § Sensitivity of LCD model to path direction · plain
      “The largest differences between evacuation times are in areas where baseline model times are the highest”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The regular sample of 3985 points with non-zero evacuation times is representative of Seward's tsunami-hazard zone.
  • Findingown data

    The magnitude of local model sensitivity to land cover SCVs is greatest in areas with higher evacuation times or with wetland or shore land cover types.

    “with the magnitude of local model sensitivity greatest in areas with higher evacuation times or with wetland or shore land cover types”
    Quote in the paperSubstantial claim
    • Rests on an assumption: No consistent, empirically based relationship between land cover and travel speeds could be identified in the literature.
  • Findingown data

    There is a very strong and statistically significant relationship between baseline model and normal-direction evacuation times (R2 = 0.976, p < 0.0001).

    “A regression analysis suggests a very strong and statistically significant relationship between the baseline model and normaldirection evacuation times (R 2 ¼ 0.976, p < 0.0001).”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: The regular sample of 3985 points with non-zero evacuation times is representative of Seward's tsunami-hazard zone.
  • It may state two separate claims as one. 22
  • Probably not a substantial claim. 2

  • It may state two separate claims as one.

    Findingown data

    The median of the distribution of mean differences between the base scenario and the Monte Carlo scenarios is 0.10 min (about 6 s), and the median absolute deviation is 0.05 min (about 3 s).

    “The median of this distribution is 0.10 min, or about 6 s, and the median absolute deviation is 0.05 min, or about 3 s.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Findingown data

    Negative values (possible underestimates) dominate almost all of the study area, while overall bias remains relatively small (largest biases about 35 s).

    “By far the dominant observable trend is that while overall bias remains relatively small (with the largest biases of about 35 s), negative values (or possible underestimates) dominate almost all of the study area.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Findingnone given

    Both uncertainty magnitude and negative bias are greater in areas with higher evacuation times farther from the safe zones.

    “Additionally, both uncertainty magnitude and negative bias are greater in areas with higher evacuation times farther from the safe zones.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Findingown data

    Areas with high uncertainty magnitude include dirt and gravel areas at the southern end of the small boat harbor and wetland areas directly north of the lagoon; the remainder of the study area has relatively low magnitude of uncertainty.

    “While there are higher magnitudes of uncertainty around the dirt and gravel areas at the southern end of the small boat harbor, as well as the wetland areas directly to the north of the lagoon, the remainder of the study area has a relatively low magnitude of uncertainty (Fig. 7a).”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Findingown data

    The most striking relationships in the hexplots are the positive linear relationship between base model evacuation times and uncertainty magnitude and the negative linear relationship between base model evacuation times and uncertainty bias.

    “The most striking relationships evident in these hexplots are the positive linear relationships between base model evacuation times and uncertainty magnitude (Fig. 8a) and a negative linear relationship with uncertainty bias (Fig. 8c).”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Findingreasoning

    The impervious category had the least impact on both uncertainty bias and magnitude.

    “In addition to being the only land cover type that improved uncertainty precision, the impervious category had the least impact on both uncertainty bias and magnitude.”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Findingown data

    Areas with evacuation paths that crossed more wetlands or unconsolidated shoreland were more likely to have a higher magnitude and worse precision of uncertainty, and greater underestimates in evacuation times.

    “Areas with higher evacuation times and evacuation paths that crossed more wetlands or unconsolidated shoreland were more likely to have a higher magnitude and worse precision of uncertainty, and greater underestimates in evacuation times.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Findingown data

    Values for uncertainty bias were predominantly negative across all land cover categories, with larger negative values overall for the wetlands, shore and grass, dirt, gravel categories.

    “Values for uncertainty bias were predominantly negative across all land cover categories, with larger negative values overall for the wetlands, shore and grass, dirt, gravel categories (Fig. 9c).”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Findingown data

    Pockets of more precise estimates are west of the small boat harbor and along the southern edge of the alluvial fan in areas primarily classified as impervious.

    “Fig. 7b shows pockets of more precise estimates to the west of the small boat harbor, and along the southern edge of the alluvial fan in areas that are primarily classified as impervious (Fig. 1D).”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Findingown data

    Negative differences were only seen for 1.1 percent of the sample points, all of which had very short evacuation times (of 1.2 min or less) and had normal direction times at most 0.01 s greater than base model evacuation times.

    “Negative differences were only seen for 1.1 percent of the sample points, all of which had very short evacuation times (of 1.2 min or less) and had normal direction times at most 0.01 s greater than base model evacuation times.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Findingown data

    Areas with higher evacuation times were more likely to have a higher magnitude and worse precision of uncertainty, and greater underestimates in evacuation times.

    “Areas with higher evacuation times and evacuation paths that crossed more wetlands or unconsolidated shoreland were more likely to have a higher magnitude and worse precision of uncertainty, and greater underestimates in evacuation times.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Findingown data

    Most evacuation paths crossed at least some pixels from the impervious, light brush, and grass, dirt, gravel land cover categories, while far fewer paths crossed pixels with wetlands, shore or heavy bush land covers.

    “These values show that while most evacuation paths crossed at least some pixels from the impervious; light brush; and grass, dirt, gravel land cover categories, far fewer paths crossed pixels with wetlands, shore or heavy bush land covers.”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Findingown data

    Areas of model imprecision include the wetland area directly north of the lagoon, at the southern end of the study area, and in many small areas abutting the safe zone classified as a mix of dirt, gravel, and grass.

    “Areas of model imprecision include the wetland area directly north of the lagoon, at the southern end of the study area, and in many small areas abutting the safe zone that are classified as a mix of dirt, gravel, and grass.”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Interpretationreasoning

    Individuals in locations with the greatest travel times (small-boat harbor and adjacent waterfront) may have enough time to evacuate from tectonically generated tsunamis, provided evacuation models correctly estimate pedestrian travel times and at-risk individuals do not delay.

    “therefore individuals in these locations may have enough time to evacuate from tectonically generated tsunamis, providing the evacuation models correctly estimate pedestrian travel times and at-risk individual do not delay during an evacuation.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: At-risk individuals do not delay during an evacuation.

  • It may state two separate claims as one.

    Interpretationown data

    Reversing the direction of the LCD search in the anisotropic pedestrian evacuation model does not result in identical evacuation times but leads to a very close approximation.

    “Results suggest that while reversing the direction of the LCD search in the anisotropic pedestrian evacuation model does not result in identical evacuation times, it does lead to a very close approximation.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The regular sample of 3985 points with non-zero evacuation times is representative of Seward's tsunami-hazard zone.

  • It may state two separate claims as one.

    Interpretationown data

    In areas with higher evacuation times or wetland or shore land cover types, model results may slightly underestimate travel times.

    “where model results may slightly underestimate travel times.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: No consistent, empirically based relationship between land cover and travel speeds could be identified in the literature.

  • It may state two separate claims as one.

    Interpretationown data

    The base model is fairly accurate and precise.

    “The distribution of the mean differences between the base scenario and the Monte Carlo scenarios suggests that the base model is fairly accurate (Fig. 6).”
    • Also stated in: § Overall model performance · hedged
      “Both values are relatively low, suggesting that the overall model has fairly good accuracy and precision”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: If the base model SCVs are approximately correct, then the accuracy and precision of the model can be understood by looking at the distribution of differences between the base model and the Monte Carlo scenarios.
    • The authors state this as an assumption: The base model SCVs are approximately correct.

  • It may state two separate claims as one.

    Interpretationreasoning

    The baseline model's bias is a solidly conservative one that would result in residents having slightly more time than estimated, which is the preferred bias from a risk-reduction perspective.

    “Although the baseline model does result in biased evacuation times, the bias seems to be a solidly conservative one that would result in residents having slightly more time than estimated, which is the preferred bias from a risk-reduction perspective.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim. It may state two separate claims as one.

    Study design and conductreasoning

    We continued with our analysis despite finding both normality and heteroscedasticity problems because these issues do not affect estimation of slope coefficients, and because of the extreme statistical significance of model results.

    “We continued with our analysis despite finding both normality and heteroscedatiscity problems both because these issues do not affect estimation of slope coefficients, and also because the extreme statistical significance of model results.”
    Quote in the paperSubstantial claim

  • It may state two separate claims as one.

    Interpretationown data

    The baseline model can slightly underestimate evacuation times, but it is far more likely to result in over-estimates.

    “This suggests that while it is possible that the baseline model can slightly underestimate evacuation times, it is far more likely to result in over-estimates.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim. It may be too vague to check against evidence. It may state two separate claims as one.

    Interpretationreasoning

    The model is performing well and increases confidence in the modeled results.

    “While this does not provide assurance that the evacuation times are correct, as an empirically measured relationship between land cover type and evacuation speed would, it does suggest that the model is performing well and increases confidence in the modeled results.”
    Quote in the paperNot a substantial claim

  • It may state two separate claims as one.

    Interpretationreasoning

    Some caution is warranted in interpreting results with higher evacuation times and where there is wetland or shore land cover.

    “Overall, these results suggest some caution is warranted in interpreting results with higher evacuation times (where they may be a greater magnitude of uncertainty), and where there is wetland or shore land cover (where there may be a greater magnitude of uncertainty, as well as a potential for underestimates of evacuation times).”
    Quote in the paperSubstantial claim
22 of 22

  • The authors state this as an assumption.

    If the base model SCVs are approximately correct, then the accuracy and precision of the model can be understood by looking at the distribution of differences between the base model and the Monte Carlo scenarios.

    “Lacking such a measure, we assume that if the base model SCVs are approximately correct, then the accuracy and precision of the model can be understood by looking at the distribution of the differences between the base model and the Monte Carlo scenarios.”

  • The paper relies on this without support.

    No consistent, empirically based relationship between land cover and travel speeds could be identified in the literature.

    “Because no consistent, empirically based relationship between land cover and travel speeds could be identified in the literature”

  • The authors state this as an assumption.

    At-risk individuals do not delay during an evacuation.

    “providing the evacuation models correctly estimate pedestrian travel times and at-risk individual do not delay during an evacuation.”

  • The paper relies on this without support.

    The safety-to-hazard approach has shorter processing times than the hazard-to-safety approach.

    “and shorter processing times of the safetyto-hazard approach make it the preferred approach.”

  • The authors state this as an assumption.

    The base model SCVs are approximately correct.

    “we assume that if the base model SCVs are approximately correct”

  • The paper relies on this without support.

    The regular sample of 3985 points with non-zero evacuation times is representative of Seward's tsunami-hazard zone.

    “a regular sample of 3985 points with non-zero evacuation times was drawn from the baseline pedestrian evacuation model.”

  • The authors state this as an assumption.

    The Soule and Goldman (1972) energy coefficients are reasonably close to representing the true relationship between land cover type and pedestrian speeds.

    “if they are even relatively close to representing the true relationship between land cover type and pedestrian speeds”

  • The authors state this as an assumption.

    The arbitrary assignment of existing energy coefficients to land cover types used in this study is adequate for the analysis.

    “analysts must essentially arbitrarily decide which of the existing coefficients will be assigned to which land cover type.”
8 of 8

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Mean travel-time difference disagrees with the reported means

    Major

    Table 2 reports mean baseline and normal-direction times of 2.43 and 2.53 minutes. Their difference is −0.10 minutes, not the reported positive 0.71 minutes, reversing the direction of the mean comparison.

    Reported 0.71 · recomputed -0.1

  • Conclusion reverses the reported precision relationship

    Minor

    The paper defines worse precision as a higher coefficient of variation. Its reported negative correlation between evacuation time and that measure implies that higher evacuation times tend to have better, not worse, precision as stated in the conclusion.

  • Baseline median differs between tables for the same sample

    Minor

    Table 2 gives a baseline median of 3.23 minutes, whereas Table 3 gives 2.43 minutes for the base evacuation times of the same 3985 sampled paths.

3 of 3

Citations

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Metadata

Authors
Schmidtlein, M. C., & Wood, N. J.
Venue
Applied Geography
Publisher
Elsevier BV
Published
2015-01
Volume
56
Pages
154-163
ISSN
0143-6228 (ISSN-L)
DOI
10.1016/j.apgeog.2014.11.014
OpenAlex
W2079638035
Cited by
40
References
15
Citations by year
2026: 3 · 2025: 3 · 2024: 1 · 2023: 7 · 2022: 7 · 2021: 6 · 2020: 3 · 2019: 2 · 2018: 4 · 2017: 4
FWCI
4.53
Type
journal article
Language
en
Open access
Licence
Locations
Applied Geography (publishedVersion, cc-by-nc-nd)
Funders
U.S. Geological Survey
Topics
  • Social Sciences › Safety Research › Evacuation and Crowd Dynamics
  • Social Sciences › Decision Sciences › Management Science and Operations Research › Facility Location and Emergency Management
  • Physical Sciences › Engineering › Civil and Structural Engineering › Earthquake and Tsunami Effects
Keywords
tsunami evacuation, model sensitivity, movement direction, Monte Carlo sensitivity analysis

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-11-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-11-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review