Estimating global arthropod species richness

Refining probabilistic models using probability bounds analysis

Andrew J. Hamilton, Vojtech Novotný, Edward K. Waters, Yves Basset, Kurt K. Benke, Peter S. Grimbacher, Scott E. Miller, G. Allan Samuelson, George D Weiblen, Jian D.L. Yen, Nigel E. Stork

Research output: Contribution to journalArticle

35 Citations (Scopus)

Abstract

A key challenge in the estimation of tropical arthropod species richness is the appropriate management of the large uncertainties associated with any model. Such uncertainties had largely been ignored until recently, when we attempted to account for uncertainty associated with model variables, using Monte Carlo analysis. This model is restricted by various assumptions. Here, we use a technique known as probability bounds analysis to assess the influence of assumptions about (1) distributional form and (2) dependencies between variables, and to construct probability bounds around the original model prediction distribution. The original Monte Carlo model yielded a median estimate of 6. 1 million species, with a 90 % confidence interval of [3. 6, 11. 4]. Here we found that the probability bounds (p-bounds) surrounding this cumulative distribution were very broad, owing to uncertainties in distributional form and dependencies between variables. Replacing the implicit assumption of pure statistical independence between variables in the model with no dependency assumptions resulted in lower and upper p-bounds at 0. 5 cumulative probability (i. e., at the median estimate) of 2. 9-12. 7 million. From here, replacing probability distributions with probability boxes, which represent classes of distributions, led to even wider bounds (2. 4-20. 0 million at 0. 5 cumulative probability). Even the 100th percentile of the uppermost bound produced (i. e., the absolutely most conservative scenario) did not encompass the well-known hyper-estimate of 30 million species of tropical arthropods. This supports the lower estimates made by several authors over the last two decades.

Original languageEnglish (US)
Pages (from-to)357-365
Number of pages9
JournalOecologia
Volume171
Issue number2
DOIs
StatePublished - Jan 1 2013

Fingerprint

probabilistic models
refining
arthropod
arthropods
species richness
species diversity
uncertainty
cumulative distribution
Monte Carlo analysis
probability distribution
analysis
confidence interval
prediction
distribution

Keywords

  • Host specificity
  • Model
  • Monte Carlo
  • Uncertainty

Cite this

Hamilton, A. J., Novotný, V., Waters, E. K., Basset, Y., Benke, K. K., Grimbacher, P. S., ... Stork, N. E. (2013). Estimating global arthropod species richness: Refining probabilistic models using probability bounds analysis. Oecologia, 171(2), 357-365. https://doi.org/10.1007/s00442-012-2434-5

Estimating global arthropod species richness : Refining probabilistic models using probability bounds analysis. / Hamilton, Andrew J.; Novotný, Vojtech; Waters, Edward K.; Basset, Yves; Benke, Kurt K.; Grimbacher, Peter S.; Miller, Scott E.; Samuelson, G. Allan; Weiblen, George D; Yen, Jian D.L.; Stork, Nigel E.

In: Oecologia, Vol. 171, No. 2, 01.01.2013, p. 357-365.

Research output: Contribution to journalArticle

Hamilton, AJ, Novotný, V, Waters, EK, Basset, Y, Benke, KK, Grimbacher, PS, Miller, SE, Samuelson, GA, Weiblen, GD, Yen, JDL & Stork, NE 2013, 'Estimating global arthropod species richness: Refining probabilistic models using probability bounds analysis', Oecologia, vol. 171, no. 2, pp. 357-365. https://doi.org/10.1007/s00442-012-2434-5
Hamilton AJ, Novotný V, Waters EK, Basset Y, Benke KK, Grimbacher PS et al. Estimating global arthropod species richness: Refining probabilistic models using probability bounds analysis. Oecologia. 2013 Jan 1;171(2):357-365. https://doi.org/10.1007/s00442-012-2434-5
Hamilton, Andrew J. ; Novotný, Vojtech ; Waters, Edward K. ; Basset, Yves ; Benke, Kurt K. ; Grimbacher, Peter S. ; Miller, Scott E. ; Samuelson, G. Allan ; Weiblen, George D ; Yen, Jian D.L. ; Stork, Nigel E. / Estimating global arthropod species richness : Refining probabilistic models using probability bounds analysis. In: Oecologia. 2013 ; Vol. 171, No. 2. pp. 357-365.
@article{000a9e7f09ba407fbab6009566486c04,
title = "Estimating global arthropod species richness: Refining probabilistic models using probability bounds analysis",
abstract = "A key challenge in the estimation of tropical arthropod species richness is the appropriate management of the large uncertainties associated with any model. Such uncertainties had largely been ignored until recently, when we attempted to account for uncertainty associated with model variables, using Monte Carlo analysis. This model is restricted by various assumptions. Here, we use a technique known as probability bounds analysis to assess the influence of assumptions about (1) distributional form and (2) dependencies between variables, and to construct probability bounds around the original model prediction distribution. The original Monte Carlo model yielded a median estimate of 6. 1 million species, with a 90 {\%} confidence interval of [3. 6, 11. 4]. Here we found that the probability bounds (p-bounds) surrounding this cumulative distribution were very broad, owing to uncertainties in distributional form and dependencies between variables. Replacing the implicit assumption of pure statistical independence between variables in the model with no dependency assumptions resulted in lower and upper p-bounds at 0. 5 cumulative probability (i. e., at the median estimate) of 2. 9-12. 7 million. From here, replacing probability distributions with probability boxes, which represent classes of distributions, led to even wider bounds (2. 4-20. 0 million at 0. 5 cumulative probability). Even the 100th percentile of the uppermost bound produced (i. e., the absolutely most conservative scenario) did not encompass the well-known hyper-estimate of 30 million species of tropical arthropods. This supports the lower estimates made by several authors over the last two decades.",
keywords = "Host specificity, Model, Monte Carlo, Uncertainty",
author = "Hamilton, {Andrew J.} and Vojtech Novotn{\'y} and Waters, {Edward K.} and Yves Basset and Benke, {Kurt K.} and Grimbacher, {Peter S.} and Miller, {Scott E.} and Samuelson, {G. Allan} and Weiblen, {George D} and Yen, {Jian D.L.} and Stork, {Nigel E.}",
year = "2013",
month = "1",
day = "1",
doi = "10.1007/s00442-012-2434-5",
language = "English (US)",
volume = "171",
pages = "357--365",
journal = "Oecologia",
issn = "0029-8549",
publisher = "Springer Verlag",
number = "2",

}

TY - JOUR

T1 - Estimating global arthropod species richness

T2 - Refining probabilistic models using probability bounds analysis

AU - Hamilton, Andrew J.

AU - Novotný, Vojtech

AU - Waters, Edward K.

AU - Basset, Yves

AU - Benke, Kurt K.

AU - Grimbacher, Peter S.

AU - Miller, Scott E.

AU - Samuelson, G. Allan

AU - Weiblen, George D

AU - Yen, Jian D.L.

AU - Stork, Nigel E.

PY - 2013/1/1

Y1 - 2013/1/1

N2 - A key challenge in the estimation of tropical arthropod species richness is the appropriate management of the large uncertainties associated with any model. Such uncertainties had largely been ignored until recently, when we attempted to account for uncertainty associated with model variables, using Monte Carlo analysis. This model is restricted by various assumptions. Here, we use a technique known as probability bounds analysis to assess the influence of assumptions about (1) distributional form and (2) dependencies between variables, and to construct probability bounds around the original model prediction distribution. The original Monte Carlo model yielded a median estimate of 6. 1 million species, with a 90 % confidence interval of [3. 6, 11. 4]. Here we found that the probability bounds (p-bounds) surrounding this cumulative distribution were very broad, owing to uncertainties in distributional form and dependencies between variables. Replacing the implicit assumption of pure statistical independence between variables in the model with no dependency assumptions resulted in lower and upper p-bounds at 0. 5 cumulative probability (i. e., at the median estimate) of 2. 9-12. 7 million. From here, replacing probability distributions with probability boxes, which represent classes of distributions, led to even wider bounds (2. 4-20. 0 million at 0. 5 cumulative probability). Even the 100th percentile of the uppermost bound produced (i. e., the absolutely most conservative scenario) did not encompass the well-known hyper-estimate of 30 million species of tropical arthropods. This supports the lower estimates made by several authors over the last two decades.

AB - A key challenge in the estimation of tropical arthropod species richness is the appropriate management of the large uncertainties associated with any model. Such uncertainties had largely been ignored until recently, when we attempted to account for uncertainty associated with model variables, using Monte Carlo analysis. This model is restricted by various assumptions. Here, we use a technique known as probability bounds analysis to assess the influence of assumptions about (1) distributional form and (2) dependencies between variables, and to construct probability bounds around the original model prediction distribution. The original Monte Carlo model yielded a median estimate of 6. 1 million species, with a 90 % confidence interval of [3. 6, 11. 4]. Here we found that the probability bounds (p-bounds) surrounding this cumulative distribution were very broad, owing to uncertainties in distributional form and dependencies between variables. Replacing the implicit assumption of pure statistical independence between variables in the model with no dependency assumptions resulted in lower and upper p-bounds at 0. 5 cumulative probability (i. e., at the median estimate) of 2. 9-12. 7 million. From here, replacing probability distributions with probability boxes, which represent classes of distributions, led to even wider bounds (2. 4-20. 0 million at 0. 5 cumulative probability). Even the 100th percentile of the uppermost bound produced (i. e., the absolutely most conservative scenario) did not encompass the well-known hyper-estimate of 30 million species of tropical arthropods. This supports the lower estimates made by several authors over the last two decades.

KW - Host specificity

KW - Model

KW - Monte Carlo

KW - Uncertainty

UR - http://www.scopus.com/inward/record.url?scp=84872597729&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84872597729&partnerID=8YFLogxK

U2 - 10.1007/s00442-012-2434-5

DO - 10.1007/s00442-012-2434-5

M3 - Article

VL - 171

SP - 357

EP - 365

JO - Oecologia

JF - Oecologia

SN - 0029-8549

IS - 2

ER -