Trouble estimating abundance and reading 'size' column

[Molly McDevitt]

Hello Distance Sampling Team, 

I am trying to estimate abundance and density for pronghorn in our study area. However, I am getting errors with functions ds(), dht() and dht2() when I include column 'size' in my data. 'Size' in this case, is the number of individuals counted in each group detected. I do not get errors when I run the functions without 'size' in the dataframe.

The data are from aerial systematic line transects (attached as dat.RDS) and my code, including errors I get, is simplified and attached as dsDensityEstHelp.r. 

I have previously been able to run ds() & dht2() without a problem. All said, I'm a bit embarrassed to ask this question here, and am hoping collective brains might help me troubleshoot!

Please let me know if I can provide further information and thank you, in advance, for your help!

Molly

[Eric Rexstad]

Molly

Thanks for you note with code and data. No need to be embarrassed about your question. I ran your code and encountered the same error. It seems to be related to an error with which we are already acquainted and hope to be able to fix shortly:

https://github.com/DistanceDevelopment/mrds/issues/79

Uniform detection function without adjustments giving error · Issue #79 · DistanceDevelopment/mrds data(book.tee.data) region <- book.tee.data$book.tee.region egdata <- book.tee.data$book.tee.dataframe samples <- book.tee.data$book.tee.samples obs <- book.tee.data$book.tee.obs # fit a half-norma... github.com

Your analysis does flag up a couple of other matters for you to consider:

• the estimates of density/abundance seem awkward (when they are produced)
• presumably because of a units problem. See the convert_units​ function in the package to help produce more meaningfully scaled estimates
• Abundance:
    Label     Estimate           se         cv          lcl          ucl       df
  1   20A 2.003822e-04 8.788461e-06 0.04385848 1.838212e-04 0.0002184353 304.0000
  2   35A 1.511085e-04 2.189446e-05 0.14489228 1.136239e-04 0.0002009593 133.0000
  3   35B 9.611589e-05 9.428838e-06 0.09809864 7.924424e-05 0.0001165796 193.0000
  4   53A 1.114966e-04 5.027526e-06 0.04509128 1.020057e-04 0.0001218706 170.0000
• As an approximation you could fit a half normal key function to your data. It will not produce a detection probability of 1.00, as the uniform does, but the estimated detection probability perhaps reasonable P=0.968, with a fit that is plausible (CvM P-value =0.928).
• 
• The AIC score for the HN model (18142.89) is close to the AIC score for the problematic UNIF model (18142.16), so I would think inference from this model would be valid.

I suspect if you tidy up your units (I'm guessing what units were used in your study)


convert_units(distance_units="meter", effort_units="kilometer", area_units="square kilometer")​

You should be good to go.

---

From: distance...@googlegroups.com <distance...@googlegroups.com> on behalf of Molly McDevitt <mollymc...@gmail.com>
Sent: 11 May 2023 22:06
To: distance-sampling <distance...@googlegroups.com>
Subject: [distance-sampling] Trouble estimating abundance and reading 'size' column

 

--
You received this message because you are subscribed to the Google Groups "distance-sampling" group.
To unsubscribe from this group and stop receiving emails from it, send an email to distance-sampl...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/distance-sampling/0bb95d77-97da-4b4b-b9b8-50031e2fcc1cn%40googlegroups.com.

[Eric Rexstad]

Apologies Molly, I hit the 'send' button too soon, before addressing your final question.

Your survey was designed with strata, so (presumably) you want stratum-specific estimates along with an estimate for the total study area. This is a geographic form of stratification. Using the ds​ function will produce these estimates for you, without the need to explicitly call dht​ or dht2​. It is only if you have used non-geographic stratification that you need to bring dht2​ into play.

---

From: distance...@googlegroups.com <distance...@googlegroups.com> on behalf of Molly McDevitt <mollymc...@gmail.com>
Sent: 11 May 2023 22:06
To: distance-sampling <distance...@googlegroups.com>
Subject: [distance-sampling] Trouble estimating abundance and reading 'size' column

 

[Molly McDevitt]

Hi Eric!

Thank you very, very much for the help/insight, and as you noted, addressing the unif distribution problem solved my problem. I'm sure I'll be back with more questions, and for now, back to the data!

Best,

Molly