Content of review 1, reviewed on November 07, 2025
General comments
I enjoyed reading the manuscript. It presents an extension of the time-to-detection occupancy models which can incorporate covariates. While this model extension makes total sense, the consequences of ignoring heterogeneity in observation and occupancy components are established and the solution (incorporating covariates into time-to-detection models) is not new either. I struggle to find the novelty of this methodological development. I think restructuring the model description helps address this issue. It would be helpful to present the model without the covariate and then present the new model with the covariate for an easier comparison.
Specific comments
1. Please consider an alternative title that captures the jist of the study. How about:
- Mixed gamma-exponential time-to-detection occupancy model with covariates
2. In the abstract, please include the simulation results about the amount of improvement in precision and bias when the three models were compared. It is not clear that you compared four models and quantified precision and bias.
3. The last three paragraphs of the introduction fit better in the methods. I suggest reducing it to at most one concise paragraph, closing the introduction. I am missing some information about the study system and how and why it is a good case, in the introduction.
4. The term “community parameter” needs to be defined at the first use
5. The introduction text largely ignores the literature about heterogeneity in occupancy and detection, in the time-to-event and presence/absence contexts.
6. L 107-109 Which simulation studies? Please cite the relevant study
Methods
7. In the methods, a clearer distinction between the existing model versus the new development would be helpful. You could use subheadings to separate the descriptions of the different models. The results section clearly compares three models and I think it could be clearer in the methods too.
8. A summary of the number of scenarios for the first and second simulation studies would be helpful. I suggest a similar structure for describing the simulations.
9. I am missing citations to original time-to-detection papers at the start of the methods
10. Part of the methods section fits better in the introduction (131-135).
11. L 140. I already suggested an alternative structure for the methods to describe specific models in their own sections. By reading the methods, I cannot tell which part of the model is for the first, second, or third model.
12. L 304 – 325 I am curious to know which simulation scenario this case study represented.
Results
13. I am missing a paragraph at the start of the results to summarize the number of scenarios, number of models, and the overall findings.
14. L 328 – 329 Was the point estimate mean? “average” could be more specific.
15. L 337 – 340 It would be helpful to include equations or citations about relative bias and precision in the methods
16. Fig 4. The results are compared between 4 models. Until this point, it is not clear how many models are compared.
17. L 341 – 345, 353-354, L 373 – 378 Please report the amount of improvement in terms of bias and precision.
Discussion
18. The first paragraph of the discussion could be focused on the main findings of this study, somewhere around L 505 – 512. The existing models could move to later in the discussion.
19. L 513 – 519 The main finding of this study could be closer to the start of the discussion. That said, it does not read as a novel finding.
Source
© 2025 the Reviewer.
References
Dinusha, P., Res, A., K., L. A. T., Wen-Han, H. 2026. A covariate-augmented mixed gamma-exponential time-to-detection occupancy model. Methods in Ecology and Evolution.
