A Review of Driving Factors, Scenarios, and Topics in Urban Land Change Models

Round 1

Reviewer 1 Report

This manuscript reviews literature from 1945 to 2017 about land change models (LCMs). Specifically, the purpose, used driving factors and scenario story lines in the selected articles and chapters are analyzed. I can see that a lot of work has gone into this paper. However, the current framing of the paper makes it hard to see what the added value of this review is in relation to the current state of literature and for whom this review could be beneficial. In addition, the search terms used to select the articles for the review seem arbitrary and have not been justified. Please find my elaboration of these points below as main comments. Furthermore, I have a few minor comments, provided at the end.

Main comments

Timeliness of introduction: I appreciate that you are honest about the fact that the review was performed a while ago, in 2017. I find it up to the journal whether or not they consider this a problem. However, the introduction does not seem entirely up to date with the current state of literature either, which I do consider problematic. For example, in line 45-61 reviews of LCM methodologies and drivers are discussed, but they are from 2002, 2010 and 2014. More recent reviews include Verburg et al. 2019 (https://doi.org/10.1016/j.cosust.2019.05.002) and Tong & Feng 2020 (https://doi.org/10.1080/13658816.2019.1684499). As a second example, l. 30 states "but has been severely limited to only utilizing the single most likely prediction [3], a condition sometimes referred to the “predict and plan” approach [4]." The two references used here are 10 years old, and I don't believe this statement is true anymore. Approaches that use more than the single most likely 'prediction' are: model ensembles (e.g. used in climate modelling), stochastic models (e.g. used in hydrological modelling), Pareto fronts (optimization), a combination of Pareto fronts with simulation models, and Scenario Discovery. See for example: Steinmann et al. 2020 (https://doi.org/10.1016/j.techfore.2020.120052), Verstegen et al. 2017 (https://doi.org/10.1016/j.envsoft.2017.08.006), or Kim et al. 2020 (https://doi.org/10.1016/j.wace.2020.100269).

Motivation of the work: In the introduction, you motivate the use of land change models and scenarios, but you don't motivate your own aim, i.e. the review. Who would be interested in your results; modelers, policy makers, urban planners? And why would someone be interested in the differences in driving factors between case studies? Isn't it obvious that different case studies have different drivers? Don't see this comment as critique; I am not saying that nobody will be interested, but I am asking you to elaborate to your reader why this review would be useful and for whom.

Prediction vs. Projection: Whereas a prediction assumes that future changes in a system’s conditions will not influence the future system state, a projection specifically accounts for changes in the conditions. The sets of possible conditions are then captured in scenarios. As such, a weather model makes predictions but an LCM or climate model makes projections. See for example: https://sciencepolicy.colorado.edu/zine/archives/1-29/26/guest.html. Please use the two terms correctly and consistently throughout the text and/or provide your own definitions.

Search keywords: The search keywords were "land use change," "land covers," "future urban growth," "urban land change," "land use prediction," and "future urban expansion." The choice of search terms is not explained and seems arbitrary to me. For example, why do you search for "land use change", but not "land cover change" or "urban expansion" (without prediction)? Why don't you search for "land change model", the main topic of your research judging by the research questions? Along the same lines, why isn't the term scenario included in any of the search terms?

Minor comments
l. 40: "The use of LCM can be used to ...". Please correct grammar.
l. 49: Verb missing
l. 86: As the knowledge gap to justify your research you write here: "Despite the existing reviews of land-change based drivers, each routinely only focuses on a specific topic and/or within a specific context." Please make more explicit what you mean with specific topic and specific context. It is important that the knowledge gap is clearly defined.
l. 102: Why did you search for "land covers" plural? The term is typically used in the singular form.
l. 110: "During the full-text review, 17 additional articles were found so the total number of articles to review this research totaled 133." Please explain how you happened to find new articles during the review.
Section 3.1.1: How come that you have driving factors from empirical studies? You said that you searched driving factors in LCM studies (reported in section 3.1.2), so I am surprised to find a result section on empirical work. Please harmonize research questions, methods and results.
l. 266: "to prediction models". Prediction is not a verb. Please correct grammar.
l. 278: "combining different land change models (calibration methods)". Combining models, i.e. model coupling, is something different from model calibration, i.e. finding optimal parameter values of a model.
l. 314: Font size has changed here.
l. 428: A names is listed regarding contribution to software and validate. What software did you develop and what validation did you perform? I don't find anything about this in the methodology.
l. 434: Remove "Please add".

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

This paper provides an extensive literature review of the state-of-the-art in land change modeling. The authors analyzed a total of 133 studies in different disciplinary fields. The work is structured with an emphasis on the driving factors of urban growth, the topics in land change modeling and the use of scenarios.

This paper is well positioned to make a new contribution and valuable addition to the extant literature in this highly fuzzy and complex research area. It is also pretty well structured and well written. I think this can be accepted with some minor revisions:

• I would like to ask the authors to add some thoughts concerning the relevance of different land uses in section 3.1 (driving factors). It is both trivial and fundamental that specific land uses (such as housing, commercial or industrial uses) are linked to specific To give an example, distance to settlement or urban center is a plausible driver of residential development. However, this is not the case for all kinds of industrial facilities. It seems that the authors mainly focus on papers which addressed residential uses but the diversity of (urban) land uses in terms of natural or socio-economic drivers should at least be briefly mentioned.
• I wonder why the authors didn’t address the issue of land change modeling quality. I guess that there are at least a few papers which evaluate the model outcomes (predictions) in terms of accuracy. It could be interesting for the reader to learn about the improvement of predictive capacities of more recent generations of modeling approaches.

A minor point: Please try of avoid the line break in table 1 (second and sixth column).

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

it need to explain better, with more examples, what factors are used as driving forces behind urban land change in urban LCM

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Dear authors,

Thank you for answering to my comments. You managed to perform an impressive amount of additional work in a couple of days. Overall, I am happy with your responses to my comments. Two remain:

Motivation of the work: This is still not clear enough in my opinion. Your aim to "advance urban LCM projection and its scenario capability" is extremely broad, and it is not clear how exactly researchers and urban planners will benefit from your review. Can you make the aim more focused and be more explicit (and/or give examples) about the benefits for these groups? You do provide some examples in your response letter, but not in the paper.

Search keywords: You have now (partly) justified the used keywords to me in your response letter, but you did not include the justification in the method section of the paper. The search keywords are the main part of your methodology; the reader should be able to understand why these particular search terms were selected.
Furthermore, I see that you have changed a search term in the paper from "land use prediction" to "land use projection" (based on one of my other comments). Can you please confirm that you actually redid the whole search based on this change? I find that hard to believe because you say that you don't have access to the SCOPUS database anymore. What you write in the method section has to be in line with what you did; otherwise the research is not reproducible.

Author Response

Please see the attachment. Thanks!

Author Response File: Author Response.docx