Review Reports - Optimising Long-Range Agricultural Land Use Under Climate Uncertainty

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Please see document attached.

Comments for author File: Comments.pdf

Author Response

Please see response to the reviewer in cover letter.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The paper is overall well-written, and deals with an important topic. Before it can be considered for publication, a number of overarching issues need to be addressed. Below, you will also find comments on a few minor issues.

Overarching issues

Food production security: In the Abstract (line 2) you frame your paper as addressing food production security, and in your Introduction, you start with sustainable food production (lines 17-19). Your paper does not in fact deal with sustainable food production, but with the issue of how to ensure high net revenue under climate change at the level of a region in Australia. Non-food agricultural production (for fibre or energy) competes with food production in many locations, while higher revenue crops do not necessarily correspond to food needs for low-income global populations (see your own relevant remark lines 640-641). I recommend that you change the framing to better reflect the focus and objectives of your study, but keep your observations on the importance of wheat and legumes as staples globally in your Discussion (lines 706 and 600).

Economic arguments and methodology: your approach aims at economic optimisation. However, future dynamics and prices on global markets for individual export crops are extremely uncertain. These issues should be addressed in the introductory sections (with suitable references). In your methodology, I do not see any clear description of how you estimate future prices that underlie your estimates of net revenue. (These uncertainties of course also apply to other agricultural inputs, but I do not see a major problem with simplifying inputs to costs for irrigation water, provided that you are very clear about this in the framing and introduction).

Regional strategies versus farmer strategies: Although these sometimes align (and, to some extent, are interdependent), regional authorities and individual farmers operate under very different constraints, and also have different priorities. My understanding is that your paper is primarily addressed to regional authorities (and your methodology - 2.1. Optimisation Model and Solution Construction, footnote 1 - uses a unit that is described as appropriate for this purpose). This should be clarified from the outset. You can of course still argue in your Discussion (lines 635-636) that farmers may also find your model useful as a DSS, but to provide actionable options at farm level, a range of other considerations would need to be included, which lie outside the scope of this study. I would however include in your Limitations that to serve strategic planning for regional authorities, greater attention would need to be paid to the constraints and priorities of farmers in the region. Your model does also not necessarily maximise farm revenues (as suggested in your Abstract line 11), but operates at aggregate levels with other units.

Ecological Intensification of Agriculture: lines 94-96 you reference Hochman et al. 2013 who in their Abstract state that ”We define ecological intensification of agriculture (EIA) as: producing more food per unit resource use while minimising the impact of food production on the environment.” This objective diverges substantially from the optimisation goals of your own approach (which does not aim at food production) or consider environmental impacts besides immediate water use in the region. Hochman’s definition also differs from the definition that you provide lines 94-96, and though you in your own definition speak of “per unit resource input”, in fact you only appear to consider arable land and water. I recommend that you remove this concept and reference here, and instead clarify your own objectives. In your Limitations you could briefly discuss the implications of your own choice of objectives and scope with respect to the broader issue of EIA.

Crops and agricultural systems considered: At various points in your manuscript, you do clarify the scope of your study – for instance line 93 you mention broadacre cropping, line 128 you clarify that you have excluded perennial crops and only look at annual broadacre crop species. This should be clarified already both in the Abstract and framing in the Introduction. Speaking of food, agriculture and agricultural land use can additionally lead to some confusion, since agricultural land use can include livestock farming (grazing and cultivation of fodder). Also, if I understand correctly, agroforestry is not practiced in your case region, which is also a consideration, since it will affect not only shade and soil temperatures, but also wind and evapotranspiration (as well as soil microbiota and ecosystems with implications for agriculture more generally).

Uncertainties in climate projections: lines 746-747 you speak of identifying which climate model is the likely future reality. Increased weather variability under climate change is making it increasingly difficult even to make short-term forecasts, let alone projections that will apply to the end of the century. Extreme weather events are increasing, and numerous locations experience both drought and flooding in the same year. I believe that the using conditions for different decades in your modelling is useful because it illustrates a wider range of potential climate conditions (and some of their implications for the set of crops that you have considered), but with non-linear dynamics no existing climate model can claim to reliably project end-of-century conditions. An important consequence of unpredictable climate change (as well as of greater weather variability and extremes) is that there is a trade-off between optimisation (of yields or revenue), and resilience. This point should be developed in your Discussion, as well as mentioned in your Limitations as an issue that cannot be addressed with conventional optimisation approaches. One approach to increase resilience and avoid major economic losses, is to avoid broadacre cropping altogether, and plant diverse crops so that crop failure for one crop will still leave revenue from other crops (and reduce the risk in monocultures that large cultivation areas are affected by a pest).

Diversity and crop rotation: Both these terms can refer to very wide ranges of practices. Please carefully review your manuscript to clarify what exactly you are referring to. The lack of clarity also leads to some confusion in understanding your arguments. For instance, you speak of allelopathic relationships between crops (line 69), and also about pests and disease (line 75), but depending both on cropping system and specific crops, positive and negative plant-plant interactions will differ (but not have major impacts for broadacre monocultures, whereas transmission of pests and disease will differ both depending on specific crops and specific pests, and on the type of rotation.

Other comments

Several statements, particularly in the Introduction, are formulated in the form of almost telegraphic and categorical statements that are overly generalising, and also do not always clearly express what your own arguments are.

Lines 24-25 – Australia’s trade balance is dependent on foreign currency (to pay for imports), but 72% of agricultural production is exported (making the sector vulnerable to fluctuations in demand and prices on global markets). Please rephrase to clarify your argument. (also, it would be relevant to know the percentage of annual broadacre crops discussed in your paper that are produced for export).

Lines 28-37 – for international readers who are likely not familiar with these documents, please briefly describe the scenarios outlined in the report. Please also briefly describe the impacts, solutions and strategies that you are referring to.

Lines 39-40: Conditions, policies, priorities and constraints vary greatly between contexts, so there is no one-size-fits-all approach to modelling that might serve all contexts and all purposes. You do not need to make this kind of broad generalisation in order to argue that your own model can be useful. Please reword or delete.

Lines 40-41: I understand that the modelling that you use in your study draws on this particular body of work. Please reword to clarify that this is your intended meaning, and provide some indication of context, time period and purpose of the body of work in your references 9-14.

Line 44: This spatio-temporal … - please reword to The spatio-temporal … (STALS) that is proposed in the present study, is a novel multi-objective …

Lines 47 ff: I would expect more detail on your case study region here: temperatures, precipitation, agro-climatic zone, brief description of agriculture (including crops that are not covered in your own study, number/size and type of farms, a bit about the common crop rotation systems that are currently used, proportions of irrigated and rainfed land, etc).

Lines 49-50: California, Spain, Iran and Sub-Saharan Africa are huge territories that encompass a multitude of climate types, not just arid and semi-arid. Please reword. (also, with respect to agriculture it might be more relevant to consider agro-climatic zones, while for water resources (and conflicting uses for water) I would suggest introducing the notion of water stress, since the issue is not just climate but water available for agriculture.

Lines 51-54: please review the sentence structure, where the verb supplies does not correspond to all the items in your list. You do not consider any of these other demands on water resources directly in your study (nor use a nexus approach, including energy demands of the agricultural sector for instance), but these conflicting demands presumably underlie the prioritisation of water allocation directed by legislation that you mention line 92. I therefore suggest that you move the details on non-agricultural demands on water resources to the part where you mention the legislation.

Line 56: which are soil … - please reword to: which include soil … (it is understandable that your modelling does not cover every aspect, but to meaningfully interpret your findings and to be able to assess the relevance of your model for various purposes, it is not helpful to here suggest that this is comprehensive).

Line 60: please reword to: … reviews work that is relevant to the present study, in the areas of ….

Line 69: The purpose of diversifying … - do you mean that your study aims to facilitate diversification in order to reduce biotic stress …., or that: Among the reasons for diversifying land use activities that can be found in the literature are reducing biotic stress …. ? Please clarify. Also, please also spell out how the question of diversifying land use informs your own study. Diversifying “land use” is even more general and covers a much wider range of practices than diversification of land allocated to the set of annual broadacre crops considered in your study, and the subset of crop rotations that your model generates as options, so you do need to spell out and clarify your arguments here.

Lines 72-74; Rule-based approaches … - I believe this question is quite important for your study, but you need both to develop the arguments presented in the body of literature that you reference here, and to clarify how your own work draws on and makes use of this approach.

Line 92: please provide the name(s) of applicable legislation and a very brief description (besides the consequence that annual irrigation water availability for agriculture varies considerably). Also, I suspect that this paragraph is in fact connected to the preceding paragraph on regulation and access entitlement, which could be clarified in the wording, and the two paragraphs merged.

Lines 94-96: see my earlier comments under Overall issues.

Line 98: the statement that the application of optimisation approaches to agriculture “is beneficial” is debatable (see my earlier comments). Please reword to: Although it has been argued that the application ….

Line 113: please clarify that you are here referring to the Murrumbidgee Irrigation Area

Line 119: your mention of soil health here creates the expectation in the reader that this is something that you will address in your own modelling – I do not immediately see that you have done so, and if I have missed this when reading the manuscript this is a point that you may need to spell out more clearly and explicitly. (you do speak of the role of legumes – presumably as preferable to purely chemical fertilisers in supplying N? – but if this is the only dimension of soil health that you consider, you should say so from the outset).

Line 122: I do not immediately understand what you mean by “preclude land use” – please clarify

Line 129: clarify what you mean by “production capacity” here, (which indicators and measurement units do you use to operationalise the concept?)

Line 134: I do not believe that this question can be answered (see my earlier comments), but the differing conditions in the longer term projection do provide a relatively wide range of examples of possible conditions that agriculture in the case region might be operating under, which is in itself useful, provided that you in your Discussion and Limitations clarify that the projections that you use here do not cover the type of challenges that increased variability and extremes are already now starting to present (your typology of water scenarios in Table 2 does not include the mixtures of drought and flooding or short extreme spells of precipitation that are contributing to crop failures today).

Line 161: similar to much of the past

Lines 222-225: soil moisture and water held in soils (and at which depth) will depend on issues such as soil type and how much organic matter it contains, compaction, crusting, topography, roughness and vegetation covering soil, as well as the duration of rain and how heavy it is (water can be lost through runoff, while in other cases the formation of pools can lead to rot or affect the distribution of seeds and seedlings during planting).

Line 235 – the only input cost you consider is water (see earlier comments), and if the objective is to generate crop rotations that are profitable overall, it is not entirely clear if this exclusion applies to each year or to each planted crop, or to the entire ten-year period.

Line 256-257: this description does not adequately list your sources and materials for the present case study. If this is a more general statement on the types of data that the current version of the model allows, this should be more clearly expressed in the wording.

Line 274: several

Lines 285-388: I am not entirely sure if I understand your argument here – please clarify and spell out your reasoning.

Line 722: your scenarios and assumptions are at the pessimistic end of the spectrum among those presented in the CSIRO report (if I understand you correctly), which you should clarify in the wording of this sentence (since far more pessimistic scenarios and assumptions exist).

Lines 776-779: These assumptions should be presented in your Methods section, not here.

Lines 810-811 and 816-817: please remove the wording that you have copied from the journal’s instruction for authors

References:

Some of your references use capitals for each word in the publication titles – please revise for consistency.

Your reference (1) – I do not think that this is useful for your study.

Some of your references are rather old and could be replaced or supplemented with more recent publications. I believe, for instance, that you could find better references on the topic of plant-plant interactions and crop rotation than Bullock 1992.

As mentioned in my earlier comments, I think that you can make better use of your reference (34) in particular.

Comments on the Quality of English Language

Author Response

Please see response to the reviewer in cover letter.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The paper presents a study on optimising long-range agricultural land use under climate uncertainty.

Abstract: This section contains all the necessary elements required to describe the design and results of a study.

Introduction: In this section the authors described in great detail the study’s background and the study questions. The authors clearly demonstrated the research gaps they wanted to fill in.

Materials and Methods: In this section the authors clearly presented their approach and data used in the study. The methods were correctly chosen.

Results: The results are compelling and clearly presented. I suggest renaming the sub-sections 3.3.1-3 as it is not instantly clear that they relate to specific periods of time. In my opinion they should not be distinguished and should be presented under the sub-section 3.3 as part of it.

Discussion: In this section the authors made an in-depth analysis of the study results. They also presented the study limitations and future study pathways.

Conclusions: In this section the authors briefly presented the study summary.

Author Response

Please see response to the reviewer in cover letter.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Many thanks for this revised version of the manuscript, which is substantially improved and clarified.

Before publication, kindly attend to the following minor comments and suggested corrections:

line 42 - Scenario 4

line 57 - Work on previous temporal prediction models that has been foundational to the present study can be found in (10-13):

line 75 those authors' - which authors are you referring to here?

line 84 All of this previously mentioned research - please reword for clarity

line 85 This paper explicitly allocates - In this study, we will instead explicitly allocate

line 94 - if I remember correctly, the Supplementary Material is only available on request. It would therefore be helpful to briefly summarise the economic assumptions that were made.

line 123 - the comma after Sub-Saharan Africa here can lead to a generalising interpretation, please delete or reword to: ... Cape region, both in Sub-Saharan Africa

lines 165-166 now - please reword for clarity

lines 213-214 most work - overly generalising, reword to "much" and give an example of what you are referring to here

lines 323-327 - see my comments to the previous version of the manuscript on this oversimplification. Since water is the main consideration in your study and for the case region, please add some points on this aspect in your Limitations.

line 371 current responses - the word current could be read as your own assessment of responses (and their likely future development) at the time of writing, or as a summary of the arguments and findings presented by Schwalm et al. (your reference 77). I agree that RCP8.5 is better, but the argument needs to be developed and wording clarified (particularly since policy moves back and forth, with a direction that is even more uncertain at the moment of writing, and due to the earth systems feedback mechanisms). It should also be noted that Schwalm et al. are only considering that RCP8.5 will remain useful to reflect impacts of policies from 2020 to mid-century, and that they underline additional uncertainties for any period beyond that.

line 419 and Table 3 - please more explicitly address the issues with this categorisation (as well as the impacts on income estimated based on historical experience) in your Limitations (see my comments to the previous version of your manuscript)

line 727 mimics to the 2020 land use reality - please reword to more clearly and accurately express your intended meaning

line 789 - please add comma: Like all DSS's,

line 841 pessimistic - unfortunately, I do not believe that RCP8.5 can be qualified as pessimistic (see my comment to line 371). I would recommend that you here instead simply state that you have used RCP8.5 projections.

lines 860 ff (subsection 4.3.1) - I believe that you should briefly mention issues such as soil health, and that you have not considered wider environmental impacts, or the complexity of interactions between different types of land use and agricultural management systems and practices (adding a few references). Please also see earlier comments above on limitations.