| 24 Apr 2026
Serious gaming as climate service co-production: developing S2S forecasts for the Norwegian agricultural sector
Abstract. Subseasonal to seasonal (S2S) forecasts offer valuable opportunities for the agricultural sector by bridging the gap between short-term weather forecasts and long-term climate projections. However, their adoption faces significant challenges, including high uncertainty, coarse spatial and temporal resolution, and difficulties in communicating probabilistic information effectively. Establishing a meaningful dialogue between forecast producers and users is crucial to ensure these forecasts address user needs. This study explores the use of a serious game methodology to facilitate such dialogue and guide the development of S2S forecast services tailored to the Norwegian agricultural sector. Through game sessions involving farmers and agricultural advisors, we identified strategies users employ to interpret S2S forecasts and uncovered diverse use approaches for these products. Our findings highlight key recommendations for service development, including ensuring seamless integration of forecasts across different lead times, providing narrative explanations of probabilistic data, supporting user familiarization with the tools, and exploring the use of recent years' conditions as reference points instead of historical averages. The serious game approach proved effective in fostering meaningful dialogue and mutual learning among agricultural users, forecasters, and service developers, ultimately supporting forecast services' co-production.
The manuscript entitled “Serious gaming as climate service co-production: developing S2S forecasts for the Norwegian agricultural sector” focuses on the development and implementation of a serious game called “Melkeværet”. This work is clearly relevant for the readership of the journal. However, I see a number of points that the authors should address.
First important formal point: I noticed that part of the work seems to have been published as a technical report in 2024 by the same group of authors (Title: “Melkeværet. A serious game to bridge climate service providers and agricultural users”), but this finds no mention in the reference list or the acknowledgements. Some of the article’s figures are identical to the ones presented in the technical report; some parts of the reported methodology and results are clearly an earlier version of this manuscript. Publishing first results as a report is common practice as is later writing a much more elaborate article about the same work. But in any case, some kind of reference towards this earlier publication is necessary.
General feedback:
I thought about which audience(s) your article might address and I am not sure if you had the same ones in mind, but I came up with 2 groups: (1) People interested in learning how to communicate specific weather forecasts (here S2S forecasts) to specific groups (here: famers), and (2) people interested in the communication of climate-related topics through serious games. I have to admit that I rather belong to the second group. I assume that group 1 readers can already gain many insights on their questions and I think that this might have been your prioritised audience. However, since your title starts with “serious gaming…” and you report on a really well-designed game, I think it would help to introduce some more aspects of your methodology and evaluation, in order to make your interesting work more accessible for this readership. Some aspects already appear in the text, but I suggest a little restructuring to make them more visible.
Unfortunately, the methodology is not explained transparently. The game evaluation seems to be structured in a somewhat mixed-method approach: a partly quantitative questionnaire with open answers plus observation by researchers. However, the questionnaire is shared, but the answers on the closed questions are not reported. Maybe answers of the only open-ended question (last question of second questionnaire) have been used for the results, but this is not clearly communicated. Furthermore, the authors do not state how the researchers that were part of the game sessions proceeded. Did they carry out an observation? Did they use any kind of protocol for this? The methods applied should be described in section 2, the results in section 3 and the interpretation in section 4. It seems like often results and interpretation are mixed and it is unclear from which steps the authors derived these findings and their interpretation. Sometimes interpretation is given in the results section and sometimes suddenly new results are being introduced in the discussion section (I’ll comment some examples below). As far as I see, the journal does not require a division of results and discussion, so you could consider having a combined Results & Discussion (section 3) before a Conclusion (section 4). If you decide to keep your current structure, please make sure to keep both parts better separated.
Minor aspects per section:
Abstract:
The abstract gives a good overview of your paper. It would help to add the serious game’s name Melkeværet into the text directly, so that anyone interested in finding out more about the specific game knows which game you are focusing on.
1 Introduction
The structure of this section into 2 subsections makes absolute sense. You could consider dividing 1.2 into 2 sections: the last 2 paragraphs give an overview on your approach and the aim of this research; therefore, you could simply introduce a new subheading after l. 85.
l.40ff: The use type categorization by Ray and Webb was not known to me before. From reading this 1 sentence it does not become clear. These 4 use types do not seem to be so easily separable, I would assume a big overlap. It is also not clear if people are meant to belong to only one of these use type groups, which would probably not make sense. If you really want to use this theory/concept in your work, please explain it better. (See also comment in section 4 on this)
l.60-75: In this subchapter some terms are named but not always clearly defined. The topic of “co-production” (later in the text “co-development”?) is mentioned, but it does not come across if this framework has been fully applied or not in this research. Serious games are defined in l.67-68; it might help to give a reference for this definition. In l.69 and multiple times later in the text the authors write about “gaming simulations”, which is confusing. It is not clear if the authors understand “gaming simulation” as a part of “serious gaming” or if this another concept. I personally would find it easier to follow if you stick to one term throughout the whole article (preferably serious game).
l.85: As said above, it might help readability to start a new section after this called “Aim/focus of this research” or something like that.
l.100-105: This is a long preview of what is to come in the next chapter. This is probably an individual question of style; I think it could be shortened. Especially since l.107-109 also have a little preview of what is to come in the chapter, which is a bit repetitive.
2 Methodology
This section is quite long and has many subchapters and sub-subchapters, which is a bit confusing. There is some repetition. Through some restructuring I think it would be easier to grasp. Most readers probably expect to read about (1) when, (2) where, with which (3) sample (participants), and with which (4) methods research was carried out. The first three questions are only addressed marginally often in text parts which have another focus. It would help if you started with a concise section 2.1 which could be called “Sample and setting” and answer these 3 central questions.
The question on methods applied is a more complex one. The authors describe the design process in much detail (however sometimes only the results of some of the steps, not the questions behind it), but it is unclear which methods were applied to explore and analyse this process (see general point above on this).
Chapter 2.1: The figure is very helpful in understanding the process, but it looks strange that there is just one sentence mentioning but not describing it. In the following chapters (2.1.1 etc.) it seems like the outcome of these steps are described, but not always the procedure of the pertinent step. Fig. 1 suggests that in the Game scenario definition (please keep the same wording in the subheadings/text) agricultural experts were involved. In 2.1.1 however, this is not mentioned. Make sure to share who actually worked on the scenario definition (where, when, who).
For readability it would help if firstly (2.2?), the design process is described step by step (following fig. 1) and after that (2.3?) the actual serious game is described step by step.
l.118: The summer 2018 example is given but not explained. “Still engraved in Norwegian farmer’s memories” does not tell the reader (who might not be a Norwegian farmer) what actually happened. Replace this with a short mention of specific impacts of this drought.
l.129-131 (and also following figure titles): Please shorten titles and especially do not introduce new information here, which is otherwise not mentioned in the text.
l.134-135: It is unclear what Yr or YR means (same thing with different writing?), this abbreviation is not introduced.
l.157: The mention of the questionnaire is confusing. As said above, please explain methods and materials used in the beginning. Furthermore, there is a strong repetition in chapter 2.2.3. Do not report on the same thing twice.
Chapter 2.2.1 could serve as an opening section of chapter 2, if you restructure a bit.
Figure 4 can be deleted as it does not share much relevant information. Furthermore (as said before) it is a bit difficult if the authors share relevant information only in the figure title, here: “Please note that in almost all cases, agricultural advisors and representatives are also farmers themselves.” This proves the figure as somehow misleading, because this double role (being a farmer and also an agricultural advisor/representative) is not displayed properly. If 45 out of 60 participants of the game sessions are farmers and some of them also had another role, please write this in the text. The figure takes much room and shares little information. Furthermore, the authors shared no figure and also no information in the text on the participants of the game scenario definition step, which would have been just as relevant for the article.
3 Results
Throughout the text the terminology is not concise: please stick to “participants” if you refer to the participants (at the moment you jump between farmers, farmer participants, and participants: if it is important to mention that especially the group of farmers among the participants did or said something differently to the other groups, make this clear).
As said above, the methodology is not clear and most findings are reported without any mention of the underlying data, meaning: no quotes are given to explain the formulated findings. Please include quotes and also make a note on the source of the pertinent quote (e.g. written answer from questionnaire or statement from the gaming session).
One general thought: even though the authors understand their study as a co-development process, the results reported only circle around the last step of this process, the gaming session.
Figure 8: This seems to be part of the method/material section, not a result?
l.326-329: This is a bit repetitive to the text before. Please shorten.
3.4: This is a very interesting and relevant subsection. It might also be a good start for the results section.
4 Discussion
In my view, especially the first 2 pages (p.16+17) there are some results shared, that were not presented before. Also as said above, it is a bit tricky sometimes to divide the presentation and interpretation of findings, therefore it would be a possibility to have a combined results & discussion section. If not, these 2 existing sections should be clearly separated.
The paragraph starting in l.346 on Ray & Webb’s theory are confusing. The authors write on “Confirming Ray and Webb’s findings […]”, but in the text and also in Table 1 they did not confirm the whole concept, as it seems? I think it is great if researchers use (established) theories/models/concepts in their work, but in this case, I do not find it fitting. Either the theory and the findings are explained better or this reference is taken out.
l.350-351: The mention on “situational awareness” in quotation marks is a bit confusing. Maybe restructure to something like “…, farmers may want to gain situational awareness (…) by consulting …”, if this is the intended message of this sentence.
Chapter 4.3 is very interesting and the authors could position it earlier.
l.439: It is unclear for which statements these 2 sources (Flood et al., 2018; Vigna et al., 2024) stand, the sentence before seems to be about the game itself?
Appendix A – Questionnaire: If none of the results of these questions are given in the text this questionnaire is not necessary. If maybe only results of the open-ended last question are given, it is enough to give this question in section 2.
l.550: There is no statement on data availability here. Comment on the questionnaire data and the observation notes here, even if these data are not shared in a repository or elsewhere.