GC Insights: Choosing the right map projection, and what it means for scale bars and north arrows

Royé, Dominic

doi:10.5194/egusphere-2026-4209

Preprints

https://doi.org/10.5194/egusphere-2026-4209

Preprints

27 Aug 2026

| 27 Aug 2026

Status: this preprint is open for discussion and under review for Geoscience Communication (GC).

GC Insights: Choosing the right map projection, and what it means for scale bars and north arrows

Dominic Royé

Abstract. Map projection choice is usually left to software defaults, yet it determines whether spatial patterns are shown faithfully or systematically distorted. Because every projection trades area, shape, distance, and direction differently, the validity of two common map elements, the scale bar and the north arrow, depends entirely on the projection chosen. A survey of 299 maps drawn at random from a frame of 27 geoscience journals finds that 1.7 % name their projection, 44 % give no way to determine it, and only 16 % draw a graticule, the one device that shows the geometry rather than approximating it. This Insight argues that projection should be a deliberate, purpose-driven choice, and that journals should require authors to declare the coordinate reference system in figure captions.

Received: 14 Jul 2026 – Discussion started: 27 Aug 2026

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Preprint (PDF, 456 KB)

Supplement (304 KB)

Download & links

Dominic Royé

Status: open (until 22 Oct 2026)

Post a comment Subscribe to comment alert

RC1:
'Comment on egusphere-2026-4209', Fabio Crameri, 09 Sep 2026 reply
This manuscript outlines the importance and the (wide) misusage of cartographic projections (in particular relating to the lack of transparency of the chosen map projection), explains the different use cases for certain projections, and provides the author’s perspective on an educated way forward. The paper is written and illustrated well, and the supplementary provides the necessary information (including the R code) to reproduce the study. I recommend publication and it seems like a great fit for GC too.
To clarify the approach, the main text might state how the samples were judged, whether by hand/eye or somehow hard-coded.

Is there a reason why all the global maps omit showing the most southern latitudes? It makes them feel incomplete.

Is it worth mentioning, or better, discussing that the Equal Earth projection has just been voted for on a resolution by the U.N. General Assembly as the preferred map projection? This could strengthen your important point that agreeing on one map projection (out of many wrong but, depending on the cause, useful projections) misses again the point of how important its use case is for choosing any, and this, map projection.

Below are no critical comments, but just would like to raise the following points to the discussion (even though they might go beyond what a reviewer is here to judge). I like how short and punchy the manuscript currently is, so I suggest thinking carefully whether you actually want to include any of my following suggestions:
Currently, this seems to be directed at and provide a clear way forward towards journals/journal editors. I wonder if it would be useful to provide some simple and clear “to-dos” for the authors/creators of the graphics too. Maybe with one or two “good examples” that simultaneously also clarify the different elements. For example, there is neither a distance scale, nor a North arrow, nor the Tissot indicatrix, shown here, and I assume not everyone knows what is meant by that. If you can add this, at least to the supplement, I think that would be helpful for the understanding of the paper’s messages.

I agree that there are direct parallels to how people use colour scales: it might be worth mentioning the underlying cause of all these problems too (and not just the reactive measures): Most academics are never educated in graphic design or even data visualisation, even though it is a truly fundamental, likely the most common methodology across all science disciplines. I have outlined that a couple of times already myself (e.g., Crameri, F. and S. Hason (2024), Navigating color integrity in data visualization, Patterns, 5(5), https://doi.org/10.1016/j.patter.2024.100972), but it would be worth reiterating, I think.

That’s a huge problem, in my opinion, and it keeps being neglected. I mean, how ridiculous is it to have to write scientific papers about how to use a proper colour scale or a map projection; two fundamentals. If you think about it, these are primary school level discussions, really.

Having stated that, as long as we won’t offer University-level mandatory classes on academic graphic design, manuscripts like this are (and will remain) fundamental to fill the (most basic) educational gaps.

Thanks for the effort and a wonderful read!
Fabio Crameri

09.09.2026

Reply
Citation: https://doi.org/10.5194/egusphere-2026-4209-RC1
- AC1: 'Reply on RC1', Dominic Royé, 15 Sep 2026 reply
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2026/egusphere-2026-4209/egusphere-2026-4209-AC1-supplement.pdf
  Reply
  
  Citation: https://doi.org/10.5194/egusphere-2026-4209-AC1
RC2:
'Comment on egusphere-2026-4209', Fritz Kessler, 13 Sep 2026 reply
The following is a review of "Choosing the right map projection, and what it means for scale bars and north arrows." The paper discusses an interesting conundrum involving the map projection - whether map projections are named on maps accompanying journal articles in the geosciences. The author argues that the reader should be informed of the map projection as the projection choice has important implications for the accuracy of the scale bar and north arrow. Moreover, the inclusions of the scale bar and north arrow help communicate important spatial geometries and relationships to the reader when using the map for specific purposes. The author addresses this conundrum by examining 27 geoscience journals paying attention to the maps that appeared in these journals specifically noting whether the map projection name was mentioned and reporting the presence/absence of the graticule.

Overall, the call for authors of scientific journals to include map metadata, such as the map projection name, class, property, etc. is well intended, at least from a scientific standpoint. Knowing the name of the projection (or at least seeing the graticule's appearance) is a reasonable way to help the reader better determine the underlying structure of the map (via the map projection) and what properties and characteristics that structure possesses. These fundamental criteria are important for the map reader to successfully carry out specific tasks using the map. The present effort sets a good foundation but lacks much in the way of supporting structure. However, I offer ways that the author could substantially revise the present work and make a meaningful contribution to help future authors add important map projection information to their maps.

General Comments
- The article title suggests a focus on the choice of the right map projection and its impact on scale bars and north arrows. However, the abstract doesn't necessarily address the title's focus. In fact, the abstract somewhat contradicts the reader's understanding of the article's scope and purpose (as inferred by the title). The main point here is that the results that are reported in the abstract say nothing about the map projection choice, the scale bars, or north arrows that were included in the study. Instead, the results section mentions whether the map projection was named and the presence/absence of the graticule. Either the abstract needs to be refined so that the results directly reflects back on the title or the title should be revised to better conform to what was mentioned in the abstract.

- The discussion under the "why projections matter" does not get to the crux of the research problem in a succinct and purposeful manner. This discussion includes topics such as criticizing the Mercator projection, use of geographic coordinates, projection properties, geographic scale, and Tissot's indicatrix. The linkage between this discussion and the choice of a map projection, the reporting of the map projection name, and the accuracy and understanding of scale bars and north arrows included on maps in peer-reviewed journals is not clearly stated. If these ideas are in fact important to the stated title of this paper, then this introductory section needs to better address the scope and purpose of the title and abstract.

- The research methodology needs refinement. Specifically, I note the following issues with the sampling of the journals, what is reported, and the general conclusions.
There needs to be a rational for why the 2016-2025 time period was selected. Why start at 2016? This start year seems arbitrary.

It isn't clear what the author means by "what a reader can see on the page." Does this mean the entirety of the text and map on any given page or is this statement more limited to what was shown on the map or included in the figure caption? To me, any of these page locations would suffice as an opportunity to report the map projection name.

I didn't see or was directed to where the "Supplement" was located.

What does the author consider as the projection name and how were the 1.7% of the identified map projections named in the 299 articles? For example, did the author consider the "UTM" as sufficient for a projection name? In Figure 1, the author uses "LaEA' as a projection name. However, I suspect that some readers unfamiliar with map projection would not be able to connect "LaEA" with the Lambert azimuthal equal area projection. Were any of the map projection names abbreviated? The author also includes the EPSG id code. Again, for readers that are not familiar with EPSG, did any of the named map projections include the EPSG code. There should be a more complete accounting of what examples were learned from the sampling effort.

The author reports on the percentage of local-maps (68%) and global maps (3%) that included a scale bar. No comparable reporting of the north arrow finding is reported in this section. The author should also report the total number of maps that are local vs global in geographic scale and define what exactly is a local geographic scale. However, assuming there was a scale threshold used in this context, I am puzzled over what constituted a map scale intermediate to local and global (e.g., a map showing a country)?

For those named projections, what percentage was the Mercator (since this projection was specifically called out on line 33), or any other named projection from the maps sampled?

Overall, the writing contains many errors, is overly informal, and at time difficult to understand.

Ultimately, the author leaves the reader adrift in that aside from suggesting that journal articles require authors to name the map projection (to which I totally agree), there is no guidance on what to do with the scale bar or north arrow. The author states that "neither the bar nor the arrow should be included or omitted, by default." However, this doesn't help the reader or reflect back on the overall stated focus of the article.

- I suggest that the article would be greatly improved if the author would provide guidance on the following matters.
Provide specific guidance on how authors should state the map projection name. I propose that the name should follow this example when possible: Common name, class, and property (e.g., Mercator, cylindrical. conformal). For those grid systems such as the UTM, the naming convention should be "UTM, zone number, and north or south." The author could also suggest that, when possible, to include the EPSG code, but realize that not all map projections have an associated EPSG code. I would definitely avoid suggesting that the map projection name follow any proprietary software naming conventions (e.g., Esri).

Provide specific guidance on how scale bars should be included, but also statements that should be made regarding their accuracy. Essentially, there should be a statement about the location of the standard lines along which the scale is valid.

For all maps, the graticule and the corresponding DMS or DD values should be included along the margins to help the reader better understand the spatial contextualization of the map.

For north arrows, their inclusion is a bit more tricky as their alignment with the meridians complicates their association with "north" on the map. Perhaps including several north arrows aligned to the meridians east to west would be a good recommendation, along with a short statement indicating that "north" follows the alignment of the meridians.

Specific Comments
Abstract, line 7: the author suggests that map projections possesses the property of shape without qualifying what "shape" means in this context. Far too often, authors use the idea of shape without providing the context of its use. Map projections do not preserve shape, it is mathematically impossible to quantify a metric that can be used to preserve, for example, the shape of Iceland. Too often authors conflate "shape" with "angles" - the two ideas could not be further apart in the vernacular use of "shape."

Line 18, need a citation to support the claim about the neglected issue of projection choice in scientific cartography.

Line 28, a better wording would be "map actually informs the read which projection..."

Line 30, again, the use of the word "shape" is being misconstrued in this statement. The author needs to remove this word and replace it with "angles," which is the more mathematically correct usage related to the map projection properties.

Line 30, a better wording would be "each projection is a compromise of distortion" as some projections preserve more than one property.

Line 33, a better wording would be "preserves local angles, a property essential..." Shape is not essential for navigational activities.

Line 43, equal area is not just limited to global maps that show a spatial distribution of some phenomenon.

Line 45, again, the use of the word "shape" takes on a different meaning here but should be contextualized.

Line 46, I am not sure why "European statistical cartography" emphasizes statistical. Isn't statistical cartography practiced in other countries?

Line 48, the use of shape, again?

Line 53, I don't see the indicatrix appearing on the maps in Figure 1 (I don't follow what the Fig. S1 means - what does the "S" mean here?)

Line 63, it isn't clear to me what the author meant with the use of "by domain," domain of what exactly?

Line 64, "whether the coordinate system is reported." It is also inconsistent to switch between coordinate system and map projection as these phrases have different contextual meanings that are not interchangeable.

Line 66. I don't understand the use of 44% and then parenthetically reference (38-50). To what is this 38-50 referring?

Line 67, The author states that "no coordinate system is visible" to which I assume means the graticule isn't displayed. However, I could be wrong in my assumption, be consistent with the use of graticule.

Line 68, why is the focus suddenly change to distance and direction?

Line 69, the "local scale" needs to be defined.

Line 70, the author states "Global maps are, in turn, where the projection is least often chosen with any care." I completely disagree here, a citation is needed to support this claim.

Line 70, the author should be consistent with the use the graticule instead of using "coordinate lines" to also refer to the graticule.

Line 73, the statement "A map centred [centered] on a pole cannot make that mistake, since plate carrée stretches the pole itself into a line" doesn't make sense. Unless the author is talking about a polar centered plate carrée? The rest of the sentence is similarly illogical "such maps are necessarily projected."

Line 77, I don't understand what "perfectly equidistant" means, what woudl be "partially" equidistant in the projection nomenclature?

Line 77, this statement is illogical "the bar is accurate where it is placed." Scale bars are usually placed along the margins of a map.

Line 78, the statement "In the Mercator projection, the linear scale at 60° latitude is twice that at the equator, so a bar placed at the bottom of a world map is correct near the equator and wrong by 100 % in Scandinavia" is not correct. A scale bar appearing on the Mercator projection usually represents that scale along (not near) the equator is 100% accurate. Based on my calculations, the north-south scale distortion on the Mercator projection for Scandinavia ranges from 77% to 150%.

Line 82, this phrase doesn't make sense "not the absurd bar," what is absurd about a bar?

Line 83, I strongly disagree with the author's assertion that "This is a projection problem, not a design problem." In fact, the projection should not be faulted for this, but the map maker should be regarding their design decisions (shouldn't the map maker be stating what the map projection is, include the graticule, scale bar, and north arrow?).

Line 89, the author states "conic projections meridians converge, so north is vertical only along the central meridian." While meridians do converge to a pole on conic projections, they are straight vertical lines; so, the context of this statement needs some clarification.

Line 91, I have no idea what the following phrase means "This rarely bites in the surveyed literature."

Line 91/92: the author states "three-quarters of the world maps that draw any coordinate lines draw them as a straight, rectangular grid, meridians vertical by construction." However, where is the proof? A citation is needed to support this claim.

Line 92, the author states "Where meridians are straight, the arrow is redundant." This is not correct as the meridians on conic projections are straight.

Line 93, the author states "the graticule is the honest device" to which I more or less agree but this situation should be regardless of the geometry of the graticule. I don't follow the author's assertion here.

Line 100, the author refers to "shape" again without qualification.

Line 104, be consistent and avoid mixing "coordinate reference system" with the map projection as the two ideas are not the same.

Line 104, I strongly disagree with the author's assertion that "tools exist to make the choice easy." Part of the problem is that these "tools" are still wrapped up using complex and confusing projection terminology (much like the author's insistence to use "shape"). The "easy" association needs further explanation or at least a supporting citation.

Technical Comments
Line 11 in the Abstract, "Insight" should not be capitalized

Figure 1, I don't know why these projections were included in this figure. The descriptive titles are not consistent in their content (e.g., the top row suggests their thematic utility but the rest do not, why?). I don't follow the meaning behind the statement that the "Graticule lines are shown in geographic coordinates (WGS84)." This statement makes no sense. All of the graticule lines are projected in the same map projection that shows the landmasses. Furthermore, the statement "area-preservation properties are indicated in each panel" is also not correct as two of the four map projections are conformal and not equal area. Overall, I don't understand the role of Figure 1 in the context of the author's article. I would have assumed to see examples of maps that report the projection, show the graticule, or include a scale bar/north arrow.

Reply
Citation: https://doi.org/10.5194/egusphere-2026-4209-RC2
- AC2: 'Reply on RC2', Dominic Royé, 15 Sep 2026 reply
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2026/egusphere-2026-4209/egusphere-2026-4209-AC2-supplement.pdf
  Reply
  
  Citation: https://doi.org/10.5194/egusphere-2026-4209-AC2

Dominic Royé

Supplement

https://doi.org/10.5194/egusphere-2026-4209-supplement

Dominic Royé

Viewed

Total article views: 312 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
208	74	30	312	38	31	35

HTML: 208
PDF: 74
XML: 30
Total: 312
Supplement: 38
BibTeX: 31
EndNote: 35

Views and downloads (calculated since 27 Aug 2026)

Month	HTML	PDF	XML	Total
Aug 2026	59	18	6	83
Sep 2026	149	56	24	229

Cumulative views and downloads (calculated since 27 Aug 2026)

Month	HTML	PDF	XML	Total
Aug 2026	59	18	6	83
Sep 2026	149	56	24	229

Viewed (geographical distribution)

Total article views: 306 (including HTML, PDF, and XML) Thereof 306 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 17 Sep 2026

Short summary

Scientists often keep a map's default projection without question, even though every projection changes how sizes, shapes, and distances appear. Scale bars and direction arrows are only accurate under certain projections, so a poor choice can make them misleading. Motivated by how often this goes unnoticed in published research, this paper argues that projections should be chosen deliberately and stated in each figure, so readers can trust what a map is showing.


Total:	0
HTML:	0
PDF:	0
XML:	0