Transparent text example

[rolyp]

Candidate paragraphs for a "transparent text" example based on NAP3/CCRA3.

Some source documents:

• UK Climate Risk Independent Assessment (CCRA3)
• Independent Assessment of the Third National Adaptation Programme (NAP3)

Some of the data of interest (from Alan Kennedy-Asser):

• Evaluation-of-NAP3-and-CCRA3-ADAS.xlsx
• Independent-Assessment-of-UK-Climate-Risk-Advice-to-Government-Full-Risk-Table-Scoring-and-Priorities.xlsx
• Independent-Assessment-of-UK-Climate-Risk-Advice-to-Government-Charts-and-Data.xlsx
• The-Sixth-Carbon-Budget-Dataset_v2.xlsx

Data imported so far:

• See folder website/literate-execution/dataset/NAP3-CCRA3

See also:

• Paragraph: headings fluid#1443
• IPCC AR6 WG1 Fig TS.12(f) #6

[rolyp]

@trevormuwonge Let's capture our initial ideas here. Once we settle on a plan we'll rename the issue to something more specific.

[trevormuwonge]

Thanks @rolyp, I'll update shortly with my progress or questions.

[trevormuwonge]

@rolyp Just wanted to share some initial candidates I’ve shortlisted from the AR6 SPM for the transparent text example. These combine quantifiable statements with linked figures and seem well-suited to Fluid’s current chart capabilities (mainly bar/line).

---

Candidate 1: Warming over recent decades

Text (SPM A.1.2)

Figure: SPM.1

Justification:

• Strong temporal structure → ideal for a line chart.
• Compares observed vs modelled temperatures (human + natural vs natural only).
• Could demonstrate linked views (clicking a decade reveals model spread, for example).

---

Candidate 2: Human-caused temperature contributions

Text (SPM A.1.3)

Figure: SPM.2

Justification:

• Classic attribution chart with bars for GHGs, aerosols, natural variability etc.
• Visually decomposable and likely reproducible in Fluid.
• Useful for highlighting causality — “what drove how much”.

---

Candidate 3: Scenario-driven future warming

Text (SPM B.1.1 + Table SPM.1)

Figure: SPM.4 (Panel b)

Justification:

• Bar charts across SSP scenarios showing contribution from CO₂, aerosols, etc.
• Very clear structure, good for dropdowns or brushing interactions.
• Useful to show warming breakdown by emission driver per scenario.

---

Next steps

I’ll now:

• Confirm the datasets in the CEDA archive are usable and well-structured.
• Keep reviewing the SPM for more viable text–figure pairs, and will then move on to scan the TS and NAP3/CCRA3 for comparable examples.
• Let you know if I run into any issues or need to simplify further.

Happy to adjust focus depending on what’s most useful for the PROPL presentation or linking experiments.

[rolyp]

Hi Trevor, great start and really useful to have the analysis/justification. Thanks. I think these are all worth considering for the paper. SPM.4 is one we did make a start on, but as I recall the data availability was a bit limited. One issue you will soon encounter is that often the "data" in the CEDA Archive is essentially just the figure in another form. Finding some underlying data (from which the figure/text data was calculated) is harder! I think it's fine if the text linking sometimes just refers to a quantity in the input (this can be useful for comprehension/disambiguation), but the case isn't very convincing if all of the examples are like that.

Bearing that in mind, it would be most useful if you could isolate a paragraph or two where (a) there is data available in the CEDA Archive, but (b) the data isn't all already in the summary/aggregate form presented directly in the figure and/or described in the paragraph (so that to arrive at the numbers in the text some kind of calculation was involved).

Would paragraph A.1.2 would fall into this category? For example the statement about global surface temperature in 2001-2020 being 0.99 higher than 1850-1900 might involve some calculation. But if the "data" in the CEDA Archive already has a 20-year rolling average, then 0.99 might already be a point in that dataset.

For PROPL we may have to concentrate on the text only, given the time it will take to reconstruct the charts (i.e. I'll show text linking and visualisation linking separately, not how they interact). For the paper it might be nice to aim to implement #987 so we can recreate SPM.1.

In case it's helpful, here is the example I showed you and Alan the other day:

https://explorable-viz.github.io/fluid-article/scigen-1805.02474v1-10/

and the associated preprint:

https://dynamicaspects.org/papers/authoring-assistant.pdf

[rolyp]

@trevormuwonge Btw we do have "annual temperature anomaly" data in the NAP3/CCRA3 dataset, which we could potentially reuse for a IPPC example (if data isn't otherwise to hand).

(I've been experimenting with a new directory structure where the Fluid sources files and data files are stored in the same folder as the webpage -- so the page literate-execution/evaluation-nap3 as well as containing the web pages now contains .fld files and a dataset folder with some .json and the original .csv exports. If this seems like a better structure going forward, we can apply the same pattern to the other web pages.)

[trevormuwonge]

Thanks, @rolyp. I appreciate the detailed feedback and examples. That helps sharpen the focus a lot. I'll revisit the SPM candidates with this in mind and start isolating paragraphs that involve a real calculation step (rather than just pointing at pre-aggregated values). I'll also check A.1.2 again more carefully for what’s available in the CEDA archive.

Noted on the separation of text vs. figure linking for PROPL, and the idea of implementing #987 for the paper sounds exciting!

Also, cheers for the pointer on the NAP3/CCRA3 temperature anomaly data and the new directory structure. It sounds more streamlined. I'll take a look and keep that structure in mind going forward.

I’ll share another round of updates as soon as possible.

[trevormuwonge]

Hi @rolyp,

Here are some paragraphs for your consideration, drafted for PROPL, based on data from ccra-risks.json and ccra-actions-for-nap3.json.

---

CCRA Risks

How Risk Severity Increases Across Time and Temperature

Of the 47 climate risks assessed, 13 were rated High in the present day. This rises to 22 under a 2°C scenario by 2050, and 26 under 4°C. By 2080, the number of High-rated risks increases to 29 under 2°C, and 37 under 4°C. In total, 27 risks increased in severity between now and 2080 under a 4°C pathway. The data shows that both time and temperature contribute to escalation, with most additional High risks appearing after 2050, especially in the higher emissions scenario. This shift affects nearly 60% of all risks, and highlights where planning will need to anticipate worsening outcomes over the long term.

Near-Term Gaps in Adaptation to High-Risk Hazards

While 34 risks were projected to be high magnitude by 2050 under 4°C, nearly half (47%) of these were marked as ‘Partially’ or ‘Low’ in adaptation progress. This includes risks to critical infrastructure, natural ecosystems, and supply chains. Moreover, 61.7% of the risks that escalate under 4°C by 2080 had ‘More action needed’ as their urgency classification, revealing a dual gap in both capability and pace. Importantly, only 18 of these risks were listed as Committee Priorities, suggesting potential blind spots in formal prioritisation. Closing the gap between projected severity and current adaptation effort remains a critical policy opportunity for the upcoming NAP4 cycle.

Assessing the Timing of Policy Opportunity Against Escalating Risk

21 risk areas were identified as having high delivery opportunity in the next two years, and of these, 4 in 5 are flagged as Committee Priority risks. In contrast, none of the risks rated ‘Low’ in policy opportunity were committee priorities, despite more than half of them projected to reach High severity by 2080 under 4°C warming. This includes risk areas such as community preparedness, food security, and transport infrastructure, where future impacts are expected to grow, but no immediate intervention window has been noted.

---

CCRA Actions

Disparities in Action Allocation Across Risks and Opportunities

Out of 251 adaptation actions mapped across the CCRA3 evidence base, prioritisation was uneven both across and within risk categories. Risks to the built environment received the most concentrated attention, averaging 6.4 actions per risk, while infrastructure-related risks lag with under 3 actions on average, despite representing the most numerous group. Certain nodes such as biodiversity and natural systems, and food systems, were highly saturated, each receiving 10 or more actions, while others, such as health, were addressed by a single action. The median number of actions per risk was 4, but the range spans from 1 to 15, suggesting an unequal distribution where a few risks absorb disproportionate planning effort. Notably, opportunity-oriented issues accounted for 85 of the 251 actions (34%), indicating a strong tilt toward nature-based and cross-cutting adaptation frames. Such disparities underscore the importance of rebalancing efforts in NAP4 to ensure coherence between risk magnitude, urgency, and actionable response.

Limited Progress Despite Broad Recognition of Risk

Across the entire landscape of evaluated climate adaptation actions (n=251), the data revealed a persistent gap between recognition and effective implementation. While 33% of actions were rated as “minimally addressed” and another 27% as “significantly addressed,” only 6% were “fully addressed.” Markedly, nearly one-third (31%) of actions remained either unaddressed or only nominally recognized. This uneven pattern holds among actions explicitly aimed at government and arm’s-length bodies (n=62), where just 4 initiatives were considered fully addressed. The issue is not one of inaction alone, but of insufficient depth: the bulk of adaptation efforts are fixed at low-to-moderate engagement levels, highlighting a structural reluctance or inability to move beyond baseline interventions. This pattern suggests that while climate risks are widely acknowledged across sectors, genuine transformation is being hampered by a lack of sustained follow-through and cross-governmental coordination.

Sectoral and Departmental Disparities Signal Policy Misalignment

Disaggregating by department revealed acute disparities in government response. Natural environment and infrastructure sectors performed relatively well, with over 70% of their actions at least “minimally addressed.” In contrast, business and industry actions lagged severely, with nearly half either unaddressed or only nominally recognized. This suggests an institutional lapse where private sector resilience is concerned. At the departmental level, Defra (Dept. for Environment, Food and Rural Affairs) dominated with over 130 actions, although a small portion was fully or significantly addressed. Meanwhile, departments with key cross-sectoral mandates such as DSIT (Science & Technology) and HMT (Treasury) had a smaller action base and weaker scores. No department had more than 10% of its actions fully addressed, and in many, the largest category was “minimally addressed.” These gaps point to a fragmented governance ecosystem where adaptation is siloed, under-resourced, or misaligned with the urgency of the climate crisis.

Implementation Status Across All Actions

6% of the assessed actions (15 out of 251) were deemed fully addressed within current UK national adaptation plans, with a further 27% (68 actions) significantly addressed. In contrast, over half of the actions (more than 53%) remained either minimally addressed, recognised but not addressed, or not addressed at all. Specifically, 90 actions were tagged as minimally addressed, 35 as merely recognised, and 43 as not addressed. This distribution signals a significant implementation gap, especially given that many of these actions are linked to priority climate risks and opportunities already acknowledged in the CCRA3 framework. Transparent tracking of these gaps, particularly where actions have stagnated despite earlier recognition, is essential for closing the policy–delivery divide.

---