2025-11-10

Before

How to plan for futute landscapes

\[ \text{maximize CI} = \color{Red}{W_1 \times \text{Biodiversity}} + \color{Purple}{W_2 \times \text{Carbon}} \times \color{Blue}{W_3 \times\text{Contiguity}_n} \]

  • Biodiversity (Nature for Nature W1 = 1, W3 = 1, W2 = 0)
    • Species richness
    • Phylogenetic Diversity
    • Rarity
  • Carbon (Nature for Society W1 = 1, W3 = 1, W2 = 1)
    • Above ground carbon
    • Bellow ground carbon
  • Contiguity (Always there)
    • Contiguity to new solutions
    • Contiguity to current nature

Results

## Rows: 15 Columns: 16
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## dbl (16): weight, Biodiversity, CarbonRaw, CarbonWeighted, Contiguity, Obj, ...
## 
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

Optimization elbow

Percentage critical

We are being very conservative

  • This should be for worldwide distribution
  • This only includes what is in our solution not what already exists

Beta diversity

  • Sørensen and Jaccard weigh overlap differently but here they agree on the elbow near w≈0.4.

  • Interpretation: around that weight you still retain high between-site distinctness while boosting carbon—after that, extra carbon comes with rapidly diminishing \(\beta\) diversity.

Problem:

  • In this optimization the final potential is a snapshot, it should be a trajectory
  • End-state winners can be near-term violators unless we filter by the window.

Two intertwined timelines:

  • Climate emergency (near‑term): minimize additional warming in the next ~20 years.
  • Biodiversity recovery (longer‑term): secure space and processes for ecosystems to persist and adapt.
  • If we optimize biodiversity only without near‑term climate guardrails, we risk actions that raise early warming and undermine long‑term gains.

Working “Emergency Windows”

  • Window A (2026–2045): No‑additional‑warming guardrail at national land‑use scale.

    • Practical implication: constrain cumulative CO₂‑equivalent over this period to ≤ 0 (or to a tight carbon budget).
    • Prioritize measures that cut CO₂ fast and avoid big CH₄ spikes unless offset by strong co‑benefits.

  • Window B (2046–2075): consolidation & scaling of nature networks; continue decarbonization and stabilize land carbon.

  • Window C (2076–2100): resilience under residual climate change; maintain connectivity, hydrology, and evolutionary potential.

Rationale: aligns near‑term with global 1.5 °C pathways and mid‑/late‑century with EU nature restoration ambitions. (Exact dates are modeling choices; see Feedback slide.)

Some ideas for the temporal aspect

Network flow

  • Incorporating species dispersal (Species on the move)
  • Considering GWP (Global Warming Potential)
  • Quadratic Network flow
  • Multiple scenarios

Geographic representation

Global Warming Potential (GWP)

Global Warming Potential (GWP, help wanted)

  • How do we actually model it for Denmark for each intervention
  • How to model it for each landuse type

Non linearity

  • Rewetting curve: Adapted from Kalhori et al. (2024), Communications Earth & Environment
  • SOC response functions: After Jost et al. (2021), Journal of Environmental Management
  • Biomass parameters: IPCC (2006) Guidelines, Vol. 4 carbon fraction & root:shoot defaults.

Biodiversity modeling

  • Normally you could use an SDM with climate
  • But climate is not the main driver in restoration

  • Is it fine to use metanalysis code
  • Can we do it species by species

Biodiversity

Curves parameterized from meta-analyses: * peatland enhanced revegetation timelines from Allan et al. 2023; * grassland passive vs seeding from Ladouceur et al. 2023 and a 2023 synthesis of restoration methods * forest restoration trajectories from Crouzeilles et al. 2016, 2020; * wetland recovery context from Moreno-Mateos et al. 2012.

Example with limits

Filters

  • Carbon floor 20y: Keeps only trajectories whose minimum carbon over years 0–20 is ≥ carbon_floor_20 (default 0 tCO₂e/ha). Why: enforce “no near-term net carbon loss.”

  • Climate veil (top 50% @H): focus on long-run climate-robust futures.

  • Biodiv veil (top 50% @H): emphasize long-run biodiversity performance. Effect: hides the lower half of biodiversity outcomes at the horizon.

  • All three filters Intersection of all the above: passes the 20-year floor and is above-median in both carbon and biodiversity at H_sel. Why: strict shortlist of futures that are near-term safe and long-term strong on both axes.

References {.refs}