2  Structure

For the related research project Black et al. (2025), we couple three main steps: Land Use Simulation, combined NCP Estimations and Species Distribution Modeling. These are shown bold in the Figure 2.1. evoland-plus HPC covers the LULC Simulation and steps dependent on its output, namely the NCP estimations, intensity analysis and focal window preparation. These steps, shown in blue, are unified to be automated for command line interface (CLI) and high performance computing (HPC) compatibility, to be scalable and reproducible.

flowchart TD
  LULCC[LULCC:<br>Land Use Simulation] --> Focal(Focal LULC<br>prep.)
  LULCC --> NCP[NCP Estimation]
  LULCC --> CheckLULCC[Intensity Analysis]
  NCP --> Clust(Clustering)
  LULCC --> NSDM[N-SDM:<br>Species Distribution<br>Modelling]
  Focal --> NSDM
  NSDM --> AggSp(Species Maps Aggregation)
  AggSp --> Clust

  style LULCC color:#2780e3, fill:#e9f2fc, stroke:#000000, stroke-width:3px, font-weight:bold
  style Focal color:#2780e3, fill:#e9f2fc, stroke:#000000
  style NCP color:#2780e3, fill:#e9f2fc, stroke:#000000, stroke-width:3px, font-weight:bold
  style Clust color:#b51c89, fill:#f9e8f5, stroke:#5f5f5f
  style NSDM color:#b51c89, fill:#f9e8f5, stroke:#5f5f5f, stroke-width:3px, font-weight:bold
  style AggSp color:#b51c89, fill:#f9e8f5, stroke:#5f5f5f

Figure 2.1: Pipeline Overview – Steps in blue are part of the evoland-plus HPC pipeline, red steps are external. From this structure, a clear order of execution can be derived.

First, we give an overview of each step in the pipeline. In contrast, the Pipeline section details and documents the individual steps and code used, in terms of input, output and execution.

2.1 LULCC: Land Use Simulation

The land use land cover change (LULCC) model for Switzerland (Black et al. 2023; Black et al. 2024) is the first step in the evoland-plus HPC pipeline. For given climate scenarios, it simulates land use changes in Switzerland until a given future year (e.g., 2060), based on historical data and future projections. The generated land use maps are then used as input for the following steps.

2.2 NCP: Nature’s Contributions to People

A range of NCP are estimated from the land use maps. Some NCP use further data for the estimation, e.g., precipitation and temperature projections.

The code is based on Külling et al. (2024), but has been adapted for automation and HPC compatibility.

Table 2.1: Nature’s Contributions to People (NCP) estimated from the LULCC model.

NCP	Name	Indicator
CAR	Regulation of climate	Carbon stored in biomass and soil
FF	Food and feed	Crop production potential (ecocrop)
HAB	Habitat creation and maintenance	Habitat quality index
NDR	Nutrient Delivery Ratio	Annual nutrient retention by vegetation
POL	Pollination and dispersal of seeds	Habitat abundance for pollinators
REC	Recreation potential	Recreation potential (RP) provided by ecosystems
SDR	Formation, protection and decontamination of soils	Erosion control by sediment retention
WY	Regulation of freshwater quantity, location and timing	Annual water yield

2.3 Intensity Analysis

The Intensity Analysis (IA) is based on x, and serves as

2.4 Focal LULC Preparation

Explain why These focal layers are used as inputs for X and Y, and act as NCP, specifically Z.

2.5 N-SDM: Nested Species Distribution Modelling

The nested species distribution modelling (N-SDM) (Adde et al. 2023) step simulates the distribution of species in Switzerland. Species occurrence used to fit models and covariate data are selected, both at different spatial scales, to predict the species distribution. This part is not integrated into the evoland-plus HPC pipeline, and needs to be consulted separately.

2.6 Species Maps Aggregation

The species maps are then aggregated to …

2.7 Clustering

To analyze the scenarios, the species maps are clustered to … The code for this step can be found in the separate repository reponame. These scripts are adapted to the case of Switzerland and the evoland-plus HPC project. …

Adde, Antoine, Pierre-Louis Rey, Philipp Brun, Nathan Külling, Fabian Fopp, Florian Altermatt, Olivier Broennimann, et al. 2023. “N-SDM : A High-Performance Computing Pipeline for Nested Species Distribution Modelling.” Ecography 2023 (6): e06540. https://doi.org/10.1111/ecog.06540.

Black, Benjamin, Antoine Adde, Daniel Farinotti, Antoine Guisan, Nathan Külling, Manuel Kurmann, Caroline Martin, et al. 2024. “Broadening the Horizon in Land Use Change Modelling: Normative Scenarios for Nature Positive Futures in Switzerland.” Regional Environmental Change 24 (3). https://doi.org/10.1007/s10113-024-02261-0.

Black, Benjamin, Antoine Adde, Nathan Külling, Carlson Büth, Manuel Kurmann, Anthony Lehmann, Florian Altermatt, Antoine Guisan, and Adrienne Gret-Regamey. 2025. “Identifying Robust Conservation Strategies to Secure Ecosystem Service Provision Under Uncertainties.”

Black, Benjamin, Maarten J. Van Strien, Antoine Adde, and Adrienne Grêt-Regamey. 2023. “Re-Considering the Status Quo: Improving Calibration of Land Use Change Models Through Validation of Transition Potential Predictions.” Environmental Modelling & Software 159 (January): 105574. https://doi.org/10.1016/j.envsoft.2022.105574.

Külling, Nathan, Antoine Adde, Audrey Lambiel, Sergio Wicki, Antoine Guisan, Adrienne Grêt-Regamey, and Anthony Lehmann. 2024. “Nature’s Contributions to People and Biodiversity Mapping in Switzerland: Spatial Patterns and Environmental Drivers.” Ecological Indicators 163: 112079. https://doi.org/10.1016/j.ecolind.2024.112079.