Plotting & Summary Visualization¶

Visualization is a central part of the PyPSA-Earth workflow. After solving a network, built-in scripts let you generate geographic maps of the power system, produce aggregated cost and energy summaries, and explore results interactively via Jupyter notebooks. This page documents those tools, explains how to invoke them via Snakemake, and highlights the relevant configuration options.

Overview of Visualization Scripts¶

PyPSA-Earth currently ships three main scripts for plotting and result analysis, located in the scripts/ directory:

Script	Snakemake Rule	Purpose
`plot_network.py`	`plot_network`	Renders geographic maps of solved networks
`make_summary.py`	`make_summary`	Aggregates optimization results into `.csv` summary tables
`plot_summary.py`	`plot_summary`	Generates summary charts from the `costs` and `energy` tables

All three scripts are driven by Snakemake wildcards and the plotting: section of config.yaml, so no manual Python invocation is needed for standard use.

Network Maps: `plot_network`¶

The plot_network rule creates static geographic maps of a solved network. Bus sizes represent installed capacity and line widths represent transfer capacity.

Running it¶

# Plot a solved network using the supported attr wildcard p_nom
snakemake -j 1 "results/plots/elec_s_10_ec_lcopt_Co2L-3H_p_nom.pdf"

# Plot the same map as a PNG
snakemake -j 1 "results/plots/elec_s_10_ec_lcopt_Co2L-3H_p_nom.png"

The {attr} wildcard is currently required and only supports p_nom. The {ext} wildcard controls the output format. Supported formats depend on your matplotlib backend, typically including pdf, png, and svg.

What is plotted¶

Buses as proportionally sized circles, split into pie segments by carrier.
Lines and links with widths proportional to their transfer capacity.
A geographic base map whose colors are controlled in the plotting config.

Relevant config options¶

plotting:
  map:
    figsize: [7, 7]
    boundaries: [-10.2, 29, 35, 72]
    p_nom:
      bus_size_factor: 5.e+4
      linewidth_factor: 3.e+3
    color_geomap:
      ocean: white
      land: whitesmoke

Tip

Adjust boundaries to zoom into your region of interest. If bus circles or line widths look too large or too small for your clustering level, tune bus_size_factor and linewidth_factor.

Summary Generation: `make_summary`¶

The make_summary rule aggregates solved networks into .csv files stored under results/summaries/. These files can be consumed by plot_summary.py or read directly in a notebook or data analysis workflow.

Running it¶

# Summarize all solved networks defined in config.yaml
snakemake -j 1 make_summary

# Summarize a specific scenario for one country (for example Nigeria)
snakemake -j 1 "results/summaries/elec_s_all_ec_lall_Co2L-3H_NG"

The {country} wildcard narrows the summary to buses belonging to a given two-letter ISO country code. Use all to include every country in the model.

Output files¶

make_summary.py writes several .csv files per scenario combination:

File	Description
`costs.csv`	Annualized system costs by technology
`curtailment.csv`	Curtailed energy by carrier
`energy.csv`	Annual generation and dispatch by carrier
`capacity.csv`	Installed capacity by carrier
`supply.csv`	Supply grouped by bus carrier and component
`supply_energy.csv`	Supply energy grouped by bus carrier and component
`prices.csv`	Mean marginal prices by bus carrier
`weighted_prices.csv`	Load-weighted prices by carrier
`metrics.csv`	High-level system metrics such as line volume and CO2 shadow prices

Relevant config options¶

costs:
  default_exchange_rate: 0.7532
  discountrate: 0.07

electricity:
  max_hours:
    battery: 6
    H2: 168

Summary Plots: `plot_summary`¶

Once make_summary has run, the plot_summary rule reads the summary tables and produces charts for the costs and energy summaries.

Running it¶

# Generate all configured summary plots
snakemake -j 1 plot_summary

# Plot the cost summary for a specific country as PDF
snakemake -j 1 "results/plots/summary_costs_elec_s_all_ec_lall_Co2L-3H_NG.pdf"

What is plotted¶

Cost summary as a stacked bar chart of annualized system cost by technology.
Energy summary as a stacked chart of annual generation and storage dispatch.

Small contributors can be grouped into Other using the configured thresholds:

plotting:
  costs_max: 10
  costs_threshold: 0.2
  energy_max: 20000
  energy_min: -20000
  energy_threshold: 15

Technology Color Palette¶

All plotting scripts share a color palette defined under plotting: tech_colors: in config.default.yaml. These colors are applied consistently across network maps and summary charts.

Key colors from the default configuration:

Carrier	Config key	Default color
Onshore Wind	`onwind`	`#235ebc`
Offshore Wind (AC)	`offwind-ac`	`#6895dd`
Offshore Wind (DC)	`offwind-dc`	`#74c6f2`
Solar PV	`solar`	`#f9d002`
Hydro	`hydro`	`#08ad97`
Pumped Hydro Storage	`PHS`	`#08ad97`
Run of River	`ror`	`#4adbc8`
OCGT	`OCGT`	`#d35050`
CCGT	`CCGT`	`#b80404`
nuclear	`nuclear`	`#ff9000`
Coal	`coal`	`#707070`
Oil	`oil`	`#262626`
Battery	`battery`	`slategray`
Hydrogen	`H2`	`#ea048a`

To override a color, add or edit the entry under plotting: tech_colors: in your config.yaml:

plotting:
  tech_colors:
    solar: "#e8c800"
    onwind: "#1a3e8a"

The helper _helpers.sanitize_carriers() in scripts/_helpers.py reads these values and assigns them to the PyPSA network's carriers component, making them automatically available to n.plot().

Technology Groupings¶

Plotting scripts sort carriers into named groups that control ordering and labelling of stacked charts. The default groupings are defined under plotting: in config.default.yaml and can be overridden in config.yaml.

plotting:
  vre_techs:
    - onwind
    - offwind-ac
    - offwind-dc
    - solar
    - ror
  conv_techs:
    - OCGT
    - CCGT
    - nuclear
    - Nuclear
    - coal
    - oil
  storage_techs:
    - hydro+PHS
    - battery
    - H2
  renewable_storage_techs:
    - PHS
    - hydro

Jupyter Notebook Examples¶

The pypsa-meets-earth/documentation repository contains Jupyter notebooks for exploring PyPSA-Earth outputs. These examples are useful for validation workflows and custom post-processing beyond the built-in Snakemake targets.

git clone https://github.com/pypsa-meets-earth/documentation.git
cd documentation
jupyter lab

Custom Plotting with PyPSA¶

For analyses beyond the built-in scripts, load a solved network directly in Python and use PyPSA's n.plot() method.

Load a solved network¶

import pypsa

n = pypsa.Network("results/networks/elec_s_10_ec_lcopt_Co2L-3H.nc")

Static network map¶

import matplotlib.pyplot as plt

fig, ax = plt.subplots(figsize=(8, 8))

bus_sizes = n.generators.groupby(["bus", "carrier"])["p_nom_opt"].sum() / 5e4

n.plot(
    ax=ax,
    bus_sizes=bus_sizes,
    bus_colors=n.carriers["color"],
    line_widths=n.lines["s_nom"] / 3e3,
    geomap=True,
)
plt.tight_layout()
plt.savefig("my_network_map.png", dpi=150)

Generation mix bar chart¶

import matplotlib.pyplot as plt

gen = (
    n.generators_t.p.multiply(n.snapshot_weightings.generators, axis=0)
    .sum()
    .groupby(n.generators.carrier)
    .sum()
    / 1e6
)

colors = n.carriers.loc[gen.index, "color"]

gen.plot.bar(color=colors, figsize=(10, 5))
plt.ylabel("Annual generation (TWh)")
plt.title("Generation mix")
plt.tight_layout()
plt.savefig("generation_mix.png", dpi=150)

Line loading map¶

import matplotlib.pyplot as plt

fig, ax = plt.subplots(figsize=(8, 8))

loading = n.lines_t.p0.abs().divide(n.lines.s_nom, axis=1).mean()

n.plot(
    ax=ax,
    line_colors=loading,
    line_cmap="YlOrRd",
    line_widths=1.5,
    geomap=True,
)
plt.savefig("line_loading.png", dpi=150)

Note

For a full reference of n.plot() parameters, see the PyPSA plotting documentation.

Quick Reference¶

Task	Command
Plot a solved network map	`snakemake -j 1 "results/plots/elec_s_10_ec_lcopt_Co2L-3H_p_nom.pdf"`
Generate summary CSVs	`snakemake -j 1 make_summary`
Country-specific summary	`snakemake -j 1 "results/summaries/elec_s_all_ec_lall_Co2L-3H_NG"`
Plot summary charts	`snakemake -j 1 plot_summary`
Plot a specific summary figure	`snakemake -j 1 "results/plots/summary_costs_elec_s_all_ec_lall_Co2L-3H_NG.pdf"`

Explore valid wildcard combinations for your configuration with a dry-run:

snakemake -j 1 plot_summary -n

For more on the wildcards used in these rules, see Wildcards.

Plotting & Summary Visualization¶

Overview of Visualization Scripts¶

Network Maps: plot_network¶

Running it¶

What is plotted¶

Relevant config options¶

Summary Generation: make_summary¶

Running it¶

Output files¶

Relevant config options¶

Summary Plots: plot_summary¶

Running it¶

What is plotted¶

Technology Color Palette¶

Technology Groupings¶

Jupyter Notebook Examples¶

Custom Plotting with PyPSA¶

Load a solved network¶

Static network map¶

Generation mix bar chart¶

Line loading map¶

Quick Reference¶

Network Maps: `plot_network`¶

Summary Generation: `make_summary`¶

Summary Plots: `plot_summary`¶