som_plots

Methods to plot SOM results.

build_display(res, data, out_dir, use_parallel_processing=False, selection=None)

Constructs plots to visualize results.

Parameters:

Name	Type	Description	Default
res	dict	SOM results.	required
data	dict	SOM input data.	required
out_dir	str	plot output directory.	required
use_parallel_processing	bool	toggle for multiprocessing.	False
selection	dict	filtering options.	None

Source code in src/som_plots.py

def build_display(res: dict[str, dict[str, pd.DataFrame]], data: dict[str, pd.DataFrame], out_dir: str, use_parallel_processing: bool = False, selection: dict[str, list] = None):
    """
    Constructs plots to visualize results.

    Arguments:
        res (dict): SOM results.
        data (dict): SOM input data.
        out_dir (str): plot output directory.
        use_parallel_processing (bool): toggle for multiprocessing.
        selection (dict): filtering options.
    """
    res = copy.deepcopy(res)
    data = copy.deepcopy(data)

    if selection is not None:
        print('\t\tFiltering results...')
        res = filter_results(res, selection)
        data = filter_ids(data, selection)

    areas = data['area']['ID']

    cpu_count = multiprocessing.cpu_count()     # available cpu cores
    with multiprocessing.Manager() as manager:
        progress = manager.Namespace()
        progress.current = 0
        progress.total = len(areas)
        lock = manager.Lock()
        if use_parallel_processing:
            with multiprocessing.Pool(processes=(min(cpu_count - 2, len(areas)))) as pool:
                jobs = [(area, res, data, out_dir, progress, lock) for area in areas]
                progress.current = 0
                display_progress(progress.current / progress.total, text='\t\tTPL: ')
                pool.starmap(plot_total_pressure_load_levels, jobs)
                display_progress(progress.current / progress.total, text='\t\tTPL: ')
                progress.current = 0
                display_progress(progress.current / progress.total, text='\n\t\tPressures: ')
                pool.starmap(plot_pressure_levels, jobs)
                display_progress(progress.current / progress.total, text='\t\tPressures: ')
                progress.current = 0
                display_progress(progress.current / progress.total, text='\n\t\tThresholds: ')
                pool.starmap(plot_thresholds, jobs)
                display_progress(progress.current / progress.total, text='\t\tThresholds: ')
                progress.current = 0
                display_progress(progress.current / progress.total, text='\n\t\tStatePressures: ')
                pool.starmap(plot_state_pressure_levels, jobs)
                display_progress(progress.current / progress.total, text='\t\tStatePressures: ')
                progress.current = 0
                display_progress(progress.current / progress.total, text='\n\t\tActivityContributions: ')
                pool.starmap(plot_activity_contributions, jobs)
                display_progress(progress.current / progress.total, text='\t\tActivityContributions: ')
                progress.current = 0
                display_progress(progress.current / progress.total, text='\n\t\tPressureContributions: ')
                pool.starmap(plot_pressure_contributions, jobs)
                display_progress(progress.current / progress.total, text='\t\tPressureContributions: ')
        else:
            progress.current = 0
            display_progress(progress.current / progress.total, text='\t\tTPL: ')
            for area in areas:
                plot_total_pressure_load_levels(area, res, data, out_dir, progress, lock)
            display_progress(progress.current / progress.total, text='\t\tTPL: ')
            progress.current = 0
            display_progress(progress.current / progress.total, text='\n\t\tPressures: ')
            for area in areas:
                plot_pressure_levels(area, res, data, out_dir, progress, lock)
            display_progress(progress.current / progress.total, text='\t\tPressures: ')
            progress.current = 0
            display_progress(progress.current / progress.total, text='\n\t\tThresholds: ')
            for area in areas:
                plot_thresholds(area, res, data, out_dir, progress, lock)
            display_progress(progress.current / progress.total, text='\t\tThresholds: ')
            progress.current = 0
            display_progress(progress.current / progress.total, text='\n\t\tStatePressures: ')
            for area in areas:
                plot_state_pressure_levels(area, res, data, out_dir, progress, lock)
            display_progress(progress.current / progress.total, text='\t\tStatePressures: ')
            progress.current = 0
            display_progress(progress.current / progress.total, text='\n\t\tActivityContributions: ')
            for area in areas:
                plot_activity_contributions(area, res, data, out_dir, progress, lock)
            display_progress(progress.current / progress.total, text='\t\tActivityContributions: ')
            progress.current = 0
            display_progress(progress.current / progress.total, text='\n\t\tPressureContributions: ')
            for area in areas:
                plot_pressure_contributions(area, res, data, out_dir, progress, lock)
            display_progress(progress.current / progress.total, text='\t\tPressureContributions: ')

    #
    # Measure effects
    #

    print('\n\t\tMeasure effects...')

    # plot settings
    capsize = 3
    capthick = 1
    elinewidth = 1
    ecolor = 'salmon'
    label_angle = 60
    char_limit = 25
    bar_width = 0.4
    edge_color = 'black'

    fig, ax = plt.subplots(figsize=(100, 14), constrained_layout=True)

    bar_width = 0.8
    edge_color = 'black'
    activity_font_size = 8

    # adjust data
    df = res['MeasureEffects']['Mean'].merge(res['MeasureEffects']['Error'], on=['measure', 'pressure', 'state', 'activity'], how='left', suffixes=('_mean', '_error'))
    df = df.sort_values(by=['measure', 'pressure', 'state', 'activity'])
    suffixes = ('', '_name')
    for col in ['measure', 'activity', 'pressure', 'state']:
        df = df.merge(data[col].loc[:, [col, 'ID']], left_on=col, right_on='ID', how='left', suffixes=suffixes)
        df = df.drop(columns=[col, 'ID'])
        df = df.rename(columns={col+'_name': col})
        df.loc[:, col] = np.array([(x[:char_limit]+'...' if len(x) > char_limit else x) if type(x) == str else 'All' for x in df.loc[:, col].values])
    df['index'] = np.arange(len(df))
    x_ticks = {x: df[df['measure'] == x]['index'].mean() for x in df['measure'].unique()}

    # set colors
    df['color_key'] = df['pressure'].astype(str) + '_' + df['state'].astype(str)
    unique_keys = df['color_key'].unique()
    colors = plt.cm.viridis(np.linspace(0, 1, len(unique_keys)))
    color_map = {key: colors[i] for i, key in enumerate(unique_keys)}

    # create plot
    for key in unique_keys:
        subset = df[df['color_key'] == key]
        bars = ax.bar(subset['index'], subset['reduction_mean'] * 100, width=bar_width, color=color_map[key], label=key if key not in ax.get_legend_handles_labels()[1] else '', edgecolor=edge_color)
        ax.errorbar(subset['index'], subset['reduction_mean'] * 100, yerr=subset['reduction_error'] * 100, linestyle='None', marker='None', capsize=capsize, capthick=capthick, elinewidth=elinewidth, ecolor=ecolor)
        for bar, (_, row) in zip(bars, subset.iterrows()):
            ax.text(bar.get_x() + bar.get_width() / 2, bar.get_height() / 2, str(row['activity']), 
                    ha='center', va='center', rotation=90, fontsize=activity_font_size, color='white')

    ax.set_xlabel('Measure')
    ax.set_ylabel('Reduction effect (%)')
    ax.set_title(f'Measure Reduction Effects')
    ax.set_xticks(list(x_ticks.values()), list(x_ticks.keys()), rotation=label_angle, ha='right')
    ax.yaxis.grid(True, linestyle='--', color='lavender')
    ax.legend(title='Pressure/State', bbox_to_anchor=(1.05, 1), loc='upper left')

    # adjust axis limits
    x_lim = [- 0.5, len(df) - 0.5]
    ax.set_xlim(x_lim)
    y_lim = [0, 100]
    ax.set_ylim(y_lim)

    # export
    for area in areas:
        area_name = data['area'].loc[areas == area, 'area'].values[0]
        temp_dir = os.path.join(out_dir, f'area_{area}_{area_name}')
        plt.savefig(os.path.join(temp_dir, f'area_{area}_{area_name}_MeasureEffects.png'), dpi=200)

    plt.close(fig)

filter_ids(input_data, selection)

Filter input data id dataframes.

Parameters:

Name	Type	Description	Default
input_data	dict	SOM input data.	required
selection	dict	filtering options.	required

Returns:

Name	Type	Description
input_data	dict	filtered input data.

Source code in src/som_plots.py

def filter_ids(input_data: dict[str, pd.DataFrame], selection: dict[str, list]) -> dict[str, pd.DataFrame]:
    """
    Filter input data id dataframes.

    Arguments:
        input_data (dict): SOM input data.
        selection (dict): filtering options.

    Returns:
        input_data (dict): filtered input data.
    """
    for key, values in [
        ('measure', selection['measure']), 
        ('activity', selection['activity']), 
        ('pressure', selection['pressure']), 
        ('state', selection['state']), 
        ('area', selection['area'])
    ]:
        if values != []:
            input_data[key] = input_data[key].loc[input_data[key]['ID'].isin(values), :]
    return input_data

filter_results(res, selection)

Filter results for more selective output.

Parameters:

Name	Type	Description	Default
res	dict	SOM results.	required
selection	dict	filtering options.	required

Returns:

Name	Type	Description
res	dict	filtered results.

Source code in src/som_plots.py

def filter_results(res: dict[str, pd.DataFrame], selection: dict[str, list]) -> dict[str, pd.DataFrame]:
    """
    Filter results for more selective output.

    Arguments:
        res (dict): SOM results.
        selection (dict): filtering options.

    Returns:
        res (dict): filtered results.
    """
    # Pressure, TPL, TPLRed, Thresholds
    for key, values in [
        ('Pressure', selection['pressure']), 
        ('TPL', selection['state']), 
        ('TPLRed', selection['state']), 
        ('Thresholds', selection['state'])
    ]:
        for r in ['Mean', 'Error']:
            if values != []:
                if selection['area'] != []:
                    res[key][r] = res[key][r].loc[res[key][r]['ID'].isin(values), ['ID'] + selection['area']]
                else:
                    res[key][r] = res[key][r].loc[res[key][r]['ID'].isin(values), :]
    # StatePressure
    if selection['pressure'] != []:
        for s in res['StatePressure']:
            for r in ['Mean', 'Error']:
                if selection['area'] != []:
                    res['StatePressure'][s][r] = res['StatePressure'][s][r].loc[res['StatePressure'][s][r]['ID'].isin(selection['pressure']), ['ID'] + selection['area']]
                else:
                    res['StatePressure'][s][r] = res['StatePressure'][s][r].loc[res['StatePressure'][s][r]['ID'].isin(selection['pressure']), :]
    # MeasureEffects, ActivityContributions, PressureContributions
    for key, cols in {
        'MeasureEffects': {
            'measure': selection['measure'], 
            'activity': selection['activity'], 
            'pressure': selection['pressure'], 
            'state': selection['state']
        }, 
        'ActivityContributions': {
            'activity': selection['activity'], 
            'pressure': selection['pressure'], 
            'area_id': selection['area']
        }, 
        'PressureContributions': {
            'state': selection['state'], 
            'pressure': selection['pressure'], 
            'area_id': selection['area']
        }
    }.items():
        for r in ['Mean', 'Error']:
            for col, values in cols.items():
                if values != []:
                    res[key][r] = res[key][r].loc[res[key][r][col].isin(values), :]
    return res

plot_activity_contributions(area, res, data, out_dir, progress, lock)

Plots activity contributions.

Source code in src/som_plots.py

def plot_activity_contributions(area, res, data, out_dir, progress, lock):
    """
    Plots activity contributions.
    """
    # create new directory for the plots
    area_name = data['area'].loc[data['area']['ID'] == area, 'area'].values[0]
    out_dir = os.path.join(out_dir, f'area_{area}_{area_name}', 'contributions', 'activity')
    os.makedirs(out_dir, exist_ok=True)

    # plot settings
    char_limit = 25

    for pressure in res['ActivityContributions']['Mean']['pressure'].unique():
        pressure_name = data['pressure'].loc[data['pressure']['ID'] == pressure, 'pressure'].values[0]

        out_path = os.path.join(out_dir, f'area_{area}_{area_name}_pressure_{pressure}_ActivityContributions.png')

        fig, ax = plt.subplots(figsize=(12, 9), constrained_layout=True)

        # adjust data
        suffixes = ('_mean', '_error')
        df = pd.merge(res['ActivityContributions']['Mean'].loc[:, :], res['ActivityContributions']['Error'].loc[:, :], on=['activity', 'pressure', 'area_id'], suffixes=suffixes)
        contributions = df.loc[(df['area_id'] == area) & (df['pressure'] == pressure), :]
        labels = contributions['activity']
        labels = contributions['activity'].apply(lambda x: data['activity'].loc[data['activity']['ID'] == x, 'activity'].values[0])
        labels = np.array([x[:char_limit]+'...' if len(x) > char_limit else x for x in labels])     # limit characters to char_limit
        vals = contributions['contribution_mean'] * 100    # convert to %

        # create plot
        ax.pie(vals, labels=labels, autopct='%1.1f%%')
        ax.set_title(f'Activity Contributions to Pressure ({pressure_name})\n({area_name})')

        # export
        plt.savefig(out_path, dpi=200)
        plt.close(fig)

    with lock:
        progress.current += 1
        display_progress(progress.current / progress.total, text='\t\tActivityContributions: ')

plot_pressure_contributions(area, res, data, out_dir, progress, lock)

Plots pressure contributions.

Source code in src/som_plots.py

def plot_pressure_contributions(area, res, data, out_dir, progress, lock):
    """
    Plots pressure contributions.
    """
    # create new directory for the plots
    area_name = data['area'].loc[data['area']['ID'] == area, 'area'].values[0]
    out_dir = os.path.join(out_dir, f'area_{area}_{area_name}', 'contributions', 'pressure')
    os.makedirs(out_dir, exist_ok=True)

    # plot settings
    char_limit = 25

    for state in res['PressureContributions']['Mean']['state'].unique():
        state_name = data['state'].loc[data['state']['ID'] == state, 'state'].values[0]

        out_path = os.path.join(out_dir, f'area_{area}_{area_name}_state_{state}_PressureContributions.png')

        fig, ax = plt.subplots(figsize=(12, 9), constrained_layout=True)

        # adjust data
        suffixes = ('_mean', '_error')
        df = pd.merge(res['PressureContributions']['Mean'].loc[:, :], res['PressureContributions']['Error'].loc[:, :], on=['state', 'pressure', 'area_id'], suffixes=suffixes)
        contributions = df.loc[(df['area_id'] == area) & (df['state'] == state) & (pd.notna(df['contribution_mean'])), :]
        if len(contributions) > 0:
            labels = contributions['pressure']
            labels = contributions['pressure'].apply(lambda x: data['pressure'].loc[data['pressure']['ID'] == x, 'pressure'].values[0])
            labels = np.array([x[:char_limit]+'...' if len(x) > char_limit else x for x in labels])     # limit characters to char_limit
            vals = contributions['contribution_mean'] * 100    # convert to %

            # create plot
            ax.pie(vals, labels=labels, autopct='%1.1f%%')
            ax.set_title(f'Pressure Contributions to State ({state_name})\n({area_name})')

            # export
            plt.savefig(out_path, dpi=200)
        plt.close(fig)

    with lock:
        progress.current += 1
        display_progress(progress.current / progress.total, text='\t\tPressureContributions: ')

plot_pressure_levels(area, res, data, out_dir, progress, lock)

Plots pressures.

Source code in src/som_plots.py

def plot_pressure_levels(area, res, data, out_dir, progress, lock):
    """
    Plots pressures.
    """
    # create new directory for the plots
    area_name = data['area'].loc[data['area']['ID'] == area, 'area'].values[0]
    out_path = os.path.join(out_dir, f'area_{area}_{area_name}', f'area_{area}_{area_name}_PressureLevels.png')
    os.makedirs(os.path.dirname(out_path), exist_ok=True)

    # plot settings
    marker = 's'
    markersize = 5
    markercolor = 'black'
    capsize = 3
    capthick = 1
    elinewidth = 1
    ecolor = 'salmon'
    label_angle = 60
    char_limit = 25

    fig, ax = plt.subplots(figsize=(25, 12), constrained_layout=True)

    # adjust data
    suffixes = ('_mean', '_error')
    df = pd.merge(res['Pressure']['Mean'].loc[:, ['ID', area]], res['Pressure']['Error'].loc[:, ['ID', area]], on='ID', suffixes=suffixes)
    x_vals = data['pressure'].loc[:, 'pressure'].values
    x_vals = np.array([x[:char_limit]+'...' if len(x) > char_limit else x for x in x_vals])     # limit characters to char_limit
    y_vals = df[str(area)+'_mean'] * 100    # convert to %
    y_err = df[str(area)+'_error'] * 100    # conver to %

    # create plot
    ax.errorbar(np.arange(len(x_vals)), y_vals, yerr=y_err, linestyle='None', marker=marker, capsize=capsize, capthick=capthick, elinewidth=elinewidth, markersize=markersize, color=markercolor, ecolor=ecolor)
    ax.set_xlabel('Pressure')
    ax.set_ylabel('Level (%)')
    ax.set_title(f'Pressure Levels\n({area_name})')
    ax.set_xticks(np.arange(len(x_vals)), x_vals, rotation=label_angle, ha='right')
    ax.yaxis.grid(True, linestyle='--', color='lavender')

    # adjust axis limits
    x_lim = [- 0.5, len(x_vals) - 0.5]
    ax.set_xlim(x_lim)
    y_lim = [-5, 105]
    ax.set_ylim(y_lim)

    # export
    plt.savefig(out_path, dpi=200)
    plt.close(fig)

    with lock:
        progress.current += 1
        display_progress(progress.current / progress.total, text='\t\tPressures: ')

plot_state_pressure_levels(area, res, data, out_dir, progress, lock)

Plots state pressures.

Source code in src/som_plots.py

def plot_state_pressure_levels(area, res, data, out_dir, progress, lock):
    """
    Plots state pressures.
    """
    # create new directory for the plots
    area_name = data['area'].loc[data['area']['ID'] == area, 'area'].values[0]
    out_dir = os.path.join(out_dir, f'area_{area}_{area_name}', 'state')
    os.makedirs(out_dir, exist_ok=True)

    # plot settings
    marker = 's'
    markersize = 5
    markercolor = 'black'
    capsize = 3
    capthick = 1
    elinewidth = 1
    ecolor = 'salmon'
    label_angle = 60
    char_limit = 25

    for state in res['StatePressure']:
        state_name = data['state'].loc[data['state']['ID'] == state, 'state'].values[0]

        out_path = os.path.join(out_dir, f'area_{area}_{area_name}_state_{state}_PressureLevels.png')

        fig, ax = plt.subplots(figsize=(25, 12), constrained_layout=True)

        # adjust data
        suffixes = ('_mean', '_error')
        df = pd.merge(res['StatePressure'][state]['Mean'].loc[:, ['ID', area]], res['StatePressure'][state]['Error'].loc[:, ['ID', area]], on='ID', suffixes=suffixes)
        x_vals = data['pressure'].loc[:, 'pressure'].values
        x_vals = np.array([x[:char_limit]+'...' if len(x) > char_limit else x for x in x_vals])     # limit characters to char_limit
        y_vals = df[str(area)+'_mean'] * 100    # convert to %
        y_err = df[str(area)+'_error'] * 100    # conver to %

        # create plot
        ax.errorbar(np.arange(len(x_vals)), y_vals, yerr=y_err, linestyle='None', marker=marker, capsize=capsize, capthick=capthick, elinewidth=elinewidth, markersize=markersize, color=markercolor, ecolor=ecolor)
        ax.set_xlabel('Pressure')
        ax.set_ylabel('Level (%)')
        ax.set_title(f'Pressure Levels ({state_name})\n({area_name})')
        ax.set_xticks(np.arange(len(x_vals)), x_vals, rotation=label_angle, ha='right')
        ax.yaxis.grid(True, linestyle='--', color='lavender')

        # adjust axis limits
        x_lim = [- 0.5, len(x_vals) - 0.5]
        ax.set_xlim(x_lim)
        y_lim = [-5, 105]
        ax.set_ylim(y_lim)

        # export
        plt.savefig(out_path, dpi=200)
        plt.close(fig)

    with lock:
        progress.current += 1
        display_progress(progress.current / progress.total, text='\t\tStatePressures: ')

plot_thresholds(area, res, data, out_dir, progress, lock)

Plots thresholds comparison.

Source code in src/som_plots.py

def plot_thresholds(area, res, data, out_dir, progress, lock):
    """
    Plots thresholds comparison.
    """
    # create new directory for the plots
    area_name = data['area'].loc[data['area']['ID'] == area, 'area'].values[0]
    out_path = os.path.join(out_dir, f'area_{area}_{area_name}', f'area_{area}_{area_name}_Thresholds.png')
    os.makedirs(os.path.dirname(out_path), exist_ok=True)

    # plot settings
    capsize = 3
    capthick = 1
    elinewidth = 1
    ecolor = 'salmon'
    label_angle = 60
    char_limit = 25
    bar_width = 0.4
    bar_color_1 = 'turquoise'
    bar_color_2 = 'seagreen'
    edge_color = 'black'

    fig, ax = plt.subplots(figsize=(16, 12), constrained_layout=True)

    # adjust data
    x_labels = np.array([x[:char_limit]+'...' if len(x) > char_limit else x for x in data['state'].loc[:, 'state'].values])     # limit characters to char_limit
    x_vals = np.arange(len(x_labels))
    suffixes = ('_mean', '_error')
    df = pd.merge(res['TPLRed']['Mean'].loc[:, ['ID', area]], res['TPLRed']['Error'].loc[:, ['ID', area]], on='ID', suffixes=suffixes)
    y_vals_tpl = df[str(area)+'_mean'] * 100    # convert to %
    y_err_tpl = df[str(area)+'_error'] * 100    # conver to %
    df = pd.merge(res['Thresholds']['Mean'].loc[:, ['ID', area]], res['Thresholds']['Error'].loc[:, ['ID', area]], on='ID', suffixes=suffixes)
    y_vals_ges = df[str(area)+'_mean'] * 100    # convert to %
    y_err_ges = df[str(area)+'_error'] * 100    # convert to %

    # create plot
    label_tpl = 'Reduction with measures'
    ax.bar(x_vals-bar_width/2, y_vals_tpl, width=bar_width, align='center', color=bar_color_1, label=label_tpl, edgecolor=edge_color)
    ax.errorbar(x_vals-bar_width/2, y_vals_tpl, yerr=y_err_tpl, linestyle='None', marker='None', capsize=capsize, capthick=capthick, elinewidth=elinewidth, ecolor=ecolor)
    label_ges = 'Target'
    ax.bar(x_vals+bar_width/2, y_vals_ges, width=bar_width, align='center', color=bar_color_2, label=label_ges, edgecolor=edge_color)
    ax.errorbar(x_vals+bar_width/2, y_vals_ges, yerr=y_err_ges, linestyle='None', marker='None', capsize=capsize, capthick=capthick, elinewidth=elinewidth, ecolor=ecolor)
    ax.set_xlabel('Environmental State')
    ax.set_ylabel('Reduction (%)')
    ax.set_title(f'Total Pressure Load Reduction vs. GES Reduction Thresholds\n({area_name})')
    ax.set_xticks(x_vals, x_labels, rotation=label_angle, ha='right')
    ax.yaxis.grid(True, linestyle='--', color='lavender')
    ax.legend()

    # adjust axis limits
    x_lim = [- 0.5, len(x_vals) - 0.5]
    ax.set_xlim(x_lim)
    y_lim = [0, 100]
    ax.set_ylim(y_lim)

    # export
    plt.savefig(out_path, dpi=200)
    plt.close(fig)

    with lock:
        progress.current += 1
        display_progress(progress.current / progress.total, text='\t\tThresholds: ')

plot_total_pressure_load_levels(area, res, data, out_dir, progress, lock)

Plots Total Pressure Load.

Source code in src/som_plots.py

def plot_total_pressure_load_levels(area, res, data, out_dir, progress, lock):
    """
    Plots Total Pressure Load.
    """
    # create new directory for the plots
    area_name = data['area'].loc[data['area']['ID'] == area, 'area'].values[0]
    out_path = os.path.join(out_dir, f'area_{area}_{area_name}', f'area_{area}_{area_name}_TotalPressureLoadLevels.png')
    os.makedirs(os.path.dirname(out_path), exist_ok=True)

    # plot settings
    marker = 's'
    markersize = 5
    markercolor = 'black'
    capsize = 3
    capthick = 1
    elinewidth = 1
    ecolor = 'salmon'
    label_angle = 60
    char_limit = 25

    fig, ax = plt.subplots(figsize=(16, 12), constrained_layout=True)

    # adjust data
    suffixes = ('_mean', '_error')
    df = pd.merge(res['TPL']['Mean'].loc[:, ['ID', area]], res['TPL']['Error'].loc[:, ['ID', area]], on='ID', suffixes=suffixes)
    x_vals = data['state'].loc[:, 'state'].values
    x_vals = np.array([x[:char_limit]+'...' if len(x) > char_limit else x for x in x_vals])     # limit characters to char_limit
    y_vals = df[str(area)+'_mean'] * 100    # convert to %
    y_err = df[str(area)+'_error'] * 100    # conver to %

    # create plot
    ax.errorbar(np.arange(len(x_vals)), y_vals, yerr=y_err, linestyle='None', marker=marker, capsize=capsize, capthick=capthick, elinewidth=elinewidth, markersize=markersize, color=markercolor, ecolor=ecolor)
    ax.set_xlabel('Environmental State')
    ax.set_ylabel('Level (%)')
    ax.set_title(f'Total Pressure Load on Environmental States\n({area_name})')
    ax.set_xticks(np.arange(len(x_vals)), x_vals, rotation=label_angle, ha='right')
    ax.yaxis.grid(True, linestyle='--', color='lavender')

    # adjust axis limits
    x_lim = [- 0.5, len(x_vals) - 0.5]
    ax.set_xlim(x_lim)
    y_lim = [-5, 105]
    ax.set_ylim(y_lim)

    # export
    plt.savefig(out_path, dpi=200)
    plt.close(fig)

    with lock:
        progress.current += 1
        display_progress(progress.current / progress.total, text='\t\tTPL: ')