Source code for ezclimate.bau

import numpy as np
from abc import ABCMeta, abstractmethod

class BusinessAsUsual(object):
    """Abstract BAU class for the EZ-Climate model.

    Parameters
    ----------
    ghg_start : float
        today's GHG-level
    ghg_end : float
        GHG-level in the last period
    
    Attributes
    ----------
    ghg_start : float
        today's GHG-level
    ghg_end : float
        GHG-level in the last period
    emission_by_decisions : ndarray
        emissions at decision time periods??
    emission_per_period : ndarray
        total emission at decision time period??
    emission_to_ghg : ndarray
        GHG levels in decision time period??
    emission_to_bau : float
        constant for converting GHG to emission??

    """
    __metaclass__ = ABCMeta
    def __init__(self, ghg_start, ghg_end):
        self.ghg_start = ghg_start
        self.ghg_end = ghg_end
        self.emission_by_decisions = None
        self.emission_per_period = None
        self.emission_to_ghg = None
        self.emission_to_bau = None
        self.bau_path = None

    @abstractmethod
    def emission_by_time(self):
        pass


class DLWBusinessAsUsual(BusinessAsUsual):
    """Business-as-usual scenario of emissions. Emissions growth is assumed to slow down 
    exogenously - these assumptions represent an attempt to model emissions growth in a 
    business-as-usual scenario that is in the absence of incentives.

    Parameters
    ----------
    ghg_start : float
        today's GHG-level
    ghg_end : float
        GHG-level in the last period
    emit_time : ndarray or list
        time, in years, from now when emissions occurs
    emit_level : ndarray or list
        emission levels in future times `emit_time`

    Attributes
    ----------
    ghg_start : float
        today's GHG-level
    ghg_end : float
        GHG-level in the last period
    emission_by_decisions : ndarray
        emissions at decision time periods??
    emission_per_period : ndarray
        total emission at decision time period??
    emission_to_ghg : ndarray
        GHG levels in decision time period??
    emission_to_bau : float
        constant for converting GHG to emission??
    emit_time : ndarray or list
        time, in years, from now when emissions occurs
    emit_level : ndarray or list
        emission levels in future times `emit_time`

    """
    def __init__(self, ghg_start=400.0, ghg_end=1000.0, emit_time=[0, 30, 60], emit_level=[52.0, 70.0, 81.4]):
        super(DLWBusinessAsUsual, self).__init__(ghg_start, ghg_end)
        self.emit_time = emit_time
        self.emit_level = emit_level

    def emission_by_time(self, time):
        """Returns the BAU emissions at any time

        Parameters
        ----------
        time : int 
            future time period in years

        Returns
        -------
        float
            emission

        """
        if time < self.emit_time[1]:
            emissions = self.emit_level[0] + float(time) / (self.emit_time[1] - self.emit_time[0]) \
                        * (self.emit_level[1] - self.emit_level[0])
        elif time < self.emit_time[2]:
            emissions = self.emit_level[1] + float(time - self.emit_time[1]) / (self.emit_time[2] 
                        - self.emit_time[1]) * (self.emit_level[2] - self.emit_level[1])
        else:
            emissions = self.emit_level[2]
        return emissions

    def bau_emissions_setup(self, tree):
        """Create default business as usual emissions path. The emission rate in each period is 
        assumed to be the average of the emissions at the beginning and at the end of the period.

        Parameters
        ----------
        tree : `TreeModel` object
            provides the tree structure used
            
        """
        num_periods = tree.num_periods
        self.emission_by_decisions = np.zeros(num_periods)
        self.emission_per_period = np.zeros(num_periods)
        self.bau_path = np.zeros(num_periods)
        self.bau_path[0] = self.ghg_start
        self.emission_by_decisions[0] = self.emission_by_time(tree.decision_times[0])
        period_len = tree.decision_times[1:] - tree.decision_times[:-1]

        for n in range(1, num_periods):
            self.emission_by_decisions[n] = self.emission_by_time(tree.decision_times[n])
            self.emission_per_period[n] = period_len[n] * (self.emission_by_decisions[n-1:n].mean())

        #the total increase in ghg level of 600 (from 400 to 1000) in the bau path is allocated over time
        self.emission_to_ghg = (self.ghg_end - self.ghg_start) * self.emission_per_period / self.emission_per_period.sum()
        self.emission_to_bau = self.emission_to_ghg[-1] / self.emission_per_period[-1]
        for n in range(1, num_periods):
            self.bau_path[n] = self.bau_path[n-1] + self.emission_per_period[n]*self.emission_to_bau