Download source code: geothermal_climsig_wrong.py
"""
Geothermal: Climate signal: What happens when assuming a climate change is
linear, when in fact it was abrupt?
"""
import numpy
from fatiando import utils
from fatiando.geothermal.climsig import abrupt, SingleChange
from fatiando.vis import mpl
# Generating synthetic data using an ABRUPT model
amp = 3
age = 54
# along a well at these depths
zp = numpy.arange(0, 100, 1)
temp, error = utils.contaminate(abrupt(amp, age, zp), 0.02,
percent=True, return_stddev=True)
# Preparing for the inversion
data = SingleChange(temp, zp, mode='linear').config('levmarq', initial=[1, 1])
amp_, age_ = data.fit().estimate_
print "Trying to invert an abrupt change as linear"
print " true: amp=%.3f age=%.3f" % (amp, age)
print " estimated: amp=%.3f age=%.3f" % (amp_, age_)
mpl.figure(figsize=(4, 5))
mpl.title("Residual well temperature")
mpl.plot(temp, zp, 'ok', label='Observed')
mpl.plot(data.predicted(), zp, '--r', linewidth=3, label='Predicted')
mpl.legend(loc='lower right', numpoints=1)
mpl.xlabel("Temperature (C)")
mpl.ylabel("Z (m)")
mpl.ylim(100, 0)
mpl.show()