deletions | additions       diff --git a/Becancour like1.tex b/Becancour like1.tex  index 7388990..6b43fd4 100644  --- a/Becancour like1.tex  +++ b/Becancour like1.tex 
...
\subsubsection{Bécancour-like scenario}  Figure \ref{fig:seismbec} shows the corridor stack for 7 offsets. For each corridor stack, black traces refer to the baseline survey and red traces refer to CO$_2$ injected scenario. The effect of CO$_2$ is mainly observed by a travel-time delay. Indeed, a travel-time delay due to the CO$_2$ injection is well detected starting at 550 ms, (due to heterogeneity in the reservoir that produces a reflection) and at the bottom at the reservoir. Other authors reported similar observations \citep{Yang2014,Arts2004}. We can also notice an AVO anomaly with an amplitude increase with the offset at the basement reflector. Since the saturation and the CO$_2$ plume shape do not evolve too much in the time-lapse modeling, we do not have significant variation between repeated monitoring. monitoring.\\  Figure \ref{fig:stochvsblock} shows the comparison of the seismograms obtained from a stochastic geological model (left) versus a blocky model (right) for a 5 years time-lapse. As expected, the lack of details in the blocky model is reflected in the seismograms. For example, the time-delay at 550 ms is not observable in the blocky seismogram and the CO$_2$ effect is only detected by an amplitude variation. variation.\\  Moreover, at the Potsdam bottom reflector, the blocky model shows an amplitude decrease with offset. However, for the stochastic seismogram, the same reflector shows an increasing amplitude with offset. This is confirmed by the Zoeppritz analysis (Figure \ref{fig:zoeppritz}) that shows how the reflection coefficient change with angle of incidence.