diff --git a/papers/exceeding_snow_loads/check_mle_convergence_for_trends/qqplot/plot_qqplot.py b/papers/exceeding_snow_loads/check_mle_convergence_for_trends/qqplot/plot_qqplot.py
index 5e862f5527a7315a8005810b12d49b2d67f435e4..e9ec74bd310c4bc72e189d475fa05dd45f156364 100644
--- a/papers/exceeding_snow_loads/check_mle_convergence_for_trends/qqplot/plot_qqplot.py
+++ b/papers/exceeding_snow_loads/check_mle_convergence_for_trends/qqplot/plot_qqplot.py
@@ -3,6 +3,7 @@ from typing import Dict
 
 import matplotlib.pyplot as plt
 import numpy as np
+import pandas as pd
 from matplotlib.ticker import PercentFormatter
 
 from experiment.meteo_france_data.scm_models_data.crocus.crocus import CrocusSnowLoadTotal
@@ -157,6 +158,14 @@ def last_quantile(psnow):
     return standard_gumbel.quantile(last_proba)
 
 
+def non_stationarity_psnow(altitude_to_visualizer):
+    all_relative_change_in_psnow = []
+    for a, v in altitude_to_visualizer.items():
+        all_relative_change_in_psnow.extend(list(v.massif_name_to_relative_change_in_psnow.values()))
+    s = pd.Series(all_relative_change_in_psnow)
+    print(s.describe())
+
+
 if __name__ == '__main__':
     """
     Worst examples:
@@ -172,7 +181,7 @@ if __name__ == '__main__':
     300 Haut_Var-Haut_Verdon 0.8446498197950775
 
     """
-    altitudes = [300, 600, 900, 1200, 1500, 1800][:-2]
+    altitudes = [300, 600, 900, 1200, 1500, 1800][:]
     # altitudes = ALL_ALTITUDES_WITHOUT_NAN
     # altitudes = [900, 1800, 2700]
     altitude_to_visualizer = {altitude: StudyVisualizerForNonStationaryTrends(CrocusSnowLoadTotal(altitude=altitude),
@@ -183,7 +192,8 @@ if __name__ == '__main__':
 
     # plot_qqplot_wrt_standard_gumbel(altitude_to_visualizer)
     # plot_hist_psnow(altitude_to_visualizer)
-    plot_exceedance_psnow(altitude_to_visualizer)
+    # plot_exceedance_psnow(altitude_to_visualizer)
+    non_stationarity_psnow(altitude_to_visualizer)
 
     # plot_qqplot_for_time_series_examples(altitude_to_visualizer)
     # plot_intensity_against_gumbel_quantile_for_time_series_with_missing_zeros(altitude_to_visualizer, nb_worst_examples=3)
diff --git a/papers/exceeding_snow_loads/study_visualizer_for_non_stationary_trends.py b/papers/exceeding_snow_loads/study_visualizer_for_non_stationary_trends.py
index 7d71e1b7a33c54be571b892697ef45878f6de42d..2b82833c377eccfdf26bacf7621983c6ed5d71d2 100644
--- a/papers/exceeding_snow_loads/study_visualizer_for_non_stationary_trends.py
+++ b/papers/exceeding_snow_loads/study_visualizer_for_non_stationary_trends.py
@@ -474,3 +474,11 @@ class StudyVisualizerForNonStationaryTrends(StudyVisualizer):
             percentage = 100 * np.array(eurocode_uncertainty.triplet) / eurocode_value
             percentages.append(percentage)
         return np.round(np.mean(percentages, axis=0))
+
+    @property
+    def massif_name_to_relative_change_in_psnow(self):
+        def compute_relative_change_in_psnow(maxima):
+            maxima_before, maxima_after = maxima[:30], maxima[30:]
+            psnow_before, psnow_after = [np.count_nonzero(s) / len(s) for s in [maxima_before, maxima_after]]
+            return 100 * (psnow_after - psnow_before) / psnow_before
+        return {m: compute_relative_change_in_psnow(self.massif_name_to_years_and_maxima[m][1]) for m in self.massifs_names_with_year_without_snow}