hopefully fixed the images not rendering

MarkMelotto · MarkMelotto · commit 3d5add312c7c · 2025-04-07T11:43:53.000+02:00
diff --git a/book/thesis_projects/BSc/2025_Q3_IschaHollemans_CEG/Report/2_literature_study.md b/book/thesis_projects/BSc/2025_Q3_IschaHollemans_CEG/Report/2_literature_study.md
@@ -10,7 +10,7 @@ of the river, the mean discharge in January is around 1,800 m³/s, while in Augu
 difference in mean monthly flow between January and August. The average monthly flow in January is
 approximately 560 m³/s, whereas in August, it drops to around 115 m³/s.
 
-![figure2](figures/figure2.png)
+![figure2](figures/figure2.PNG)
 *Figure 2: Monthly mean discharge for January and August to show the difference in river flow during low
 precipitation month (August), and high precipitation month (January).*
 
diff --git a/book/thesis_projects/BSc/2025_Q3_IschaHollemans_CEG/Report/3_historical_droughts.md b/book/thesis_projects/BSc/2025_Q3_IschaHollemans_CEG/Report/3_historical_droughts.md
@@ -44,7 +44,7 @@ $$ D_{max} = \text{max} \left( |D_{cum,list(t)}| \right) \text{,   } t \in [1,n]
 The drought period is defined by the amount of time it takes for the system to replenish the amount of
 lost water. In figure 3, a visualisation of the length of a drought $T_{drought}$, and $D_{max}$ is displayed:
 
-![figure4](figures/figure4.png)
+![figure4](figures/figure4.PNG)
 *Figure 4: Visualisation of 'Drought Analyser' algorithm. This figure shows how the algorithm detects the
 beginning of a drought using the $Q_{crit}$, and how the deficit and replenishment periods are defined to calculate
 the $D_{max}$.*
@@ -58,7 +58,7 @@ detected at Montjean, which is 200 km downstream of Blois. The algorithm is used
 slight differences are expected due to inflow of tributaries between Montjean and Blois. The results are
 shown in figure 5.
 
-![figure5](figures/figure5.png)
+![figure5](figures/figure5.PNG)
 *Figure 5: Droughts in the period of 1975-2022*
 
 After rearranging the results, based on severity, the algorithm gives the following output for the five
@@ -104,7 +104,7 @@ catchment. During calibration, 2000 sets of parameters are generated using ‘La
 After running the calibration, the optimal modelled discharge is then validated. The results of modelled
 discharge and the observed discharge in the validation period is displayed in figure 6.
 
-![figure6](figures/figure6.png)
+![figure6](figures/figure6.PNG)
 *Figure 6: Visual validation of the modelled discharge for the period 2015-2019. In this graph, it is visible
 that the modelled discharge is correctly simulated for low water flows.*
 
@@ -114,15 +114,15 @@ not of interest.
 The overview of figure 7 displays the droughts for the calibration and validation period. Also, the relation
 between duration and deficit is plotted using a fitted first-degree polynomial.
 
-![figure7](figures/figure7.png)
+![figure7](figures/figure7.PNG)
 *Figure 7: The relationship between droughts duration and deficit to validate the model using a first-degree
 polynomial. The model slightly overestimates the extreme droughts for both calibration and validation.*
 
 It is visible that model correctly detects the droughts in calibration and validation, yet there is a slight
 overestimation of drought severity and duration, especially for the validation period. The cause of this overestimation can be linked to the smaller size of data for this period. This is also visible in the
 distribution overview in figure 8.
 
-![figure8](figures/figure8.png)
+![figure8](figures/figure8.PNG)
 *Figure 8: Distribution overview for calibration and validation. The left side of this overview displays the
 distributions for the calibration period, and the right side displays the validation period.*
 
diff --git a/book/thesis_projects/BSc/2025_Q3_IschaHollemans_CEG/Report/4_future_droughts.md b/book/thesis_projects/BSc/2025_Q3_IschaHollemans_CEG/Report/4_future_droughts.md
@@ -34,7 +34,7 @@ duration and deficit. Yet, these methods did not clarify the results, so another
 appendix B for these results). For this purpose, the cumulative distribution is used as it better displays
 the differences between the chosen forcings. The results are visible in figure 9.
 
-![figure9](figures/figure9.png)
+![figure9](figures/figure9.PNG)
 *Figure 9: Cumulative distributions for Duration and Deficit. The graphs display the impact of the different
 scenarios. The distribution will go faster and further to the right, the more extreme droughts occur. This is
 especially the case for ‘SSP245’ and SSP585’.*
@@ -57,7 +57,7 @@ duration and $F_S(s)$ for drought severity (deficit).
 To make sure that the modelled return periods are precise, the historical CMIP6 droughts are validated
 using the observed past droughts. The validation is displayed in figure 10:
 
-![figure10](figures/figure10.png)
+![figure10](figures/figure10.PNG)
 *Figure 10: Validation of the return periods for 1942-2014. The graphs show a significant discrepancy for
 both drought duration and deficit.*
 
@@ -73,7 +73,7 @@ A better alternative is to define a correction factor based on the calibration p
 more recent droughts are represented correctly and the more extreme droughts before 1990 are
 adjusted on more recent data, making them more applicable. 
 
-![figure11](figures/figure11.png)
+![figure11](figures/figure11.PNG)
 *Figure 11: Correction of return period based on 1990-2014. For drought duration there was no correction
 factor needed. For drought deficit, a correction factor of 2.2 is applied.*
 
diff --git a/book/thesis_projects/BSc/2025_Q3_IschaHollemans_CEG/Report/5_analysing_results.md b/book/thesis_projects/BSc/2025_Q3_IschaHollemans_CEG/Report/5_analysing_results.md
@@ -4,7 +4,7 @@ In this chapter the results of the historical and future return periods are anal
 for each scenario are compared to the historical return periods, to conduct the impact of each climate
 scenario on droughts. 
 
-![figure12](figures/figure12.png)
+![figure12](figures/figure12.PNG)
 *Figure 12: Return period for duration and deficit for all scenarios using the return period equations denoted
 in formula 5 and 6. For drought deficit, $CF = 2.2$ is applied.*
 
@@ -29,7 +29,7 @@ the return periods are summarized in the table below:
 *Table 4: Overview of return periods for duration and deficit for each scenario. The increase is calculated
 based on the historical droughts.*
 
-![table4](figures/table4.png)
+![table4](figures/table4.PNG)
 
 The results indicate that droughts with a 10-year return period are projected to become significantly
 more severe. For ‘SSP585’ this can go up to an increase of 236% in drought duration days. For 50 and
