diff --git a/plotting/panel.R b/plotting/panel.R
index 3c601b1ec15b288396648892d941752c86565aa2..2d2cc42ea895d5fb5773a3f6907df37b3e95b116 100644
--- a/plotting/panel.R
+++ b/plotting/panel.R
@@ -1,11 +1,12 @@
 # Usefull library
 library(ggplot2)
+library(scales)
 library(qpdf)
 library(gridExtra)
 
 
 # Time panel
-panel = function (df_data, df_meta, figdir, p_threshold=0.1, filedir_opt='', filename_opt='', variable='', df_trend=NULL, unit2day=365.25, is_sqrt=FALSE) {
+panel = function (df_data, df_meta, figdir, p_threshold=0.1, filedir_opt='', filename_opt='', variable='', df_trend=NULL, unit2day=365.25, is_sqrt=FALSE, missRect=FALSE) {
 
     outfile = "Panels"
     
@@ -42,12 +43,28 @@ panel = function (df_data, df_meta, figdir, p_threshold=0.1, filedir_opt='', fil
 
         dDate = df_data_code$Date[length(df_data_code$Date)] -
             df_data_code$Date[1]
-        datebreak = round(as.numeric(dDate) / unit2day / 12 , 0)
+        datebreak = round(as.numeric(dDate) / unit2day / 11 , 0)
         
-        p = ggplot() +
+        p = ggplot() + theme_bw() +
             geom_line(aes(x=df_data_code$Date, y=df_data_code$Qm3s),
                       color='black')
 
+
+        if (missRect) {
+            NAdate = df_data_code$Date[is.na(df_data_code$Qm3s)]
+            dNAdate = diff(NAdate)
+            NAdate_Down = NAdate[append(Inf, dNAdate) != 1]
+            NAdate_Up = NAdate[append(dNAdate, Inf) != 1]
+
+
+            p = p +
+                geom_rect(aes(xmin=NAdate_Down, 
+                              ymin=0, 
+                              xmax=NAdate_Up, 
+                              ymax=max(df_data_code$Qm3s, na.rm=TRUE)*1.1),
+                          linetype=0, fill='Wheat', alpha=0.3)
+        }
+
         if (!is.null(df_trend)) {
             if (df_trend[df_trend$code == code,]$p < p_threshold) {
 
@@ -65,15 +82,52 @@ panel = function (df_data, df_meta, figdir, p_threshold=0.1, filedir_opt='', fil
         }}
         
         p = p + 
-            ggtitle(paste(variable, 'station', 
-                          as.character(code), sep=' ')) +
+            # ggtitle(paste(variable, 'station', 
+                          # as.character(code), sep=' ')) +
             ylab(expression(paste('débit [', m^{3}, '.', 
                                   s^{-1}, ']', sep=''))) +
             xlab('date') + 
-            scale_x_date(date_breaks=paste(as.character(datebreak), 'year', sep=' '), date_labels="%Y")
+            scale_x_date(date_breaks=paste(as.character(datebreak),
+                                           'year', sep=' '),
+                         date_labels="%Y",
+                         limits=c(min(df_data_code$Date), 
+                                  max(df_data_code$Date)),
+                         expand=c(0, 0)) 
+            
+            if (is_sqrt) {
+                p = p +
+                    scale_y_continuous(breaks=seq(0, 100, 10),
+                                       minor_breaks=seq(0, 100, 5),
+                                       limits=c(0,
+                                                max(df_data_code$Qm3s,
+                                                    na.rm=TRUE)*1.1),
+                                       expand=c(0, 0))
+            } else {
+                p = p +
+                    scale_y_continuous(breaks=seq(0, 10000, 1000),
+                                       minor_breaks=seq(0, 10000, 500),
+                                       limits=c(0, 
+                                                max(df_data_code$Qm3s,
+                                                    na.rm=TRUE)*1.1),
+                                       expand=c(0, 0))
+            }
+       
+        df_meta_code = df_meta[df_meta$code == code,]
+        text = textGrob(paste(
+            'station ', code, '\n',
+            'nom : ', df_meta_code$nom, '\t', 
+            'territoire : ', df_meta_code$territoire, '\n',
+            'position : (', df_meta_code$L93X, '; ', df_meta_code$L93Y, ')', '\t',
+            'surface : ', df_meta_code$surface_km2, ' km2',
+            sep=''),
+            just='left',
+            gp=gpar(col="darkgrey", fontsize=10))
+
+        plot = grid.arrange(text, p, ncol=1, nrow=2, heights=c(1/4, 3/4))
 
         # Saving
-        ggsave(path=outdirTmp,
+        ggsave(plot=plot, 
+               path=outdirTmp,
                filename=paste(as.character(code), '.pdf', sep=''),
                width=29.7, height=21, units='cm', dpi=100)
 
diff --git a/script.R b/script.R
index e02c8e289f746dd477529ef9e4994046d8f5cfc1..cc1a9da3bf92039520d367b01081a0e82247579b 100644
--- a/script.R
+++ b/script.R
@@ -139,8 +139,8 @@ df_meta = df_join$meta
 
 # TIME PANEL #
 # Plot time panel of debit by stations
-# panel(df_data, df_meta, figdir, "")
-# panel(df_data, df_meta, figdir, "", is_sqrt=TRUE)
+panel(df_data, df_meta, figdir, "", missRect=TRUE)
+# panel(df_data, df_meta, figdir, "", missRect=TRUE, is_sqrt=TRUE)
 ### /!\ Removed 185 row(s) containing missing values (geom_path) -> remove NA ###
 
 
@@ -150,12 +150,12 @@ df_meta = df_join$meta
 
 
 # QA TREND #
-res_QAtrend = get_QAtrend(df_data, period)
-panel(df_data=res_QAtrend$data, 
-      df_meta=df_meta,
-      df_trend=res_QAtrend$trend,
-      p_threshold=0.1,
-      figdir=figdir, filedir_opt='QA', variable='QA')
+# res_QAtrend = get_QAtrend(df_data, period)
+# panel(df_data=res_QAtrend$data, 
+#       df_meta=df_meta,
+#       df_trend=res_QAtrend$trend,
+#       p_threshold=0.1,
+#       figdir=figdir, filedir_opt='QA', variable='QA')
 
 # QMNA TREND #
 # res_QMNAtrend = get_QMNAtrend(df_data, period)
diff --git a/script_install.R b/script_install.R
index b520c2c24389cb7adce0edbc476ea9e40f20cde0..268717774b2c2c80f80c801804155e8aa65d82f5 100644
--- a/script_install.R
+++ b/script_install.R
@@ -8,6 +8,7 @@ install.packages("officer")
 install.packages("lubridate")
 install.packages('zoo')
 install.packages("qpdf")
+install.packages("scales") 
 
 library(devtools)
 install_github("https://github.com/benRenard/BFunk") #type '1'