MEDUSA-DOCU-:-Rivers.md

@@ -105,4 +105,24 @@ first runs did not work because of something in the newly made river input files
   * I've tried 1 run with all Andrew's new LS200 files (let's call it `LS200`) 
   * and also 1run where i've switched `DIC` and `ALK` to the widerly spreaded LS400 (so let's call it `LS400`).      
   Both failed. `LS400` few days later though.      
+  An interesting point is that the model ran with no river nutrients, in both advection scheme. What means that something in the river induces the crash.       
+  The DIC field in the `no_river` run (fct advection) looks like this :      
 
+<img src="uploads/681ad9fee40424c0bf4401cb1a9d81e4/image.png" width=49%>            
+
+  The difference with here is the diff with respectively `LS200` and `LS400` bellow.             
+
+<img src="uploads/5a6d4c51dd7da71bf1d6e443cfac97de/image.png" width=49%>   
+<img src="uploads/5ffeab748e149f99aeb8942cf20a7dbd/image.png" width=49%>  
+              
+<img src="uploads/0583ccd90f5e2fb294626a0adc958ddc/image.png" width=49%>   
+<img src="uploads/b4ace1fc3265be5248ea5b1050d2aec2/image.png" width=49%>           
+  The difference decrease in `LS400`, but the interesting bit here, is that we loose DIC compare to the `no-river` run, although we add the river nutrients...         
+   Looking at the river, we might understand what's going on :     
+   Here `LS200`:    
+       
+<img src="uploads/21b7054deefbdae89b3eaf6395662829/image.png" width=49%>    
+<img src="uploads/d627423cc33edc7804850527fdaa68de/image.png" width=49%>
+  
+  The problem seems to come from the mismatch between DIN and DIC riverine inputs.  PP is boosted by the N supply, where no DIC is added and where fresh water arrives. DIC is driven down by the PP, and probably bellow 0.0 because of the fresh water and advection.      
+I am doing another test with add mixing at river mouth down to 20m instead of 10, but best might be to make sure both DIN and DIC river input match.
\ No newline at end of file