Flooded areas 1- Processing of information One   of   the   main   factors   involved   in   the   flood-vegetation   association   relationship   is   the   flood   height.   It   is   possible   to   modelise   the flooded   areas   for   different   flood-heights   on   the   Mopti   flood   gauge   by   assigning   to   each   of   the   14,535   vegetation   units   its   average flood   depth.   In   such   a   model,   "intergrade   mosaics"   are   problematical.   We   therefore   decided   to   consider   that   an   intergrade   mosaic,   for example   BP/VB,   whose   constituent   associations   are   set   at   flood   level   7   for   BP   (flood   depth   between   4   m   and   2.8   m)   and   at   level   6   for VB   (flood   depth   (between   2.8   m   and   1.5   m),   was   to   be   assigned   a   composite   level   76   (flood   depth   between   4   m   and   1.5   m).   In   the calculation   of   flooded   areas,   it   is   always   possible   to   “decompose”   the   composite   levels   by   allocating   a   share   of   the   area   to   each   of   the component   levels.   Thus   for   a   mosaic   at   level   76   such   as   BP/VB,   we   can   by   convention   allocate   half   of   the   area   of   the   unit   to   BP, therefore   at   level   7,   and   the   other   half   to VB,   at   level   6.   If   the   mosaic   spans   over   a   stronger   gradient,   for   example   a   mosaic   at   level   53, a   third   of   the   area   is   attributed   to   level   5,   another   third   to   level   4   and   the   last   third   to   level   3,   since   an   area   bearing   a   mosaic   at   level 53   cannot   physically   pass   from   level   3   to   level   5   without   having   a   share   of   its   area   in   intermediate   level   4.   However,   while   it   is possible   to   decompose   the   mosaic   composition   by   following   these   set   rules   in   a   somewhat   arbitrary   but   plausible   way,   it   is   very difficult   or   even   impossible   to   assign   a   precise   spatial   location   to   each   of   the   components   (see   the   method   described   in   detail   in Table 1A and summarized in Table 1, which can be downloaded) Table 1: Flooded areas by levels after mosaic decomposition The total area of the study is 2,229,950 ha The   preceding   table   establishes   a   first   "model"   of   potentially   floodable   areas   in   relation   to   different   levels   of   flood   depth.   We must   first   define   the   expression   "potentially   floodable   areas"   we   have   chosen   to   use   instead   of   the   simpler   "flooded   areas".   This comes   from   the   fact   that,   in   the   model,   when   each   of   the   conditions   of   the   flood   depth   –   namely   flood   levels   other   than   level   1   –      is successively   reached,   all   the   component   vegetation   associations   characterised   by   that   “average”   flooding   level   are   assumed   to   be flooded.   However,   we   can   only   refer   to   potential   flooding   because   the   model   treats   each   vegetation   unit   as   an   independent   entity. As   a result,   the   effects   of   topographic   thresholds   which   would   prevent   a   basin   from   being   flooded,   even   if   the   flood   levels   corresponding to   the   vegetation   associations   it   contains   are   reached,   are   not   and   cannot   be   taken   into   account.   One   can   only   think   that   the   presence of   these   specific   vegetation   associations   at   this   precise   location   indicates   that   the   spot   is   usually   flooded   under   the   conditions described by the model, though without any certainty. Besides,   the   relationship   between   vegetation   associations   and   flood   height   is   based   on   a   single   flood   gauge:   that   of   Mopti,   which serves   as   a   reference.   This   assumes   that   the   so-called   reference   flood   is   also   valid   for   the   other   gauges   in   the   Delta:   Ke   Macina   at   the entry   of   the   Niger   River   into   the   Delta,   Beneny   Kegni   or   Sofara   on   the   Bani   River,   and Akka   at   the   exit   of   the   Debo   lake.   To   define the   corresponding   reference   floods   for   these   three   gauges,   we   relied   on   the   work   of   J.P.   Lamagat   "Analyse   de   la   vitesse   de propagation   des   crues,   application   à   la   prévision   des   crues   et   des   étiages" ,   Orstom,   1983.   This   work   makes   it   possible   to   define reference   floods   for   each   of   those   three   gauges   corresponding   to   different   flood   heights   as   measured   in   Mopti,   but,   as   we   will   see later, "real" floods rarely correspond to reference floods. This   also   leads   us   to   reflect   on   the   meaning   of   the   “zero”   reference,   which   marks   the   limit   between   flooded   and   non-flooded vegetation   areas.   It   is   defined   as   the   maximum   flood   height   most   regularly   reached   and   its   correspondence   was   established   with   a flood   height   of   660   cm   at   the   Mopti   gauge.   The   relationship   between   the   “zero”   reference   and   vegetation   associations   –   under   the conditions   of   validation   of   multivariate   floristic   profile/state   of   ecological   variables   analyses   –   therefore   applies,   regardless   of   which location   is   being   considered   in   the   Delta.   However,   this   zero   flood,   other   than   at   Mopti   –   where   it   corresponds   to   an      altitude   of 267.20   m   –   as   well   as   at   the   three   other   reference   stations   (Ke   Macina,   Sofara,   Akka),   cannot   be   attached   to   a   precise   altitude everywhere   else   in   the   Delta.   To   move   from   a   relative   model,   calibrated   with   respect   to   this   “zero”,   to   a   topographic   model,   it   would first   be   necessary   to   know   the   relation   uniting   “zero”   and   altitude   at   every   single   point   of   the   Delta. As   a   first   approximation,   we   can assume   that   “zero”   represents   the   trace   in   space   of   the   surface   generated   by   the   maximum   reference   flood   wave.   This   surface   is probably   complex   corresponding   to   the   period   of   slack   between   the   end   of   the   flood   and   the   beginning   of   the   recession,   when   the slope of the flow is at its lowest. We will later see how to try to solve this problem. According   to   table   n°1,   the   potenrially   floodable   area   corresponding   to   the   reference   flood   as   defined   previously   (660cm   at   the Mopti   gauge)   covers   1,820,289   ha,   including   the   Farimaké   area   and   the   areas   initially   flooded   first   by   run-off   then   by   the   flood.   The second   lesson   to   be   derived   from   this   table   is   show   sensitive   the   Delta   proves   to   be   to   small   variations   in   the   water   heights.   Between 660   cm   and   600   cm,   potentially   flooded   areas   decrease   by   7%   to   9%   for   every   loss   of   10cm   in   water   height.   Below   600cm   at   Mopti, however,   there   seems   to   be   a   sharp   shift   in   the   regression   pattern,   with   a   less   than   3%   loss   in   flooded   areas   for   each   loss   of   10   cm   in water   height.   This   suggests   a   very   theoretical   profile   for   the   inner   Delta,   considered   as   a   single   entity   –   which   it   obviously   is   not.   Its higher   part   appears   to   have   a   weak   cross-slope,   which   makes   it   very   sensitive   to   fairly   small   variations   in   water   heights.   Beyond   level 4, however, deep basins with steeper cross-slopes are therefore less sensitive to such variations. 2- Flood mapping: the CRUE3 layer CRUE3   layer   is   derived   from   VEG4   by   copying   and   creating   specific   items:   H_0,   H_10,   H_30   ...Among   the   ecological variables,   two   were   selectively   chosen:   soil   texture,   which   has   been   briefly   dealt   with   in   the   preceding   part,   and   flood   heights   or depths. •SIGLE (fr:Sigle): is directly derived from VEG4 and matches each vegetation association with a geographical unit. •LEVEL   (fr:niveau):   concerns   the   flood   level   of   the   vegetation   association.   A   number   between   1   and   7   is   ascribed   to   each vegetation   association,   8   being   reserved   for   open   water.   (see   table   n°3   page   39 :   the   relationship   between   vegetal   associations,   water heights and the Mopti gauge measurements) The   mosaics   are   represented   by   a   two-digit   number   in   reference   to   their   component   associations.   Thus   BP/VB,   respectively belonging   to   levels   7   and   6   will   be   coded   76,   while   O/VOR   will   be   coded   55   since   its   components   both   belong   to   level   5.   For   the sake of simplicity, single vegetation associations are coded from 11 to 77, with 80 reserved for MB and 90 for open water. •HIGH   (fr:Profond):   translates   the   LEVEL   item   into   water   depth.   The   detail   of   those   heights   is   discussed   page   XXX.   Let   us   just say   that   it   corresponds   to   the   bottom   level   of   the   corresponding   water   range   for   single   vegetation   formations,   and   the   average   one   for the   mosaics.   Thus   B   will   be   given   a   depth   of   -2.80m   in   keeping   with   level   66   and   B/VOR   will   be   given   a   depth   of   -2.15   cm,   in keeping   with   level   65.   Non-flooded   vegetation   formations   (TA   to   TS)   are   conventionally   given   a   0   depth,   so   that   P/TA,   at   level   21, will be given a depth of 0.05m  (average between -0.1m and 0m) As   we   are   going   to   see,   such   conventions   required   by   cartography   tend   to   maximise   flooded   areas   by   comparison   to   table   1, which was reached by de-composing the mosaics •H_0   to   H_280:   these   fields   derive   directly   from   the   HIGH   item.   They   are   of   the   yes/no   type   and   contain   the   following numerical values: •0 when the area is not flooded under the conditions of the field H_0, H_10, etc. •1 when the area is flooded under the conditions of field H_0, or H_10 etc. •2 when the area is first flooded by run-off (PAN, PAR, PAS, PAM type formations) under field conditions H_0, H_10, etc. The same reasoning applies to items H_30, H_60, H_150, H_280. A   convention   is   set   for   mosaics   including   associations   PAN,   PAR   or   PAS,   PAM.   When   one   of   the   two   associations   is   non- flooded   as   for   PAN/TA   for   example,   the   mosaic   is   considered   first   flooded   by   run-off,   therefore   coded   2.   When   the   other association   within   the   mosaic      is   a   flooded   type   –   as   for   PAN/ZB   for   example   –   the   river   flood   prevails   on   the   run-off   flood   and the mosaic is coded 1. This scenario only concerns a very small number of polygons. For   a   flood   reaching   660   cm,   the   floodable   surfaces   occupy   almost   the   entire   Delta.   The   Farimaké   in   the   northwest   is   largely flooded   by   run-off   first,   with   the   flood   coming   in   late   (November-December.   Inside   the   basin,   the   spaces   that   remain   exposed   are mainly located in the following areas -first,   along   a   double   string   of   "toggere"   forming   an   alignment   running   parallel   to   the   main   course   of   the   Niger   river;   from Koubaye to the south (at the latitude of Mopti), this  turns into a large tree-shape area around Dialloubé, south of Lake Débo. -second, east of Djenné, the erg of Femaye, along the Bani and the highlands of southern Sébéra. -third, near Diafarabé, south of Niger and west of the Diaka, between the defluent and the western margin. For   a   flood   reaching   630   cm,   the   western   margin   of   the   flood   recedes   and   approaches   Ténenkou.   The   " Togge "   occupy   a   larger area   and   in   the   southern   part,   the   water   table   in   Djenné   is   already   visibly   beginning   to   separate   into   a   northern   basin   and   a southern basin. For   a   flood   reaching   600   cm,   the   highlands   in   the   center-east   of   the   Delta   form   a   continuous   area   from   Kouakourou   north   of Dialloubé.   The   fragmentation   of   the   water   table   which   completely   covered   the   bowl   at   660   cm   is   now   well   marked.   To   the   west, the   flood   continues   to   stretch   massively   from   Ténenkou   to   Lake   Walado;   to   the   east,   it   is   still   continuous   from   the   Bani-Niger mesopotamia,   continues   along   the   right   bank   of   the   Niger   from   Mopti   to   Konna   before   joining   Lake   Débo   in   the   north.   In   the southern   part   of   the   Delta,   the   separation   of   the   basins   to   the   right   of   Djenné   is   almost   complete   and   the   highlands   of   Diafarabé are out of water, except for a string of pools south of the river. For   a   flood   reaching   510   cm,   the   majority   of   the   Delta   basin   is   no   longer   flooded   and   the   highly   fragmented   water   surfaces occupy only the heart of the deep basins which constitute the resistant core of the inner Delta Table 2: Areas potentially flooded by levels of flood height As   a   conclusion,   the   model   we   have   sketched   allows   us   to   calculate   and   map   out   potentially   flooded   areas   for   each   class   of water-heights.   The   presence   of   intergraded   mosaics   makes   it   necessary   to   define   conventions   by   which   the   latter   are   allocated   to specific   level,   so   that   numerical   results   will   differ   between   tables   1   and   2,   with   the   cartographic   method   overestimating   the   areas concerned.   Nevertheless,   the   cartography   allows   us   to   catch   a   glimpse   of   the   way   in   which   the   inner   Delta   is   structured,   with   deep basins   (see   move   form   600   cm   to   510   cm)   whose   precise   contours,   contents      and   boundaries   cannot   be   identified.   This   matches   the conclusions   derived   from   the   analysis   of   the   map   of   vegetation   formations   which   shows   how   very   subtle   combinations   allow   the Delta’s   structure   to   display   several   distinct   vegetation   landscapes.   We   shall   attempt   to   further   establish   the   Delta’s   structure   by moving   from   a   discrete   model   to   a   continuous   one   relying   on   matrix   data,   allowing   us   to   move   to   a   3D   model   of   the   potentially flooded   areas.   We   shall   also   endeavour   to   derive   a   Digital   Elevation   Model   of   the   Delta   from   it,   after   setting   the   relevant   altitudes for the reference flood.
Flood heights ( c m)   Non - flooded Areas  (ha)   Areas floodable by  flood (ha)   Areas floodable by  runoff (ha)   660   329,640   1,742,658   157,65 2   650   452,436   1,622,200   155,314   630   737,382   1,401,238   91,330   600   1,133,880   105,665   40,405   510   1,528,074   679,626   22,250
Flooded areas
* Level 8 corresponds to water (Niger, Bani, Lakes….) **Level   2   corresponds   to   a   10   cm   layer   (0   -   10   cm),   level   3   to   two   slices   (10   -   30   cm),   level   4   to   three   layers   (30   -   60   cm),   level   5   to   nine   layers (60   -   150   cm),   level   6   to   thirteen   layers(150   -   280   cm)   and   level   7   to   twelve   layers   (280   -   400   cm).   The   height   of   380   cm   has   never   been   observed in Mopti as the maximum height of an annual flood, the lowest recorded value was 440 cm in 1984 .
Download                
Level   Correspondence  with level on the  Mopti gauge   Areas   (ha)   Cumulative  flood areas   As a% of  the total  floodable  area   Per 10 cm of  submersion**   1   > 660   409,660         2   650 - 660   1 32 , 415   1, 820,289   7.3   7.3   3   630 - 650   296,619   1, 687,875   16.3   8.1   4   600 - 630   422,284   1, 391,256   23.2   7.7   5   510 - 600   420,758   968,972   23. 1   2.6   6   380 - 510   437,783   548,214   24.1   1. 8   7   260 - 380   56, 9 33   110,4 30   3.1   0.3   8*   < 260   53,49 7   53,49 7   2.9