Commit 031bc630764497863326aff732eb4876fe45dbc2

Authored by Arthanor
1 parent 707d962b
Exists in BGCM

Notes for the BGC modeling section

Showing 1 changed file with 72 additions and 1 deletions   Show diff stats
README.md
... ... @@ -114,4 +114,75 @@ Jean Chavy:
114 114  
115 115 Essi Aboyo:
116 116 - Dynamics of the NOW polynia
117   -- Idealized model for climate change
118 117 \ No newline at end of file
  118 +- Idealized model for climate change
  119 +
  120 +## BGCM
  121 +
  122 +Plankton: "drifters" organisms who drift with currents
  123 +Phytoplankton: Plankton that can perform photosynthesis using chlorophyll-a
  124 +
  125 +### Biological Pump
  126 +C02 + H20 -> Glucose + O2
  127 +
  128 +Processes that pump atmospheric C2 into the ocean and eventually into the deep ocean
  129 +
  130 +Link between biology and climate
  131 +CO2 -> Phytoplankton -> Foodweb -> POC -> CO2 in the deep ocea
  132 +Physical pump: Through ventilation -> Northern Atlantic & Southern Ocean
  133 +
  134 +Model definition: schematic representation of processes following a reasoned approach
  135 +
  136 +Given a selection of data, pick your model? Based on understanding? (Akaike) Information Criterion?
  137 +
  138 +Sverdrup Critical Depth: Depth at which phytoplankton losses overcome gains. You should only find (happy) phytoplankton above.
  139 +
  140 +Generally, need mixing depth < critical depth to have thriving phytoplankton. BUT, it will get nutrient depleted so you need deeper mixing some times.
  141 +
  142 +### Statistical model
  143 +Using:
  144 +- Degree days
  145 +- Southern Anomaly
  146 +- ENSO
  147 +
  148 +Best predictor: Degree days
  149 +
  150 +Logistics curve: dN/dt = rN(1-N/K)
  151 +r = b - m
  152 +b = birth
  153 +m = mortality
  154 +
  155 +Predator-Prey model
  156 +dN/dt = rN(1-N/K) - aNP/(C+N)
  157 +dP/dt = baNP/(c+N) - mP
  158 +1 - more prey
  159 +2 - more food/predator
  160 +3 - more predator
  161 +4 - more prey being eaten
  162 +5 - less prey
  163 +6 - less food/predator
  164 +7 - less predator
  165 +8 - less prey being eaten
  166 +9 - back to #1
  167 +
  168 +The keep building more complicated things and instead of considering that they're all together (0D), put them in a 1-2-3D model!
  169 +
  170 +### Dany's part
  171 +
  172 +There is rarely a model ready to fit all you need.
  173 +
  174 +First chose the boxes: Who's in your small pool
  175 +Then draw the arrows: How do they interract
  176 +Then pick a currency: Carbon, nitrate, chlorophyll, etc. Maybe more then one if you know the exchange rate
  177 +Figure out the fluxes and write down the equations
  178 +
  179 +NetLogo: simulate particles and define your own rules of interactions
  180 +
  181 +Chaos is
  182 +- deterministic
  183 +- nonlinear
  184 +- >= 3D
  185 +- sensitive to initial conditions
  186 +
  187 +Sochastic systems are not chaotic
  188 +
  189 +Ergodicity: Long average in space is equivalent to long average in time. Ex.: Waves (you can average sea level over a transect of length L, or you equivalently could sit still and average what you encounter over a certain time period equal to T = L/c_p)
119 190 \ No newline at end of file
... ...