Digital Database of Modern Evaporites and their Predicted Distribution Based on Results...
157
was initialized with a salinity of 175 (equivalent to 1175 kg/m3 = 1000 kg H 2 0 +
175 kg total dissolved salts), to approximate the onset of gypsum precipitation,
and the other with an initial salinity of 385 (equivalent to 1385 kg/m3 = 1000 kg
H 2 0 + 385 kg total dissolved salts) to approximate the onset of halite precipitation (Warren 1988). The rate of total atmospheric precipitation (mm/day) calculated by GENESIS (2.0) was subtracted from the rate of evaporation in each
2° x 2° grid cell as calculated by the proxy formation model. Because this model
depends entirely on AGCM results, and knows nothing about surface drainage
patterns, it can only predict the potential for evaporite precipitation from a
brine. Positive values for E-P indicated the potential for gypsum precipitation
(Fig. 4) when the proxy formation model was initialized with a salinity of 175.
Similarly, positive values of E-P indicated potential for halite precipitation (Fig.
5) when the proxy formation model was initialized with a salinity of 385.
Figure 4 shows the results of the PFM initialized for gypsum (S=175). The
contours in Fig. 4A show where the model predicts the potential for gypsum to
precipitate based on mean annual climate (12 months of forcing). The greatest
values of E-P on the contours indicate regions with the highest potential. The
PFM predicts the potential for the evaporite to precipitate based only on climate,
the other major factor, land surface morphology, is not considered. The PFM
predicted almost all of the observed evaporites between 30 0 N and 30 0 S latitude
using 12 months of climate forcing (Fig. 4A). However, the model did not do a
very good job predicting those evaporites that lie north of 30 0 N or south of 30 0 S
latitude.
It then occurred to us that because the majority of modern evaporites occur
in closed basins or in areas with no surface runoff, we should investigate seasonal patterns of E- P. This makes sense because even if the annual average of E-P is
too small to indicate the potential for evaporite precipitation, the fact that those
conditions occur over a closed drainage area means that the water will eventually evaporate and potentially precipitate evaporite minerals. Fig. 4B is the PFM
driven by three months of forcing for June, July, and August OJA; Northern
Hemisphere summer) from the GENESIS (2.0) simulation. In this simulation,
where E-P indicates the potential for gypsum formation in the Northern Hemisphere, we find that all of the Northern Hemisphere evaporites were predicted
except for those farthest south in Mexico and those farthest north in North
America (Nevada and Utah, USA). The Mexican evaporites are predicted by the
PFM driven by 12 months of climate data. Those in Nevada and Utah are not
predicted, probably because of the inability of the AGCM to adequately resolve
mountains to the west resulting in a decreased rainshadow-effect and too high
precipitation over this area.
The PFM for evaporites initialized for gypsum and driven by Southern Hemisphere summer (DJF; December, January, February) 3-month climate is shown
in Fig. 4C. Note the strong seasonality of the PFM prediction in western and
southern South America. None of these evaporites were predicted for the JJA
forcing (Fig. 4B), and all of the South American evaporites were predicted in the
DJF case (Fig. 4C). The greatest potential for gypsum precipitation during the
157
was initialized with a salinity of 175 (equivalent to 1175 kg/m3 = 1000 kg H 2 0 +
175 kg total dissolved salts), to approximate the onset of gypsum precipitation,
and the other with an initial salinity of 385 (equivalent to 1385 kg/m3 = 1000 kg
H 2 0 + 385 kg total dissolved salts) to approximate the onset of halite precipitation (Warren 1988). The rate of total atmospheric precipitation (mm/day) calculated by GENESIS (2.0) was subtracted from the rate of evaporation in each
2° x 2° grid cell as calculated by the proxy formation model. Because this model
depends entirely on AGCM results, and knows nothing about surface drainage
patterns, it can only predict the potential for evaporite precipitation from a
brine. Positive values for E-P indicated the potential for gypsum precipitation
(Fig. 4) when the proxy formation model was initialized with a salinity of 175.
Similarly, positive values of E-P indicated potential for halite precipitation (Fig.
5) when the proxy formation model was initialized with a salinity of 385.
Figure 4 shows the results of the PFM initialized for gypsum (S=175). The
contours in Fig. 4A show where the model predicts the potential for gypsum to
precipitate based on mean annual climate (12 months of forcing). The greatest
values of E-P on the contours indicate regions with the highest potential. The
PFM predicts the potential for the evaporite to precipitate based only on climate,
the other major factor, land surface morphology, is not considered. The PFM
predicted almost all of the observed evaporites between 30 0 N and 30 0 S latitude
using 12 months of climate forcing (Fig. 4A). However, the model did not do a
very good job predicting those evaporites that lie north of 30 0 N or south of 30 0 S
latitude.
It then occurred to us that because the majority of modern evaporites occur
in closed basins or in areas with no surface runoff, we should investigate seasonal patterns of E- P. This makes sense because even if the annual average of E-P is
too small to indicate the potential for evaporite precipitation, the fact that those
conditions occur over a closed drainage area means that the water will eventually evaporate and potentially precipitate evaporite minerals. Fig. 4B is the PFM
driven by three months of forcing for June, July, and August OJA; Northern
Hemisphere summer) from the GENESIS (2.0) simulation. In this simulation,
where E-P indicates the potential for gypsum formation in the Northern Hemisphere, we find that all of the Northern Hemisphere evaporites were predicted
except for those farthest south in Mexico and those farthest north in North
America (Nevada and Utah, USA). The Mexican evaporites are predicted by the
PFM driven by 12 months of climate data. Those in Nevada and Utah are not
predicted, probably because of the inability of the AGCM to adequately resolve
mountains to the west resulting in a decreased rainshadow-effect and too high
precipitation over this area.
The PFM for evaporites initialized for gypsum and driven by Southern Hemisphere summer (DJF; December, January, February) 3-month climate is shown
in Fig. 4C. Note the strong seasonality of the PFM prediction in western and
southern South America. None of these evaporites were predicted for the JJA
forcing (Fig. 4B), and all of the South American evaporites were predicted in the
DJF case (Fig. 4C). The greatest potential for gypsum precipitation during the