Derek's Talk, mostly on: J. Phipps Morgan, A. Harding, J. Orcutt, G. Kent, and Y. J. Chen, "An Observational and Theoretical Synthesis of Magma chamber Geometry and Crustal Genesis along a Mid-ocean Ridge Spreading Center", in Magmatic Systems, pp. 139-178, 1994.

I. Background 

A. Previous View of Melt Lenses (figure of big magma chamber)
B. Quick Summary of Seismic Evidence to the Contrary (from PM and Chen)

1) There is an Axial Magma Lens at the Base of the Sheeted Dike Complex that Contains a Large (> 25%) Fraction of Melt.

a. The Magma Lens is Usually a Continuous Feature along the Axis of a Fast-SpreadingWide
b. The Magma Lens is Thin: ~50-200 meters thick.
c. The Magma Lens is (Usually) Narrow ~ 1 km Wide.
d. Sometimes the Magma Lens is Offset from the Ridge Axis--Then It is wider (2-4 km Wide) and Slopes up toward the Ridge.
e. Synthetic Experiments can Reproduce these Basic Magma Lens Features Quite Well

 2) Other Seismic constrains on crustal Accretion at a Fast-Spreading Ridge

a. Crustal Accretion and Moho Formation Occur in a Narrow Zone about the Ridge Axis
b. The Magma Lens is Underlain by a Broader Low Velocity Zone of Hot Rock and Possible Small Amounts of Melt (<3-5%).

3) "Hot Zone" beneath the Magma Lens is Needed to Support the Axis Topography at Fast-Spreading Ridges.

 II. Motivation/Intro. to Phipps Morgan and Chen Model.

 Seismic reflection experiments observe a melt lens at a depth of around 1500 m in fast spreading ridges. However, for ridges with a half spreading rate less than about 25 mm/yr, no melt lens is seen. Since the time-averaged volume of magma injection is proportional to spreading rate, it seems likely that fast spreading ridges have quite different thermal structures from slow-spreading ridges. If melt ponding is primarily caused by decompaction of the surrounding matrix as the freezing horizon is approached [Sparks and Parmentier, etc.], which can be approximated by the location of a particular isotherm, then the location of melt should vary widely with the thermal structure and thus the spreading rate. To explore the effects on ridge flow when melt lens location is defined by an isotherm, the PM Chen model was constructed.
 

III. Boundary Conditions and Assumptions

A. Type of Model: Thermal and Mechanical Model (not geochemical)
B. Primary Assumption: Freezing horizon dictates depth of melt lens..
C. Boundary Conditions

1) Melt Emplacement

a) Magma emplacement above the freezing horizon occurs at a 250 m wide "dike" centered at the axis.
b) Magma emplacement below the freezing line occurs in a melt lens located directly below the horizon. Lens is 1km wide and 250 m thick.
c) Melt flow to lens is not modeled.

2) Geometry (see figure 15)

D. Assumptions

1) Hydrothermal Cooling

a) Significant Hydrothermal Cooling Exists above 600 deg C solidus. but never deeper than 6 km.
b) Hydrothermal Cooling defined by Nusselt #

2) Newtonian Flow

E. Neglected features

1) Melt Emplacement
a) Melt rise to lens
2) Cooling

a) by heat transferred to the surface by pillow basalts, implies isotherms in model will be a little too shallow
b) variations in Nu with spreading rate (considered likely by PM and Chen) are ignored.
c) actual values of Nu poorly known.

3) All geochemistry (fractionation and diffentiation)
4) Bouyancy effects on flow are ignored.

IV. Behavior of Model

A. Iso-strain lines mimic observed ophiolite gabbro layering.

1) Models in which melt intrusion is constant with depth do not produce this type of effect.
2) Models in which intrusion is only in the lower crust do not produce this effect.

B. Melt Lens exists for spreading half-rates of > 20-30 mm/yr. ( Figure 9)

1) For slow spreading rates, the 1200 deg. isotherm (defining the location of the melt lens in this model) is pushed below the moho by hydrothermal cooling. Varying spreading rate and melt lens geometry in the model shows that thermal structure of the crust seems to be primarily dependant upon the amount of melt injected (spreading rate), and only secondarily dependant on melt lens volume.
2) For certain vaulues of spreading rate, melt lens depth varies widely with crustal thickness.

C. Crustal Strength

1) Crustal strength increases considerably for a half-spreading rate < 20 mm/yr. Since a strong crust is thought necessity for producing a medial valley [Tapponier and Francheteau, 1978] then, as is observed, slow spreading ridges should have medial valleys. (Figures 18 C & D)

 

V. Summary of Predictions

A. Ridge structure controlled by thermal balance between melt intrustion and hydrothermal cooling.
B. melt lens existance dependent on spreading rate.
C. median valley topography controlled by lithospheric strength/thickness, which is controlled by temperature. Thus, topography indicates thermal structure.
D. crustal strain, not cumulate layering, generates the structure seen in the lower crustal portions
 

VI. Failures of Model

A. Rob finds a melt pool at the Moho, this is not explained
B. Geochemistry
C.  Nu # poorly constrained and probably dependant on spreading rate.
D.  Depth of hydrothermal cooling unknown.
D.  Your complaint here.