Reservoir Soil Density to Use for Conversion SWAT Sedimentation Study

[Jonathan Serrano]

Good day to all!

I would just like to ask help here on what density to use to convert accumulated sediment volume. The only data I have is bathymetric data results (stage-storage table) with reservoir volume capacity in MCM. My study is to model sedimentation rates/accumulated sediment and I'm estimating the loss in reservoir capacity to be equal to deposited sediment. However, SWAT output viewer has metric tons for sed_out and sed_in results while I have MCM observed data. At first I tried using the SOL_BD in the FAO soil database (about 1.3 Mg/m3 which I think is also equal to 1 ton/m3?) but I find it releases a much higher value than what Im able to simulate. I've read somewhere here that a fine value for clayey soils would be 0.5. My soils' texture are mostly clay loam or loam but when I check the percentages it has more sand.

I'm not sure too if I should use the moist bulk density (the one in SOL_BD) or dry bulk density (which I think is much lower). Any help would be much appreciated! Thank you. 

Best Regards,

Jonathan

[Jim Almendinger]

I think what I wrote back in 2009 still stands: 

On Monday, April 27, 2009 at 11:17:15 PM UTC+8 Jim Almendinger wrote:

A fair number to use for the bulk density of fine-grained (silty to clayey) reservoir sediment is about 0.5 g/mL, which equals 0.5 metric tons per cubic meter.  Sandy deltaic sediments can be appreciably more dense; sediment that has been subaerially exposed, dried, and compacted can also be denser.

I could add that organic-rich lake sediment (gyttja) can be much lower in bulk density. 

Sediment bulk density is usually expressed as (dry mass) / (wet volume).  The mass of the water is not included.  When SWAT gives you the mass of sediment trapped in a reservoir (sed_in minus sed_out), it is as dry mass.  To convert that mass to a volume you need a bulk density expressed as dry mass per wet (or "fresh") volume, for which I suggest that 0.5 g/mL is a reasonable estimate for reservoir sediment [g/mL = g/cc.  I.e., a milliliter (mL) equals a cubic centimeter (cc)]. 

Keep in mind that reservoir sediment is not soil.  It is (generally) eroded soil that has been winnowed so that the grain size distribution has been altered.  Coarse grains are trapped preferentially along the flow path, and at the very least concentrated in the deltaic sediment at the head of a reservoir.  Most of the rest of the reservoir area would then receive the remaining fine grains (silts and clays), minus the fine grained sediment that remains suspended long enough to exit the reservoir at the downstream end. 

You should look up a reference for the bulk density you use.  The number I give (0.5 g/mL) is from casually reviewing the sediment analyses our lab did over many years, but you really need a citable value.  Many of our publications give values of sediment mass/area, or whole-basin accumulation rates.  While bulk density was calculated, the results were often buried in lab analysis results, where we were more interested in sediment chronology and total mass, rather than volume.  There have been a whole lot of studies published on sediment accumulation in reservoirs -- you should be able to find something.  Hopefully the number I gave is "reasonable."  There are also decades of paleoecological literature, where lake sediments were cored to determine changes in both biota and catchment inputs, but much of this work across northern Europe and North America was from natural lakes with organic rich sediment, which is commonly somewhat lighter (lower bulk density) than reservoir sediment. 

Good luck.

-- Jim

--
You received this message because you are subscribed to the Google Groups "ArcSWAT" group.
To unsubscribe from this group and stop receiving emails from it, send an email to arcswat+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/arcswat/93544ebb-9cd3-419d-bcb7-f5329376fad3n%40googlegroups.com.

[Jonathan Serrano]

Thank you very much Dr. Jim for this! This clarifies a lot. My apologies too for the late response.

Best regards,

Jonathan