julieshortridge commented 1 year ago

Task: Rough sensitivity analysis and selection of soil moisture parameters @sumanager56

Action items:
- [ ] Calculate volumetric soil moisture content for all tabs and locations/horizons in soil moisture data spreadsheet. Share results via github.
- [ ] Create function in R to plot Van Genuchten function (water retention curve) when given different values of equation parameters. Use this to gain a sense of how function responds to different parameter values. See equation 21 here: https://acsess.onlinelibrary.wiley.com/doi/10.2136/sssaj1980.03615995004400050002x
- [ ] run simulations in SWAP-WOFOST with different values of NPAR and ALFA parameters to get a sense of how changing these parameters influences soil moisture results. What happens to the soil moisture time series when you increase or decrease these parameters, in terms of both the magnitude of change (can you see a clear difference?) and the type of change (is it making values higher or lower, more or less variable, etc.).

sumanager56 commented 1 year ago

Sb_Water Retention- Wt Basis_Bulk-Density.xlsx

ryestewart commented 1 year ago

Couple of papers for van Genuchten function and parameters.

vanGenuchten-1980-Closed-form Equation for predicting hydraulic conductivity.pdf

Fuentes_1992_Parameter_Constraints_on_Closed-form_Soilwater_Relationships.pdf

sumanager56 commented 1 year ago

bulkdensity <html xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:x="urn:schemas-microsoft-com:office:excel" xmlns="http://www.w3.org/TR/REC-html40">

Sample ID | Lab ID | Horizons | Depth from top of sample, inches | % WC_volumetric | %WC_modified -- | -- | -- | -- | -- | -- | | | | | East End | JS-1 | A | 0 - 15.5" | 10.8 | 9.84 | | B | 15.5 - 29.5" | 23.75 | 21.64 | | BC | 29.5 - 35.75" | 19.3 | 17.59 | | | | | Corn Middle 2021 | JS-2 | A | 0 -15.5" | 3.43 | 3.12 | | AB | 15.5 - 21.5" | 13.53 | 12.33 | | B1 | 21.5 - 33.5" | 25.43 | 23.17 | | B2 | 33.5 - 40.5" | 22.56 | 20.56 | | | | | Corn East 2021 | JS-3 | A | 0 - 12.5" | 4.46 | 4.07 | | BA | 12.5 - 21.5" | 24.39 | 22.23 | | B1 | 21.5 - 29" | 19.69 | 17.94 | | B2 | 29 - 42" | 26.16 | 23.83 | | | | | Corn West 2021 | JS-4 | A1 | 0 - 8" | 12.78 | 11.64 | | A2 | 8 - 14.5" | 16.9 | 15.4 | | C1 | 14.5 - 25.5" | 23.41 | 21.33 | | C2 | 25.5 - 34.5" | 26.04 | 23.73 | | | | | West End | JS-5 | A1 | 0 - 6" | 4.31 | 3.93 | | A2 | 6 - 11" | 10.94 | 9.97 | | AB | 11 - 20" | 19.41 | 17.69 | | B1 | 20 - 32" | 24.11 | 21.97 | | | | | Cotton Middle 2021 | JS-6 | A | 0 -12.5" | 19.65 | 17.9 | | B1 | 12.5 - 24.5" | 23.31 | 21.24 | | B2 | 24.5 - 38.5" | 29.64 | 27.01 | | | | | Cotton East 2021 | JS-7 | A | 0 - 11" | 20.88 | 19.03 | | BA | 11 - 19" | 25.6 | 23.33 | | B1 | 19 - 27.25" | 26.58 | 24.22 | | B2 | 27.25 - 38.5" | 29.33 | 26.72 | | | | | Cotton West 2021 | JS-8 | A | 0 -14" | 19.62 | 17.87 | | BA | 14 - 22" | 23.3 | 21.23 | | B1 | 22 - 31.25" | 23.53 | 21.44

sumanager56 commented 1 year ago

Saturation <html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:x="urn:schemas-microsoft-com:office:excel" xmlns="http://www.w3.org/TR/REC-html40">

Sample ID | Lab ID | Horizons | Depth from top of sample, inches | % WC_volumetric | %WC_modified | %Porosity -- | -- | -- | -- | -- | -- | -- | | | | | | East End | JS-1 | A | 0 - 15.5" | 68.67 | 62.77 | 47.92 | | A DUP | | 58.63 | 53.59 | 47.92 | | B | 15.5 - 29.5" | 68.13 | 62.13 | 39.25 | | BC | 29.5 - 35.75" | 56.21 | 51.12 | 40.00 | | | | | | Corn Middle 2021 | JS-2 | A | 0 -15.5" | 60.98 | 55.42 | 51.70 | | AB | 15.5 - 21.5" | 48.32 | 43.93 | 36.60 | | AB DUP | | 61.41 | 55.84 | 36.60 | | B1 | 21.5 - 33.5" | 54.58 | 49.71 | 40.00 | | B2 | 33.5 - 40.5" | 68.53 | 62.31 | 39.62 | | | | | | Corn East 2021 | JS-3 | A | 0 - 12.5" | 71.67 | 65.3 | 46.08 | | BA | 12.5 - 21.5" | 59.22 | 53.96 | 39.46 | | B1 | 21.5 - 29" | 70.87 | 64.58 | 40.96 | | B2 | 29 - 42" | 54.29 | 49.47 | 46.27 | | B2 DUP | | 64.34 | 58.46 | 46.42 | | | | | | Corn West 2021 | JS-4 | A1 | 0 - 8" | 53.75 | 48.98 | 55.94 | | A2 | 8 - 14.5" | 76.12 | 69.36 | 50.02 | | C1 | 14.5 - 25.5" | 70.6 | 64.33 | 40.49 | | C2 | 25.5 - 34.5" | 77.82 | 70.91 | 39.47 | | C2 DUP | | 69.28 | 63.13 | 39.47 | | | | | | West End | JS-5 | A1 | 0 - 6" | 53.65 | 48.88 | 58.07 | | A2 | 6 - 11" | 47.41 | 43.2 | 60.70 | | AB | 11 - 20" | 65.27 | 59.47 | 44.09 | | B1 | 20 - 32" | 57.4 | 52.3 | 41.25 | | | | | | Cotton Middle 2021 | JS-6 | A | 0 -12.5" | 75.47 | 68.76 | 37.11 | | B1 | 12.5 - 24.5" | 66.76 | 60.83 | 43.16 | | B2 | 24.5 - 38.5" | 81.73 | 74.47 | 32.65 | | | | | | Cotton East 2021 | JS-7 | A | 0 - 11" | 74.1 | 67.51 | 48.54 | | BA | 11 - 19" | 64.97 | 59.19 | 41.65 | | B1 | 19 - 27.25" | 70.39 | 64.14 | 40.70 | | B2 | 27.25 - 38.5" | 69.37 | 63.2 | 42.23 | | | | | | Cotton West 2021 | JS-8 | A | 0 -14" | 64.94 | 59.17 | 39.02 | | BA | 14 - 22" | 78.76 | 71.76 | 32.34 | | B1 | 22 - 31.25" | 74.95 | 68.29 | 39.28

sumanager56 commented 1 year ago

0.33bar <html xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:x="urn:schemas-microsoft-com:office:excel" xmlns="http://www.w3.org/TR/REC-html40">

Sample ID | Lab ID | Horizons | Depth from top of sample, inches | % WC_volumetric | %WC_modified -- | -- | -- | -- | -- | -- | | | | | East End | JS-1 | A | 0 - 15.5" | 12.93 | 11.78 | | B | | 20.35 | 18.55 | | B DUP | 15.5 - 29.5" | 20.19 | 18.4 | | BC | 29.5 - 35.75" | 23.1 | 21.05 | | BC DUP | | 22.99 | 20.95 | | | | | Corn Middle 2021 | JS-2 | A | 0 -15.5" | 14.98 | 13.62 | | A DUP | | 14.48 | 13.16 | | AB | 15.5 - 21.5" | 18.51 | 16.87 | | B1 | 21.5 - 33.5" | 21.23 | 19.34 | | B2 | 33.5 - 40.5" | 24.07 | 21.93 | | | | | Corn East 2021 | JS-3 | A | 0 - 12.5" | 15.73 | 14.34 | | BA | 12.5 - 21.5" | 18.32 | 16.7 | | B1 | 21.5 - 29" | 26.71 | 24.34 | | B2 | 29 - 42" | 23.08 | 21.03 | | | | | Corn West 2021 | JS-4 | A1 | 0 - 8" | 15.47 | 14.09 | | A2 | 8 - 14.5" | 17.11 | 15.59 | | C1 | 14.5 - 25.5" | 19.83 | 18.07 | | C2 | 25.5 - 34.5" | 20.88 | 19.02 | | | | | West End | JS-5 | A1 | 0 - 6" | 12.98 | 11.83 | | A2 | 6 - 11" | 13.15 | 11.98 | | AB | 11 - 20" | 25.34 | 23.09 | | B1 | 20 - 32" | 25.75 | 23.46 | | | | | Cotton Middle 2021 | JS-6 | A | 0 -12.5" | 19.56 | 17.86 | | A DUP | | 21.65 | 19.76 | | B1 | 12.5 - 24.5" | 22.02 | 20.11 | | B1 DUP | | 21.93 | 20.03 | | B2 | 24.5 - 38.5" | 30.01 | 24.2 | | B2 DUP | | 29.53 | 23.82 | | | | | Cotton East 2021 | JS-7 | A | 0 - 11" | 18.24 | 16.57 | | A DUP | | 16.62 | 15.1 | | BA | 11 - 19" | 21.43 | 19.57 | | BA DUP | | 23.13 | 21.13 | | B1 | 19 - 27.25" | 24.72 | 22.5 | | B1 DUP | | 25.13 | 22.88 | | B2 | 27.25 - 38.5" | 24.12 | 21.96 | | B2 DUP | | 23.87 | 21.73 | | | | | Cotton West 2021 | JS-8 | A | 0 -14" | 23.5 | 21.47 | | A DUP | | 21.44 | 19.58 | | BA | 14 - 22" | 23.77 | 21.62 | | BA DUP | | 26.64 | 24.23 | | B1 | 22 - 31.25" | 23.3 | 21.25 | | B1 DUP | | 23.05 | 21.02

sumanager56 commented 1 year ago

15 bar <html xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:x="urn:schemas-microsoft-com:office:excel" xmlns="http://www.w3.org/TR/REC-html40">

Sample ID | Lab ID | Horizons | Depth from top of sample, inches | % WC_volumetric | %WC_modified -- | -- | -- | -- | -- | -- | | | | | East End | JS-1 | A | 0 - 15.5" | 2.88 | 2.62 | | B | 15.5 - 29.5" | 8.74 | 7.96 | | BC | 29.5 - 35.75" | 8.12 | 7.39 | | | | | Corn Middle 2021 | JS-2 | A | 0 -15.5" | 3.99 | 3.64 | | AB | 15.5 - 21.5" | 10.37 | 9.45 | | B1 | 21.5 - 33.5" | 13.34 | 12.16 | | B2 | 33.5 - 40.5" | 14.38 | 13.1 | | | | | Corn East 2021 | JS-3 | A | 0 - 12.5" | 4.69 | 4.28 | | BA | 12.5 - 21.5" | 10.19 | 9.28 | | B1 | 21.5 - 29" | 15.11 | 13.77 | | B2 | 29 - 42" | 11.8 | 10.75 | | | | | Corn West 2021 | JS-4 | A1 | 0 - 8" | 4.33 | 3.94 | | A2 | 8 - 14.5" | 4.08 | 3.71 | | C1 | 14.5 - 25.5" | 9.95 | 9.06 | | C2 | 25.5 - 34.5" | 5.09 | 4.63 | | | | | West End | JS-5 | A1 | 0 - 6" | 4.4 | 4.01 | | A2 | 6 - 11" | 3.86 | 3.52 | | AB | 11 - 20" | 11.51 | 10.49 | | B1 | 20 - 32" | 11.81 | 10.76 | | | | | Cotton Middle 2021 | JS-6 | A | 0 -12.5" | 5.71 | 5.2 | | B1 | 12.5 - 24.5" | 11.51 | 10.49 | | B2 | 24.5 - 38.5" | 13.38 | 12.19 | | | | | Cotton East 2021 | JS-7 | A | 0 - 11" | 5.11 | 4.66 | | BA | 11 - 19" | 13.09 | 11.93 | | B1 | 19 - 27.25" | 11.49 | 10.47 | | B2 | 27.25 - 38.5" | 8.79 | 8.01 | | | | | Cotton West 2021 | JS-8 | A | 0 -14" | 6.2 | 5.65 | | BA | 14 - 22" | 9.9 | 9.02 | | B1 | 22 - 31.25" | 9.03 | 8.23

julieshortridge commented 1 year ago

Thanks Suman. The FC and PWP values seem pretty reasonable to me. The saturation numbers look pretty high across the board, and I also notice looking at the duplicates that they seem to range pretty widely (~10% difference in results for duplicates) in the saturation samples, much more than in the FC values (~1-2% difference across duplicates). @ryestewart I'd be curious to know if that's something you see often.

It seems to me it may be worth experimenting in SWAP with using Theta_sat values equal to what we've sampled here as well as some lower values based on soil texture. The soils at the site shouldn't be saturated often, so I'd think that if we're getting the parameters that control drainage right than the exact value of theta_sat might be less important?

sumanager56 commented 1 year ago

Thanks, Dr. Shortridge, I shall look into all potential parameters. Just a quick comparison between the plot that we viewed last week and the one with an updated alpha value of 0.1058 obtained using our sample data points and van genuchten's retention curve. It was surprising that the graph didn't change much after August-might have something to do with the met data. result_1st Rplot_new_alpha

julieshortridge commented 1 year ago

Thanks Suman - when you generate these plots, can you edit them to add a text box showing the parameter values?

sumanager56 commented 1 year ago

R code to estimate Van Genuchten parameters

install.packages("remotes")

install.packages("minpack.lm")

remotes::install_github("gowusu/vadose")

library(remotes) library(vadose) library(minpack.lm) rm(list=ls(all=TRUE))

For 3 horizons in JS-1,JS-6 samples:A,B,c represents the 3 horizons

data_A <- data.frame( h = as.numeric(c("0.0001","336.5064","15295.7","0.0001","336.5064","15295.7")), Lab_ID = c("JS-1","JS-1","JS-1","JS-6","JS-6","JS-6"), theta = as.numeric(c("0.686","0.129","0.029","0.754","0.195","0.057")) ) data_B <- data.frame( h = as.numeric(c("0.0001","336.5064","15295.7","0.0001","336.5064","15295.7")), Lab_ID = c("JS-1","JS-1","JS-1","JS-6","JS-6","JS-6"), theta = as.numeric(c("0.681","0.203","0.087","0.667","0.220","0.115")) ) data_C <- data.frame( h = as.numeric(c("0.0001","336.5064","15295.7","0.0001","336.5064","15295.7")), Lab_ID = c("JS-1","JS-1","JS-1","JS-6","JS-6","JS-6"), theta = as.numeric(c("0.562","0.231","0.081","0.817","0.300","0.134")) )

write.csv(data_A,'VGdata.csv')

uses starting values of thr,ths,alp,n and predicts theta. compares it to

measured theta values at different h and minimizes the sum of squares residuals

by changing these 4 parameter values

modA=soil physical Layer1

modA <- vg(data=data_A,h="h",theta="theta",thr=0.1, ths=0.1, alp=0.1, n=1, group=NULL,Ks="nvr") modB <- vg(data=data_B,h="h",theta="theta",thr=0.1, ths=0.1, alp=0.1, n=1, group=NULL,Ks="nvr") modC <- vg(data=data_C,h="h",theta="theta",thr=0.1, ths=0.1, alp=0.1, n=1, group=NULL,Ks="nvr")

nvr=ksat(ths=model1$ths,hb=1/model1$alp,n=model1$n,model="nvr")

print(nvr)

Using rawls2006 for KS

model2 <- vg(data=data_A,h="h",theta="theta", group=NULL,Ks="rawls2006")

ABove two models provide same parameter values

plot(modA)

Creating a new function in R

van_g <- function(h,ths,thr,alp,n){ ((ths-thr)/((((alp*h)^n)+1)^(1-(1/n))))+thr }

use nls function in R to estimate fitting parameters using least squares algorith.

compares theta with van_g estimated theta.

Model <- nls(theta~van_g(h,ths,thr,alp,n),data=data_A, start=list(ths=0.1,thr=0.1,alp=0.1,n=1))

sumanager56 commented 1 year ago

Shall figure out a way to extract and just plot a few individual soil depths!! increasing alpha by one order magnitude in all physical layers compared to original value alpha_1order increasing alpha by 2 order magnitude alpha_2order increasing alpha by 3 order magnitude alpha_3order decreasing alpha by one order magnitude compared to original value alpha_minust1order decreasing alpha by two order magnitude alpha_minus2order

sumanager56 commented 1 year ago

The initial groundwater level which was previously provided at -68 cm seems to have created the trouble with persistent saturated conditions after July GWL500

sumanager56 commented 1 year ago

Couldn't get a better curve than this using larger alpha values

Can see that NPAR values are mainly related to the rate of desaturation - the curve being steeper for higher NPAR values as water content reduces from saturation (air entry pressure head) to retention level different NPAR values

The rate of desaturation almost remains constant but likely influences the air entry pressure head - soil tends to remain saturated even at higher pressure head for lower alpha values different alpha values

sumanager56 commented 1 year ago

Just figured out a way to directly extract soil moisture values for specific depths. VWCdepth

sumanager56 commented 1 year ago

Looking into soil moisture at 10 cm depth vs rainfall data combined

ryestewart commented 1 year ago

I went back to the plastic sleeves and re-measured. I think the diameter that Athena used were slightly too small, so I updated. That made the bulk densities values more reasonable. The saturated water contents are too high, but that may be because it is pretty difficult to transfer the cores from the water to the scale and not lose water (or have too much). A better approach could be to calculate porosity based on 1-bulk density/solid density, which we could assume is 2.65 g/cm3 (quartz sand).

Julie-Shortridge_Water Retention- Wt Basis_Bulk-Density_RDS.xlsx

sumanager56 commented 1 year ago

Thanks, Dr. Stewart. I modified the above tables with VWC based on new bulk density values. I added a new column for that and kept the old one as well just to compare. Also, I added an extra column in saturated VWC table that shows the porosity values based on 1-bulk density/solid density as you suggested above. These values seem more reasonable to me in terms of saturated water content.

sumanager56 commented 1 year ago

Different alpha values Soil tend to remain saturated/retain more moisture at lower alpha values alpha0 0001 alpha0 001 alpha0 1 alpha1 0 alpha10 0 alpha90 0