Adding multiple wells#
This notebook shows how a WellModel can be used to fit multiple wells with one response function. The influence of the individual wells is scaled by the distance to the observation point.
Developed by R.C. Caljé, (Artesia Water 2020), D.A. Brakenhoff, (Artesia Water 2019), and R.A. Collenteur, (Artesia Water 2018)
import os
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import pastas as ps
ps.show_versions()
Pastas version: 1.8.0b
Python version: 3.11.10
NumPy version: 2.0.2
Pandas version: 2.2.3
SciPy version: 1.15.0
Matplotlib version: 3.10.0
Numba version: 0.60.0
DeprecationWarning: As of Pastas 1.5, no noisemodel is added to the pastas Model class by default anymore. To solve your model using a noisemodel, you have to explicitly add a noisemodel to your model before solving. For more information, and how to adapt your code, please see this issue on GitHub: https://github.com/pastas/pastas/issues/735
Load and set data#
Set the coordinates of the extraction wells and calculate the distances to the observation well.
# Specify coordinates observations
xo = 85850
yo = 383362
# Specify coordinates extractions
relevant_extractions = {
"Extraction_2": (83588, 383664),
"Extraction_3": (88439, 382339),
}
# calculate distances
distances = []
for extr, xy in relevant_extractions.items():
xw = xy[0]
yw = xy[1]
distances.append(np.sqrt((xo - xw) ** 2 + (yo - yw) ** 2))
df = pd.DataFrame(
distances,
index=relevant_extractions.keys(),
columns=["Distance to observation well"],
)
df
| Distance to observation well | |
|---|---|
| Extraction_2 | 2282.070989 |
| Extraction_3 | 2783.783397 |
Read the stresses from their csv files
# read oseries
oseries = pd.read_csv(
"data_notebook_10/Observation_well.csv", index_col=0, parse_dates=[0]
).squeeze()
oseries.name = oseries.name.replace(" ", "_")
# read stresses
stresses = {}
for fname in os.listdir("data_notebook_10"):
series = pd.read_csv(
os.path.join("data_notebook_10", fname), index_col=0, parse_dates=[0]
).squeeze()
stresses[fname.strip(".csv").replace(" ", "_")] = series
Then plot the observations, together with the different stresses.
# plot timeseries
f1, axarr = plt.subplots(len(stresses.keys()) + 1, sharex=True, figsize=(10, 8))
oseries.plot(ax=axarr[0], color="k")
axarr[0].set_title(oseries.name)
for i, name in enumerate(stresses.keys(), start=1):
stresses[name].plot(ax=axarr[i])
axarr[i].set_title(name)
plt.tight_layout(pad=0)
Create a model with a separate StressModel for each extraction#
First we create a model with a separate StressModel for each groundwater extraction. First we create a model with the heads timeseries and add recharge as a stress.
# create model
ml = ps.Model(oseries)
ml.add_noisemodel(ps.ArNoiseModel())
Get the precipitation and evaporation timeseries and round the index to remove the hours from the timestamps.
prec = stresses["Precipitation"]
prec.index = prec.index.round("D")
prec.name = "prec"
evap = stresses["Evaporation"]
evap.index = evap.index.round("D")
evap.name = "evap"
Create a recharge stressmodel and add to the model.
rm = ps.RechargeModel(prec, evap, ps.Exponential(), "Recharge")
ml.add_stressmodel(rm)
Modify the extraction timeseries.
extraction_ts = {}
for name in relevant_extractions.keys():
# get extraction timeseries
s = stresses[name]
# convert index to end-of-month timeseries
s.index = s.index.to_period("M").to_timestamp("M")
# resample to daily values
new_index = pd.date_range(s.index[0], s.index[-1], freq="D")
s_daily = ps.ts.timestep_weighted_resample(s, new_index, fast=True).dropna()
name = name.replace(" ", "_")
s_daily.name = name
# append to stresses list
extraction_ts[name] = s_daily
Add each of the extractions as a separate StressModel.
for name, stress in extraction_ts.items():
sm = ps.StressModel(stress, ps.Hantush(), name, up=False, settings="well")
ml.add_stressmodel(sm)
Solve the model.
ml.solve()
Fit report Observation_well Fit Statistics
================================================
nfev 18 EVP 94.41
nobs 2844 R2 0.94
noise True RMSE 0.21
tmin 1960-04-28 12:00:00 AICc -8801.40
tmax 2015-06-29 09:00:00 BIC -8736.01
freq D Obj 63.90
warmup 3650 days 00:00:00 ___
solver LeastSquares Interp. Yes
Parameters (11 optimized)
================================================
optimal initial vary
Recharge_A 1518.467663 210.498526 True
Recharge_a 795.334707 10.000000 True
Recharge_f -1.265581 -1.000000 True
Extraction_2_A -0.000109 -0.000086 True
Extraction_2_a 1286.826020 100.000000 True
Extraction_2_b 0.032393 1.000000 True
Extraction_3_A -0.000043 -0.000171 True
Extraction_3_a 264.112435 100.000000 True
Extraction_3_b 0.827915 1.000000 True
constant_d 10.702132 8.557530 True
noise_alpha 0.005010 1.000000 True
Visualize the results#
Plot the decomposition to see the individual influence of each of the wells.
ml.plots.decomposition();
We can calculate the gain of each extraction (quantified as the effect on the groundwater level of a continuous extraction of ~1 Mm\(^3\)/yr).
for name in relevant_extractions.keys():
sm = ml.stressmodels[name]
p = ml.get_parameters(name)
gain = sm.rfunc.gain(p) * 1e6 / 365.25
print(f"{name}: gain = {gain:.3f} m / Mm^3/year")
df.at[name, "gain StressModel"] = gain
Extraction_2: gain = -0.299 m / Mm^3/year
Extraction_3: gain = -0.119 m / Mm^3/year
Create a model with a WellModel#
We can reduce the number of parameters in the model by including the three extractions in a WellModel. This WellModel takes into account the distances from the three extractions to the observation well, and assumes constant geohydrological properties. All of the extractions now share the same response function, scaled by the distance between the extraction well and the observation well.
First we create a new model and add recharge.
ml_wm = ps.Model(oseries, oseries.name + "_wm")
ml_wm.add_noisemodel(ps.ArNoiseModel())
rm = ps.RechargeModel(prec, evap, ps.Gamma(), "Recharge")
ml_wm.add_stressmodel(rm)
We have all the information we need to create a WellModel:
timeseries for each of the extractions, these are passed as a list of stresses
distances from each extraction to the observation point, note that the order of these distances must correspond to the order of the stresses.
Note: the WellModel only works with a special version of the Hantush response function called HantushWellModel. This is because the response function must support scaling by a distance \(r\). The HantushWellModel response function has been modified to support this. The Hantush response normally takes three parameters: the gain \(A\), \(a\) and \(b\). This special version accepts 4 parameters: it interprets that fourth parameter as the distance \(r\), and uses it to scale the parameters accordingly.
Create the WellModel and add to the model.
w = ps.WellModel(list(extraction_ts.values()), "WellModel", distances)
ml_wm.add_stressmodel(w)
Solve the model.
As we can see, the fit with the measurements (EVP) is similar to the result with the previous model, with each well included separately.
ml_wm.solve()
Fit report Observation_well Fit Statistics
=================================================
nfev 32 EVP 93.47
nobs 2844 R2 0.93
noise True RMSE 0.23
tmin 1960-04-28 12:00:00 AICc -13674.50
tmax 2015-06-29 09:00:00 BIC -13620.99
freq D Obj 11.53
warmup 3650 days 00:00:00 ___
solver LeastSquares Interp. Yes
Parameters (9 optimized)
=================================================
optimal initial vary
Recharge_A 1403.044778 210.498526 True
Recharge_n 1.002640 1.000000 True
Recharge_a 911.243534 10.000000 True
Recharge_f -1.999807 -1.000000 True
WellModel_A -0.000372 -0.000756 True
WellModel_a 491.124635 100.000000 True
WellModel_b -16.152562 -15.674262 True
constant_d 12.095317 8.557530 True
noise_alpha 56.686906 1.000000 True
Visualize the results#
Plot the decomposition to see the individual influence of each of the wells
ml_wm.plots.decomposition();
Plot the stacked influence of each of the individual extraction wells in the results plot
ml_wm.plots.stacked_results(
figsize=(10, 8),
stacklegend=True,
stackcolors={"Extraction_2": "C1", "Extraction_3": "C2"},
);
Get parameters for each well (including the distance) and calculate the gain. The WellModel reorders the stresses from closest to the observation well, to furthest from the observation well. We have take this into account during the post-processing.
The gain of extraction 3 is lower than the gain of extraction 2. This will always be the case in a WellModel when the distance from the observation well to extraction 3 is larger than the distance to extraction 2.
wm = ml_wm.stressmodels["WellModel"]
for i, name in enumerate(relevant_extractions.keys()):
# get parameters (note use of stressmodel for this)
p = wm.get_parameters(model=ml_wm, istress=i)
# calculate gain
gain = wm.rfunc.gain(p) * 1e6 / 365.25
name = wm.stress[i].name
print(f"{name}: gain = {gain:.3f} m / Mm^3/year")
df.at[name, "gain WellModel"] = gain
Extraction_2: gain = -0.242 m / Mm^3/year
Extraction_3: gain = -0.162 m / Mm^3/year
Calculate gain as function of radial distance for and plot the result, including the estimated uncertainty.
r = np.logspace(3, 3.6, 101)
# calculate gain and std error vs distance
params = ml_wm.get_parameters(wm.name)
gain_wells = wm.rfunc.gain(params, r=wm.distances.values) * 1e6 / 365.25
gain_vs_dist = wm.rfunc.gain(params, r=r) * 1e6 / 365.25
gain_std_vs_dist = np.sqrt(wm.variance_gain(ml_wm, r=r)) * 1e6 / 365.25
fig, ax = plt.subplots(1, 1, figsize=(10, 3))
ax.plot(r, gain_vs_dist, color="C0", label="gain")
ax.plot(
wm.distances,
gain_wells,
color="C3",
marker="o",
mfc="none",
label="wells",
ls="none",
)
ax.fill_between(
r,
gain_vs_dist - 2 * gain_std_vs_dist,
gain_vs_dist + 2 * gain_std_vs_dist,
alpha=0.35,
label="CI 95%",
)
ax.axhline(0.0, linestyle="dashed", color="k")
ax.legend(loc=(0, 1), frameon=False, ncol=3)
ax.grid(visible=True)
ax.set_xlabel("radial distance [m]")
ax.set_ylabel("gain [m / (Mm$^3$/yr)]");
Compare individual StressModels and WellModel#
Compare the gains that were calculated by the individual StressModels and the WellModel.
df.style.format("{:.4f}")
| Distance to observation well | gain StressModel | gain WellModel | |
|---|---|---|---|
| Extraction_2 | 2282.0710 | -0.2994 | -0.2421 |
| Extraction_3 | 2783.7834 | -0.1188 | -0.1621 |
Visually compare the two models, including the calculated contribution of the wells.
Note that there is some extra code at the bottom to calculate two step responses for the “WellModel” model, for comparison purposes with the “2-wells” model.
# give models descriptive name
ml.name = "2_wells"
ml_wm.name = "WellModel"
# plot well stresses together on same plot:
smdict = {0: ["Recharge"], 1: ["Extraction_2", "Extraction_3", "WellModel"]}
# comparison plot
mc = ps.CompareModels([ml, ml_wm])
mosaic = mc.get_default_mosaic(n_stressmodels=2)
mc.initialize_adjust_height_figure(mosaic=mosaic, smdict=smdict)
mc.plot(smdict=smdict)
sumwells = ml.get_contribution("Extraction_2") + ml.get_contribution("Extraction_3")
mc.axes["con1"].plot(
sumwells.index, sumwells, ls="dashed", color="C0", label="sum 2_wells"
)
mc.axes["con1"].legend(loc=(0, 1), frameon=False, ncol=4)
# remove WellModel response for r=1m and add response twice, scaled with actual
# distances, for comparison with the two responses from the first model
mc.axes["rf1"].lines[-1].remove() # remove original step response
for istress in range(2):
# get parameters and distance for istress
p = ml_wm.stressmodels["WellModel"].get_parameters(istress=istress)
# calculate step
step = ml_wm.get_step_response("WellModel", p=p)
# plot step
mc.axes["rf1"].plot(step.index, step, color="C1")
# recalculate axes limits
mc.axes["rf1"].relim()
