use global MODIS LAI for all simulations

Author: juliasloan25

Artifacts.toml

@@ -185,10 +185,3 @@ git-tree-sha1 = "c35ba0e899040cb8153226ab751f69100f475d39"
     [[mizoguchi_soil_freezing_data.download]]
     sha256 = "0027cc080ba45ba33dc790b176ec2854353ce7dce4eae4bef72963b0dd944e0b"
     url = "https://caltech.box.com/shared/static/tn1bnqjmegyetw5kzd2ixq5pbnb05s3u.gz"
-
-[modis_lai_fluxnet_sites]
-git-tree-sha1 = "cdcc1708832654d2209d267e01a3893c4b25c085"
-
-    [[modis_lai_fluxnet_sites.download]]
-    sha256 = "c74780775d98b56eed74222377604272658f59686d4d7881c7b815b6c230080b"
-    url = "https://caltech.box.com/shared/static/g6vubma6vcxulb9fl0pvzq6v0oto4iu3.gz"

NEWS.md

@@ -3,6 +3,7 @@ ClimaLand.jl Release Notes
 
 main
 -------
+- Use global MODIS LAI for all simulations; remove `modis_lai_fluxnet_sites` artifact PR[#1282](https://github.com/CliMA/ClimaLand.jl/pull/1282)
 - Use ClimaUtilities v0.1.25 to read in spatial data to a point using lat/lon PR[#1279](https://github.com/CliMA/ClimaLand.jl/pull/1279)
 - Add data handling tools to FluxnetSimulationsExt and use throughout docs and experiments PR[#1238](https://github.com/CliMA/ClimaLand.jl/pull/1238)
 - Add convenience constructors for CanopyModel and integrated models PR[#1255](https://github.com/CliMA/ClimaLand.jl/pull/1255)

docs/src/tutorials/integrated/soil_canopy_tutorial.jl

@@ -42,6 +42,8 @@ using Insolation
 # Load CliMA Packages and ClimaLand Modules:
 
 using ClimaCore
+import ClimaComms
+ClimaComms.@import_required_backends
 import ClimaParams as CP
 import ClimaTimeSteppers as CTS
 using ClimaLand
@@ -62,8 +64,20 @@ FluxnetSimulationsExt =
 const FT = Float32;
 earth_param_set = LP.LandParameters(FT);
 
-# Setup the domain for the model:
+# First provide some information about the site
+site_ID = "US-MOz"
+# Timezone (offset from UTC in hrs)
+time_offset = 7
+start_date = DateTime(2010) + Hour(time_offset)
+
+# Site latitude and longitude
+lat = FT(38.7441) # degree
+long = FT(-92.2000) # degree
+
+# Height of the sensor at the site
+atmos_h = FT(32) # m
 
+# Setup the domain for the model:
 nelements = 10
 zmin = FT(-2)
 zmax = FT(0)
@@ -75,28 +89,21 @@ h_stem = FT(9)
 h_leaf = FT(9.5)
 compartment_midpoints = [h_stem / 2, h_stem + h_leaf / 2]
 compartment_surfaces = [zmax, h_stem, h_stem + h_leaf]
-land_domain = Column(; zlim = (zmin, zmax), nelements = nelements);
+land_domain =
+    Column(; zlim = (zmin, zmax), nelements = nelements, longlat = (long, lat));
+
+# Specify the time range and dt value over which to perform the simulation.
+t0 = Float64(150 * 3600 * 24)# start mid year
+N_days = 100
+tf = t0 + Float64(3600 * 24 * N_days)
+dt = Float64(30)
 
 # - We will be using prescribed atmospheric and radiative drivers from the
 #   US-MOz tower, which we read in here. We are using prescribed
 #   atmospheric and radiative flux conditions, but it is also possible to couple
 #   the simulation with atmospheric and radiative flux models. We also
 # read in the observed LAI and let that vary in time in a prescribed manner.
 
-
-# First provide some information about the site
-site_ID = "US-MOz"
-# Timezone (offset from UTC in hrs)
-time_offset = 7
-start_date = DateTime(2010) + Hour(time_offset)
-
-# Site latitude and longitude
-lat = FT(38.7441) # degree
-long = FT(-92.2000) # degree
-
-# Height of the sensor at the site
-atmos_h = FT(32) # m
-
 # Forcing data
 (; atmos, radiation) = FluxnetSimulationsExt.prescribed_forcing_fluxnet(
     site_ID,
@@ -223,12 +230,25 @@ photosynthesis_args =
     (; parameters = FarquharParameters(FT, is_c3; Vcmax25 = FT(5e-5)));
 
 K_sat_plant = FT(1.8e-8)
-(; LAI, maxLAI) =
-    FluxnetSimulationsExt.prescribed_LAI_fluxnet(site_ID, start_date)
-maxLAI = FT(maxLAI) # convert to the float type of our simulation
+
+# Read in LAI from MODIS data
+surface_space = land_domain.space.surface
+modis_lai_ncdata_path = ClimaLand.Artifacts.modis_lai_multiyear_paths(;
+    context = ClimaComms.context(surface_space),
+    start_date = start_date + Second(t0),
+    end_date = start_date + Second(t0) + Second(tf),
+)
+LAI = ClimaLand.prescribed_lai_modis(
+    modis_lai_ncdata_path,
+    surface_space,
+    start_date,
+)
+LAI_init = ClimaCore.Fields.zeros(surface_space)
+evaluate!(LAI_init, LAI, start_date + Second(t0))
+maxLAI = maximum(LAI_init)
+
 SAI = FT(0.00242)
 RAI = (SAI + maxLAI) * f_root_to_shoot;
-# Note: LAIfunction was determined from data in the script we included above.
 ai_parameterization = PrescribedSiteAreaIndex{FT}(LAI, SAI, RAI)
 
 ψ63 = FT(-4 / 0.0098)
@@ -346,15 +366,7 @@ for i in 1:2
         augmented_liquid_fraction.(plant_ν, S_l_ini[i])
 end;
 
-# Select the timestepper and solvers needed for the specific problem. Specify the time range and dt
-# value over which to perform the simulation.
-
-t0 = Float64(150 * 3600 * 24)# start mid year
-N_days = 100
-tf = t0 + Float64(3600 * 24 * N_days)
-dt = Float64(30)
-n = 120
-saveat = Array(t0:(n * dt):tf)
+# Select the timestepper and solvers needed for the specific problem.
 
 timestepper = CTS.ARS343()
 ode_algo = CTS.IMEXAlgorithm(
@@ -374,6 +386,8 @@ set_initial_cache!(p, Y, t0);
 # how often we save output, and how often we update
 # the forcing data ("drivers")
 
+n = 120
+saveat = Array(t0:(n * dt):tf)
 sv = (;
     t = Array{Float64}(undef, length(saveat)),
     saveval = Array{NamedTuple}(undef, length(saveat)),

docs/src/tutorials/standalone/Canopy/canopy_tutorial.jl

@@ -43,6 +43,8 @@ using Insolation
 # Load CliMA Packages and ClimaLand Modules:
 
 using ClimaCore
+import ClimaComms
+ClimaComms.@import_required_backends
 import ClimaParams as CP
 import ClimaTimeSteppers as CTS
 using StaticArrays
@@ -73,7 +75,21 @@ earth_param_set = LP.LandParameters(FT);
 # stomatal conductance model, and atmospheric and radiative flux conditions
 # which may be either prescribed or simulated.
 
-# First, define the parameters of the model domain. These values are needed
+# First provide some information about the site
+# Timezone (offset from UTC in hrs)
+time_offset = 7
+start_date = DateTime(2010) + Hour(time_offset)
+
+# Site latitude and longitude
+lat = FT(38.7441) # degree
+long = FT(-92.2000) # degree
+
+# Height of the sensor at the site
+atmos_h = FT(32)
+# Site ID
+site_ID = "US-MOz";
+
+# Now we define the parameters of the model domain. These values are needed
 # by some of the component models. Here we are performing a 1-dimensional
 # simulation in a `Point` domain and will use
 # single stem and leaf compartments, but for 2D simulations, the parameters of
@@ -91,28 +107,23 @@ h_stem = FT(9)
 h_leaf = FT(9.5)
 compartment_midpoints = [h_stem / 2, h_stem + h_leaf / 2]
 compartment_surfaces = [zmax, h_stem, h_stem + h_leaf]
-land_domain = Point(; z_sfc = FT(0.0))
+land_domain = Point(; z_sfc = FT(0.0), longlat = (long, lat));
+
+# Select a time range to perform time stepping over, and a dt. As usual,
+# the timestep depends on the problem you are solving, the accuracy of the
+# solution required, and the timestepping algorithm you are using.
+
+t0 = 0.0
+N_days = 364
+tf = t0 + 3600 * 24 * N_days
+dt = 225.0;
 
 # - We will be using prescribed atmospheric and radiative drivers from the
 #   US-MOz tower, which we read in here. We are using prescribed
 #   atmospheric and radiative flux conditions, but it is also possible to couple
 #   the simulation with atmospheric and radiative flux models. We also
 # read in the observed LAI and let that vary in time in a prescribed manner.
 
-# First provide some information about the site
-# Timezone (offset from UTC in hrs)
-time_offset = 7
-start_date = DateTime(2010) + Hour(time_offset)
-
-# Site latitude and longitude
-lat = FT(38.7441) # degree
-long = FT(-92.2000) # degree
-
-# Height of the sensor at the site
-atmos_h = FT(32)
-# Site ID
-site_ID = "US-MOz";
-
 # Forcing data
 (; atmos, radiation) = FluxnetSimulationsExt.prescribed_forcing_fluxnet(
     site_ID,
@@ -185,8 +196,22 @@ AR_model = AutotrophicRespirationModel{FT}(AR_params);
 # Begin by providing general plant parameters. For the area
 # indices of the canopy, we choose a `PrescribedSiteAreaIndex`,
 # which supports LAI as a function of time, with RAI and SAI as constant.
-(; LAI, maxLAI) =
-    FluxnetSimulationsExt.prescribed_LAI_fluxnet(site_ID, start_date)
+# Read in LAI from MODIS data
+surface_space = land_domain.space.surface
+modis_lai_ncdata_path = ClimaLand.Artifacts.modis_lai_multiyear_paths(
+    start_date = start_date + Second(t0),
+    end_date = start_date + Second(t0) + Second(tf);
+    context = ClimaComms.context(surface_space),
+)
+LAI = ClimaLand.prescribed_lai_modis(
+    modis_lai_ncdata_path,
+    surface_space,
+    start_date,
+)
+LAI_init = ClimaCore.Fields.zeros(surface_space)
+evaluate!(LAI_init, LAI, start_date + Second(t0))
+maxLAI = maximum(LAI_init)
+
 SAI = FT(0.00242)
 f_root_to_shoot = FT(3.5)
 RAI = FT((SAI + maxLAI) * f_root_to_shoot)
@@ -281,16 +306,6 @@ for i in 1:2
     Y.canopy.hydraulics.ϑ_l.:($i) .= augmented_liquid_fraction.(ν, S_l_ini[i])
 end;
 
-# Select a time range to perform time stepping over, and a dt. Also create the
-# saveat Array to contain the data from the model at each time step. As usual,
-# the timestep depends on the problem you are solving, the accuracy of the
-# solution required, and the timestepping algorithm you are using.
-
-t0 = 0.0
-N_days = 364
-tf = t0 + 3600 * 24 * N_days
-dt = 225.0;
-
 # Initialize the cache variables for the canopy using the initial
 # conditions and initial time.
 

experiments/benchmarks/snowy_land.jl

@@ -203,9 +203,10 @@ function setup_prob(t0, tf, Δt; outdir = outdir, nelements = (101, 15))
     photosynthesis_args =
         (; parameters = Canopy.FarquharParameters(FT, is_c3; Vcmax25 = Vcmax25))
     # Set up plant hydraulics
-    modis_lai_ncdata_path = ClimaLand.Artifacts.modis_lai_single_year_path(;
+    modis_lai_ncdata_path = ClimaLand.Artifacts.modis_lai_multiyear_paths(;
         context = nothing,
-        year = Dates.year(Second(t0) + start_date),
+        start_date = start_date + Second(t0),
+        end_date = start_date + Second(t0) + Second(tf),
     )
     LAIfunction = ClimaLand.prescribed_lai_modis(
         modis_lai_ncdata_path,

experiments/calibration/forward_model_land.jl

@@ -216,8 +216,11 @@ function setup_prob(
         #sc = sc,
     ))
     # Set up plant hydraulics
-    modis_lai_ncdata_path =
-        ClimaLand.Artifacts.find_modis_year_paths(date(t0), date(tf); context)
+    modis_lai_ncdata_path = ClimaLand.Artifacts.modis_lai_multiyear_paths(
+        date(t0),
+        date(tf);
+        context,
+    )
     LAIfunction = ClimaLand.prescribed_lai_modis(
         modis_lai_ncdata_path,
         surface_space,

experiments/integrated/fluxnet/fluxnet_domain.jl

@@ -1,5 +1,5 @@
 """This file sets up the domain to run Clima Land on a fluxtower site, using both
-the generic parameters set in this file as well as the site-specific parameters 
+the generic parameters set in this file as well as the site-specific parameters
 given in the {site-ID}_simulation.jl file in each site directory."""
 
 # Domain setup
@@ -8,7 +8,3 @@ h_canopy = h_stem + h_leaf
 compartment_midpoints =
     n_stem > 0 ? [h_stem / 2, h_stem + h_leaf / 2] : [h_leaf / 2]
 compartment_surfaces = n_stem > 0 ? [zmax, h_stem, h_canopy] : [zmax, h_leaf]
-
-land_domain =
-    Column(; zlim = (zmin, zmax), nelements = nelements, dz_tuple = dz_tuple)
-canopy_domain = ClimaLand.Domains.obtain_surface_domain(land_domain)

experiments/integrated/fluxnet/ozark_pft.jl

@@ -8,6 +8,7 @@ import ClimaLand
 import SciMLBase
 import ClimaTimeSteppers as CTS
 using ClimaCore
+import ClimaComms
 import ClimaParams as CP
 using CairoMakie
 using Statistics
@@ -57,6 +58,14 @@ include(
     ),
 )
 
+land_domain = Column(;
+    zlim = (zmin, zmax),
+    nelements = nelements,
+    dz_tuple = dz_tuple,
+    longlat = (long, lat),
+)
+canopy_domain = ClimaLand.Domains.obtain_surface_domain(land_domain)
+
 # Define the PFT land cover percentages for the Ozark site. Currently we only
 # use the dominant PFT, which for Ozark is deciduous broadleaf temperate trees.
 pft_pcts = [
@@ -117,8 +126,22 @@ start_date = DateTime(2010) + Hour(time_offset)
     earth_param_set,
     FT,
 )
-(; LAI, maxLAI) =
-    FluxnetSimulationsExt.prescribed_LAI_fluxnet(site_ID, start_date)
+
+# Read in LAI from MODIS data
+surface_space = land_domain.space.surface
+modis_lai_ncdata_path = ClimaLand.Artifacts.modis_lai_multiyear_paths(
+    start_date = start_date + Second(t0),
+    end_date = start_date + Second(t0) + Second(tf);
+    context = ClimaComms.context(surface_space),
+)
+LAI = ClimaLand.prescribed_lai_modis(
+    modis_lai_ncdata_path,
+    surface_space,
+    start_date,
+)
+LAI_init = ClimaCore.Fields.zeros(surface_space)
+evaluate!(LAI_init, LAI, start_date + Second(t0))
+maxLAI = maximum(LAI_init)
 RAI = maxLAI * f_root_to_shoot
 capacity = plant_ν * maxLAI * h_leaf * FT(1000)
 

experiments/integrated/fluxnet/run_fluxnet.jl

@@ -58,6 +58,14 @@ include(
     ),
 )
 
+land_domain = Column(;
+    zlim = (zmin, zmax),
+    nelements = nelements,
+    dz_tuple = dz_tuple,
+    longlat = (long, lat),
+)
+canopy_domain = ClimaLand.Domains.obtain_surface_domain(land_domain)
+
 # This reads in the data from the flux tower site and creates
 # the atmospheric and radiative driver structs for the model
 include(
@@ -78,8 +86,22 @@ include(
     earth_param_set,
     FT,
 )
-(; LAI, maxLAI) =
-    FluxnetSimulationsExt.prescribed_LAI_fluxnet(site_ID, start_date)
+
+# Read in LAI from MODIS data
+surface_space = land_domain.space.surface
+modis_lai_ncdata_path = ClimaLand.Artifacts.modis_lai_multiyear_paths(;
+    context = ClimaComms.context(surface_space),
+    start_date = start_date + Second(t0),
+    end_date = start_date + Second(t0) + Second(tf),
+)
+LAI = ClimaLand.prescribed_lai_modis(
+    modis_lai_ncdata_path,
+    surface_space,
+    start_date,
+)
+LAI_init = ClimaCore.Fields.zeros(surface_space)
+evaluate!(LAI_init, LAI, start_date + Second(t0))
+maxLAI = maximum(LAI_init)
 RAI = maxLAI * f_root_to_shoot
 capacity = plant_ν * maxLAI * h_leaf * FT(1000)
 

experiments/integrated/global/global_soil_canopy.jl

@@ -105,9 +105,10 @@ conductance_args = (; parameters = Canopy.MedlynConductanceParameters(FT; g1))
 photosynthesis_args =
     (; parameters = Canopy.FarquharParameters(FT, is_c3; Vcmax25 = Vcmax25))
 # Set up plant hydraulics
-modis_lai_ncdata_path = ClimaLand.Artifacts.modis_lai_single_year_path(;
+modis_lai_ncdata_path = ClimaLand.Artifacts.modis_lai_multiyear_paths(;
     context = nothing,
-    year = Dates.year(Second(t0) + start_date),
+    start_date = start_date + Second(t0),
+    end_date = start_date + Second(t0) + Second(tf),
 )
 LAIfunction = ClimaLand.prescribed_lai_modis(
     modis_lai_ncdata_path,