def concavity_FFS_down_to_top():
# Here are the different parameters and their default value fr this script
default_param = AGPD.get_common_default_param()
default_param["already_preprocessed"] = False
default_param["save_XY_array"] = False
default_param["X_source"] = None
default_param["Y_source"] = None
default_param["min_elevation"] = None
default_param["area_threshold"] = None
default_param["flow_distance_step"] = None
default_param["min_DA"] = None
default_param["prefix"] = ""
default_param = AGPD.ingest_param(default_param, sys.argv)
if (default_param["help"] or len(sys.argv) == 1 or "help" in sys.argv):
print("""
This command-line tool run concavity analysis tools from LSDTopoTools.
Description of the algorithms in Mudd et al., 2018 -> https://www.earth-surf-dynam.net/6/505/2018/
To use, run the script with relevant options, for example:
lsdtt-concavity-down-to-top file=myraster.tif ---
option available:
file: name of the raster (file=name.tif)
path: path to the file (default = current folder)
todo
help: if written, diplay this message. Documentation soon to be written.
""")
quit()
try:
default_param["X_source"] = float(default_param["X_source"])
default_param["Y_source"] = float(default_param["Y_source"])
default_param["min_elevation"] = float(default_param["min_elevation"])
default_param["area_threshold"] = float(
default_param["area_threshold"])
default_param["flow_distance_step"] = float(
default_param["flow_distance_step"])
default_param["min_DA"] = float(default_param["min_DA"])
except:
print(default_param)
print(
"I struggle to understand your input parameters: make sure I can convert them to number"
)
raise SystemExit("I need to quit now")
mydem = LSDDEM(file_name=default_param["file"],
path=default_param["path"],
already_preprocessed=default_param["already_preprocessed"],
remove_seas=True,
sea_level=default_param["min_elevation"])
if (default_param["already_preprocessed"] == False):
print(
"I am preprocessing your raster with default options, see lsdtt-depressions for extended options."
)
print(
"Command line tools can load already processed rasters with the keyword already_preprocessed"
)
mydem.PreProcessing()
# flwo routines
print("Processing flow routines with d8 algorithm")
mydem.CommonFlowRoutines()
print("Extracting the river")
river = mydem.ExtractSingleRiverFromSource(default_param["X_source"],
default_param["Y_source"])
river = pd.DataFrame(river)
print("Saving teh river to csv file: river_for_concavity.csv")
river.to_csv("%sriver_for_concavity.csv" % (default_param["prefix"]),
index=False)
print("Initialising the concavity analysis")
river.sort_values("elevation", inplace=True)
river.reset_index(drop=True, inplace=True)
# OUtputs
global_results = {}
outlet_info = {
"X": [],
"Y": [],
"elevation": [],
"flow_distance": [],
"drainage_area": []
}
# potential saving of basins
XY_basins = {}
flow_distance_this = river["flow_distance"].min()
index_this = river["flow_distance"].idxmin()
while (flow_distance_this < river["flow_distance"].max()
and river["drainage_area"][index_this] > default_param["min_DA"]):
print("processing flow distance =", flow_distance_this)
del mydem
mydem = LSDDEM(file_name="%sdem_for_concavity.tif" %
(default_param["prefix"]),
already_preprocessed=True,
remove_seas=True,
sea_level=river["elevation"][index_this] - 5)
mydem.CommonFlowRoutines()
# mydem.save_array_to_raster_extent(mydem.cppdem.get_DA_raster(), name = "DA", save_directory = "./")
mydem.ExtractRiverNetwork(
area_threshold_min=default_param["area_threshold"])
# print(river.iloc[index_this])
mydem.DefineCatchment(method="from_XY",
X_coords=[river["X"][index_this]],
Y_coords=[river["Y"][index_this]],
coord_search_radius_nodes=0)
mydem.GenerateChi()
print(mydem.df_base_river["source_key"].unique())
mydem.cppdem.calculate_movern_disorder(0.05, 0.05, 19, 1,
default_param["area_threshold"])
all_disorder = mydem.cppdem.get_best_fits_movern_per_BK()
global_results[str(round(
river["flow_distance"][index_this], 2))] = np.array(
all_disorder[0]
) # [0] means basin 0, which is normal as I only have one
outlet_info["X"].append(river["X"][index_this])
outlet_info["Y"].append(river["Y"][index_this])
outlet_info["elevation"].append(river["elevation"][index_this])
outlet_info["flow_distance"].append(
round(river["flow_distance"][index_this], 2))
outlet_info["drainage_area"].append(river["drainage_area"][index_this])
# New index
flow_distance_this += default_param["flow_distance_step"]
index_this = river.index[
river["flow_distance"] >= flow_distance_this].values[0]
np.savez("%sdown_to_top_conc.npz" % (default_param["prefix"]),
**global_results)
pd.DataFrame(outlet_info).to_csv("%sdown_to_top_conc_basin_info.csv" %
(default_param["prefix"]),
index=False)
if (default_param["save_XY_array"]):
this_basin = mydem.cppdem.query_xy_for_each_basin()[
0] # [0] because I want a single basin there
print(this_basin)
XY_basins[str(round(river["flow_distance"][index_this],
2))] = this_basin
np.savez(default_param["prefix"] + "down_to_top_basins_XY.npz",
**XY_basins)
def spawn_XY_outlet():
"""
Command-line tool to prechoose the basins used for other analysis. Outputs a file with outlet coordinates readable from other command-line tools and a basin perimeter csv readable by GISs to if the basins corresponds to your needs.
Takes several arguments (the values after = are example values to adapt):
file=NameOfFile.tif -> The code NEEDS the neame of the raster to process.
already_preprocessed -> OPTIONAL Tell the code your raster does not need preprocessing, otherwise carve the DEM (see lsdtt-depressions for more options)
test_edges -> OPTIONAL will test if the basin extracted are potentially influenced by nodata and threfore uncomplete. WARNING, will take out ANY basin potentially cut, if you know what you are doing, you can turn off.
prefix=test -> OPTIONAL Add a prefix to each outputted file (handy for automation)
method=from_range -> DEFAULT from_range: determine the method to select basin. Can be
from_range -> select largest basins bigger than min_DA but smaller than max_DA (in m^2)
min_area -> select largest basins bigger than min_DA
main_basin -> select the largest basin
Other methods to come.
min_elevation=45 -> DEFAULT 0. Ignore any basin bellow that elevation
area_threshold=3500 -> DEFAULT 5000. River network area threshold in number of pixels (part of the basin selection is based on river junctions HIGHLY sensitive to that variable).
Example:
lsdtt-concFFS-spawn-outlets file=DEM.tif already_preprocessed min_DA=1e7 max_DA=1e9 area_threshold=3500
"""
default_param = AGPD.get_common_default_param()
default_param["already_preprocessed"] = False
default_param["test_edges"] = False
default_param["area_threshold"] = 5000
default_param["method"] = "from_range"
default_param["min_DA"] = 1e6
default_param["max_DA"] = 1e9
default_param["min_elevation"] = 0
default_param["prefix"] = ""
default_param = AGPD.ingest_param(default_param, sys.argv)
choice_of_method = ["min_area", "main_basin", "from_range"]
if (default_param["help"] or len(sys.argv) == 1 or "help" in sys.argv):
print("""
Command-line tool to prechoose the basins used for other analysis. Outputs a file with outlet coordinates readable from other command-line tools and a basin perimeter csv readable by GISs to if the basins corresponds to your needs.
Takes several arguments (the values after = are example values to adapt):
file=NameOfFile.tif -> The code NEEDS the neame of the raster to process.
already_preprocessed -> OPTIONAL Tell the code your raster does not need preprocessing, otherwise carve the DEM (see lsdtt-depressions for more options)
test_edges -> OPTIONAL will test if the basin extracted are potentially influenced by nodata and threfore uncomplete. WARNING, will take out ANY basin potentially cut, if you know what you are doing, you can turn off.
prefix=test -> OPTIONAL Add a prefix to each outputted file (handy for automation)
method=from_range -> DEFAULT from_range: determine the method to select basin. Can be
from_range -> select largest basins bigger than min_DA but smaller than max_DA (in m^2)
min_area -> select largest basins bigger than min_DA
main_basin -> select the largest basin
Other methods to come.
min_elevation=45 -> DEFAULT 0. Ignore any basin bellow that elevation
area_threshold=3500 -> DEFAULT 5000. River network area threshold in number of pixels (part of the basin selection is based on river junctions HIGHLY sensitive to that variable).
Example:
lsdtt-concFFS-spawn-outlets file=DEM.tif already_preprocessed min_DA=1e7 max_DA=1e9 area_threshold=3500
""")
return 0
# Checks if the method requested is valid or not
if (default_param["method"].lower() not in choice_of_method):
print("I cannot recognise the method! Please choose from:")
print(choice_of_method)
return 0
# Formatting parameters
area_threshold = int(default_param["area_threshold"])
min_DA = float(default_param["min_DA"])
max_DA = float(default_param["max_DA"])
min_elevation = float(default_param["min_elevation"])
# Reading DEM
mydem = LSDDEM(file_name=default_param["file"],
path=default_param["path"],
already_preprocessed=default_param["already_preprocessed"],
remove_seas=True,
sea_level=min_elevation)
if (default_param["already_preprocessed"] == False):
mydem.PreProcessing()
# Extracting basins
mydem.CommonFlowRoutines()
print("Done with flow routines")
mydem.ExtractRiverNetwork(method="area_threshold",
area_threshold_min=area_threshold)
# Get the outlet coordinates of all the extracted basins
print("Extracted rivers")
df_outlet = mydem.DefineCatchment(
method=default_param["method"],
min_area=min_DA,
max_area=max_DA,
test_edges=default_param["test_edges"]
) #, X_coords = [X_coordinates_outlets[7]], Y_coords = [Y_coordinates_outlets[7]])
print("Extracted")
for key, val in df_outlet.items():
df_outlet[key] = np.array(df_outlet[key])
# Getting the rivers
mydem.GenerateChi(theta=0.4, A_0=1)
# Saing the rivers to csv
mydem.df_base_river.to_csv(default_param["prefix"] + "rivers.csv",
index=False)
#Saving the outlet
df_outlet["area_threshold"] = np.full(df_outlet["X"].shape[0],
area_threshold)
# print(df_outlet)
pd.DataFrame(df_outlet).to_csv(default_param["prefix"] + "outlets.csv",
index=False)
# Getting the perimeter of basins
this = mydem.cppdem.extract_perimeter_of_basins()
df_perimeter = {"X": [], "Y": [], "Z": [], "IDs": []}
for key, val in this.items():
df_perimeter["X"].append(np.array(val["X"]))
df_perimeter["Y"].append(np.array(val["Y"]))
df_perimeter["Z"].append(np.array(val["Z"]))
df_perimeter["IDs"].append(np.full(np.array(val["Z"]).shape[0], key))
for key, val in df_perimeter.items():
df_perimeter[key] = np.concatenate(val)
pd.DataFrame(df_perimeter).to_csv(default_param["prefix"] +
"perimeters.csv",
index=False)
def spawn_XY_outlet_subbasins():
default_param = AGPD.get_common_default_param()
default_param["already_preprocessed"] = False
default_param["X"] = 0
default_param["Y"] = 0
default_param["area_threshold"] = 5000
default_param["min_DA"] = 1e6
default_param["max_DA"] = 1e9
default_param["min_elevation"] = 0
default_param["prefix"] = ""
default_param = AGPD.ingest_param(default_param, sys.argv)
if (default_param["help"] or len(sys.argv) == 1 or "help" in sys.argv):
print("""
Command-line tool to extract basin information about all the subbasins within a main one. Outputs a file with outlet coordinates readable from other command-line tools and a basin perimeter csv readable by GISs to if the basins corresponds to your needs.
Takes several arguments (the values after = are example values to adapt):
file=NameOfFile.tif -> The code NEEDS the neame of the raster to process.
already_preprocessed -> OPTIONAL Tell the code your raster does not need preprocessing, otherwise carve the DEM (see lsdtt-depressions for more options)
prefix=test -> OPTIONAL Add a prefix to each outputted file (handy for automation)
min_elevation=45 -> DEFAULT 0. Ignore any basin bellow that elevation
area_threshold=3500 -> DEFAULT 5000. River network area threshold in number of pixels (part of the basin selection is based on river junctions HIGHLY sensitive to that variable).
min_DA=1e7 -> minimum drainage area to extract a subbasin
max_DA=1e9 -> maximum drainage area for a subbasin
X=234 -> X Coordinate (in map unit) of the outlet (needs to be the exact pixel at the moment, will add a snapping option later)
Y=234 -> Y Coordinate (in map unit) of the outlet (needs to be the exact pixel at the moment, will add a snapping option later)
Example:
lsdtt-concFFS-spawn-outlets file=DEM.tif already_preprocessed min_DA=1e7 max_DA=1e9 area_threshold=3500
""")
return 0
area_threshold = int(default_param["area_threshold"])
X = float(default_param["X"])
min_DA = float(default_param["min_DA"])
Y = float(default_param["Y"])
max_DA = float(default_param["max_DA"])
min_elevation = float(default_param["min_elevation"])
mydem = LSDDEM(file_name=default_param["file"],
path=default_param["path"],
already_preprocessed=default_param["already_preprocessed"],
remove_seas=True,
sea_level=min_elevation)
if (default_param["already_preprocessed"] == False):
mydem.PreProcessing()
# Extracting basins
mydem.CommonFlowRoutines()
mydem.ExtractRiverNetwork(method="area_threshold",
area_threshold_min=area_threshold)
# df_outlet = mydem.DefineCatchment( method = default_param["method"], min_area = min_DA, max_area = max_DA, test_edges = default_param["test_edges"])#, X_coords = [X_coordinates_outlets[7]], Y_coords = [Y_coordinates_outlets[7]])
df_outlet = mydem.cppdem.calculate_outlets_min_max_draining_to_baselevel(
X, Y, min_DA, max_DA, 500)
mydem.check_catchment_defined = True
for key, val in df_outlet.items():
df_outlet[key] = np.array(df_outlet[key])
mydem.GenerateChi(theta=0.4, A_0=1)
mydem.df_base_river.to_csv(default_param["prefix"] + "rivers.csv",
index=False)
df_outlet["area_threshold"] = np.full(df_outlet["X"].shape[0],
area_threshold)
df_outlet = pd.DataFrame(df_outlet)
df_outlet.to_csv(default_param["prefix"] + "outlets.csv", index=False)
df_outlet["ID"] = np.array(list(range(df_outlet.shape[0])))
this = mydem.cppdem.extract_perimeter_of_basins()
df_perimeter = {"X": [], "Y": [], "Z": [], "IDs": []}
for key, val in this.items():
df_perimeter["X"].append(np.array(val["X"]))
df_perimeter["Y"].append(np.array(val["Y"]))
df_perimeter["Z"].append(np.array(val["Z"]))
df_perimeter["IDs"].append(np.full(np.array(val["Z"]).shape[0], key))
## Log from the analysis
for key, val in df_perimeter.items():
df_perimeter[key] = np.concatenate(val)
pd.DataFrame(df_perimeter).to_csv(default_param["prefix"] +
"perimeters.csv",
index=False)
# You can add any extra dependencies you need here. For example pandas, numpy or matplotlib.
# First step is to load the DEM
## The name "mydem" can be changed into whatever suits you
## Let's have a clean and organised approach and save things into variables
path_to_dem = "/adapt/here/the/path/to/your/dem/" # You need to obviously adapt that path to your case
dem_name = "whatever_name.tif" # You also need to adapt that file name...
## Now we can load the dem into LSDTopytools:
### already_preprocessed can be turn to True if you are 100% sure that your dem does not need preprocessing before flow routines
mydem = LSDDEM(path = path_to_dem, file_name = dem_name, already_preprocessed = False)
# Alright the dem is in the system and now needs to be preprocessed (if not done yet)
mydem.PreProcessing(filling = True, carving = True, minimum_slope_for_filling = 0.0001) # Unecessary if already preprocessed of course.
#Need to pregenerate a number of routines, it calculates flow direction, flow accumulation, drainage area , ...
mydem.CommonFlowRoutines()
# This define the river network, it is required to actually calculate other metrics
mydem.ExtractRiverNetwork( method = "area_threshold", area_threshold_min = 1500)
# Defining catchment of interest: it extracts the catchments by outlet coordinates. You also need to adpat these obviously!!
## they need to be in the same coordinate system than the raster.
mydem.DefineCatchment( method="from_XY", X_coords = [532297,521028], Y_coords = [6188085,6196305])
# Calculates chi coordinate with an according theta
mydem.GenerateChi(theta = 0.4, A_0 = 1)
# At that stage you can get the basic river characteristics as follow:
# my_rivers = mydem.df_base_river
# my_rivers.to_csv("name_of_my_file_containing_base_river.csv", index = False) # This saves the base rivers to csv