def save_site(site_id):
bulk_site_dir = "/home/mikey/Desktop/aus-hydro-data/downloaded_data/data.water.vic/download.20151022110838/"
all_bom_sites_file = '/home/mikey/Desktop/aus-hydro-data/downloaded_data/stations.txt'
zipped_sites_dir = '/home/mikey/Desktop/aus-hydro-data/downloaded_data/bom'
# get levels and flows from data.vic.water
# site_id = "405230"
data_types = ["MeanWaterLevel", "MeanWaterFlow"]
site_details, data_types_values = vic_water_surface.read_bulk_downloaded_sw_file(bulk_site_dir, site_id, data_types)
dates = data_types_values["MeanWaterFlow"]["dates"]
flows = data_types_values["MeanWaterFlow"]["data"]
level_dates = data_types_values["MeanWaterLevel"]["dates"]
levels = data_types_values["MeanWaterLevel"]["data"]
lat, lng = (float(site_details["Latitude"]), float(site_details["Longitude"])) # TODO could use geofabric AHGFCatchment to find center of catchment
closest_names, closest_ids, closest_locations = bom.closest_first_bom(all_bom_sites_file, lat, lng, dates[0])
# get climate from BOM
closest_rain_i, closest_rain_dates, closest_rain_data = bom.closest_obs('136', dates, closest_names, closest_ids, closest_locations, zipped_sites_dir, bom.bom_obs_types)
closest_max_temp_i, closest_max_temp_dates, closest_max_temp_data = bom.closest_obs('123', dates, closest_names, closest_ids, closest_locations, zipped_sites_dir, bom.bom_obs_types)
closest_min_temp_i, closest_min_temp_dates, closest_min_temp_data = bom.closest_obs('122', dates, closest_names, closest_ids, closest_locations, zipped_sites_dir, bom.bom_obs_types)
# keep intersections
multi_names = ["flows", "levels", "closest_min_temp", "closest_max_temp", "closest_rain"]
multi_dates = [dates, level_dates, closest_min_temp_dates, closest_max_temp_dates, closest_rain_dates]
multi_series = [flows, levels, closest_min_temp_data, closest_max_temp_data, closest_rain_data]
intersection_i = utils.intersection_indices(multi_dates)
for i in range(len(multi_dates)):
multi_dates[i] = multi_dates[i][intersection_i[i]]
multi_series[i] = multi_series[i][intersection_i[i]]
assert numpy.all(multi_dates[i] == multi_dates[0])
assert (multi_dates[i][-1] - multi_dates[i][0]).days == (len(multi_dates[i]) - 1) # check contiguous, TODO: otherwise add nans then interpolate
print "SITE", lat, lng, dates[[0,-1]]
print bom.bom_obs_types
write_data = {
"site_details": site_details,
"closest_details": {
"closest_rain": { "name": closest_names[closest_rain_i], "id": closest_ids[closest_rain_i], "location": closest_locations[closest_rain_i], },
"closest_max_temp": { "name": closest_names[closest_max_temp_i], "id": closest_ids[closest_max_temp_i], "location": closest_locations[closest_max_temp_i], },
"closest_min_temp": { "name": closest_names[closest_min_temp_i], "id": closest_ids[closest_min_temp_i], "location": closest_locations[closest_min_temp_i], },
},
"dates" : [d.strftime("%d/%m/%Y %H:%M:%S") for d in multi_dates[0]],
}
for i in range(len(multi_dates)):
write_data[multi_names[i]] = multi_series[i].tolist()
with open(os.path.join("sw_data", site_id+'.json'), 'w') as f:
f.write(json.dumps(write_data))
def load_sites(): bulk_site_dir = "/home/mikey/Desktop/aus-hydro-data/downloaded_data/data.water.vic/download.20151022110838/" multi_dates = [] multi_flows = [] multi_levels = [] for site_id in gauges: # get levels and flows from data.vic.water # site_id = "405230" data_types = ["MeanWaterLevel", "MeanWaterFlow"] site_details, data_types_values = vic_water_surface.read_bulk_downloaded_sw_file(bulk_site_dir, site_id, data_types) dates = data_types_values["MeanWaterFlow"]["dates"] flows = data_types_values["MeanWaterFlow"]["data"] level_dates = data_types_values["MeanWaterLevel"]["dates"] levels = data_types_values["MeanWaterLevel"]["data"] multi_dates.append(dates) multi_flows.append(flows) multi_levels.append(levels) flow_to_level # keep intersections intersection_i = utils.intersection_indices(multi_dates) for i in range(len(multi_dates)): multi_dates[i] = multi_dates[i][intersection_i[i]] multi_flows[i] = multi_flows[i][intersection_i[i]] for i in range(len(multi_dates)): plt.plot(multi_dates[i], multi_flows[i], label= gauges[i]) plt.legend() plt.show()
