from loguru import logger
from veros.diagnostics.diagnostic import VerosDiagnostic
from veros import veros_method
[docs]class NPZDMonitor(VerosDiagnostic):
"""Diagnostic monitoring nutrients and plankton concentrations
"""
name = 'npzd' #:
output_frequency = None #: Frequency (in seconds) in which output is written
restart_attributes = []
save_graph = False #: Whether or not to save a graph of the selected dynamics
graph_attr = { #: Properties of the graph (graphviz)
'splines': 'ortho',
'center': 'true',
'nodesep': '0.05',
'node': 'square'
}
def __init__(self, setup):
self.output_variables = []
self.surface_out = []
self.bottom_out = []
self.po4_total = 0
self.dic_total = 0
@veros_method
def initialize(self, vs):
cell_volume = vs.area_t[2:-2, 2:-2, np.newaxis] * vs.dzt[np.newaxis, np.newaxis, :] * vs.maskT[2:-2, 2:-2, :]
po4_sum = vs.phytoplankton[2:-2, 2:-2, :, vs.tau] * vs.redfield_ratio_PN\
+ vs.detritus[2:-2, 2:-2, :, vs.tau] * vs.redfield_ratio_PN\
+ vs.zooplankton[2:-2, 2:-2, :, vs.tau] * vs.redfield_ratio_PN\
+ vs.po4[2:-2, 2:-2, :, vs.tau]
self.po4_total = np.sum(po4_sum * cell_volume)
if vs.enable_carbon:
dic_sum = vs.phytoplankton[2:-2, 2:-2, :, vs.tau] * vs.redfield_ratio_CN\
+ vs.detritus[2:-2, 2:-2, :, vs.tau] * vs.redfield_ratio_CN\
+ vs.zooplankton[2:-2, 2:-2, :, vs.tau] * vs.redfield_ratio_CN\
+ vs.dic[2:-2, 2:-2, :, vs.tau]
self.dic_total = np.sum(dic_sum * cell_volume)
def diagnose(self, vs):
pass
@veros_method
def output(self, vs):
"""Print NPZD interaction graph
"""
if self.save_graph:
from graphviz import Digraph
npzd_graph = Digraph('npzd_dynamics', filename='npzd_dynamics.gv')
label_prefix = '\\tiny ' # should be selectable in settings allows for better exporting to tex
label_prefix = ''
# Create a node for all selected tracers
# Drawing edges also creates nodes, so this just ensures, we se it,
# when there are no connections to a node
for tracer, tracer_data in vs.npzd_tracers.items():
npzd_graph.node(tracer)
# If a tracer has the sinking_speed attribute indicate on the graph
if hasattr(tracer_data, 'sinking_speed'):
npzd_graph.node('Bottom', shape='square')
npzd_graph.edge(tracer, 'Bottom', label=label_prefix + 'sinking', lblstyle='sloped,above')
# Common source rules are split up into several rules
# This causes a duplication of the first registerd rule
# Which should not be shown, so if there is more than one of them selected,
# The source edge should not be displayed, because it is already there
skiprules = []
for name, rule in vs.common_source_rules.items():
# Construct the list of rule names
rule_names = [name + '_' + rule[0][1]] + [name + '_' + rule[i][2] for i in range(1, len(rule))]
# If there is more than one indicate, that it should be skipped from drawing
if len(rule_names) > 1:
skiprules.append(vs.npzd_available_rules[rule_names[0]])
# Draw primary rules
for rule in vs.npzd_rules:
if rule in skiprules:
continue
npzd_graph.edge(rule.source, rule.sink, label=label_prefix + rule.label, lblstyle='sloped, above')
# Draw pre rules dotted
for rule in vs.npzd_pre_rules:
if rule in skiprules:
continue
npzd_graph.edge(rule.source, rule.sink, label=label_prefix + rule.label, style='dotted', lblstyle='sloped, above')
# Draw post rules dashed
for rule in vs.npzd_post_rules:
if rule in skiprules:
continue
npzd_graph.edge(rule.source, rule.sink, label=label_prefix + rule.label, style='dashed', lblstyle='sloped, above')
self.save_graph = False
npzd_graph.render('npzd_graph', view=False)
"""
Total phosphorus should be (approximately) constant
"""
cell_volume = vs.area_t[2:-2, 2:-2, np.newaxis] * vs.dzt[np.newaxis, np.newaxis, :] * vs.maskT[2:-2, 2:-2, :]
po4_sum = vs.phytoplankton[2:-2, 2:-2, :, vs.tau] * vs.redfield_ratio_PN\
+ vs.detritus[2:-2, 2:-2, :, vs.tau] * vs.redfield_ratio_PN\
+ vs.zooplankton[2:-2, 2:-2, :, vs.tau] * vs.redfield_ratio_PN\
+ vs.po4[2:-2, 2:-2, :, vs.tau]
if vs.enable_carbon:
dic_sum = vs.phytoplankton[2:-2, 2:-2, :, vs.tau] * vs.redfield_ratio_CN\
+ vs.detritus[2:-2, 2:-2, :, vs.tau] * vs.redfield_ratio_CN\
+ vs.zooplankton[2:-2, 2:-2, :, vs.tau] * vs.redfield_ratio_CN\
+ vs.dic[2:-2, 2:-2, :, vs.tau]
po4_total = np.sum(po4_sum * cell_volume)
logger.diagnostic(' total phosphorus: {}, relative change: {}'.format(po4_total, (po4_total - self.po4_total)/self.po4_total))
self.po4_total = po4_total[...]
if vs.enable_carbon:
dic_total = np.sum(dic_sum * cell_volume)
logger.diagnostic(' total DIC: {}, relative change: {}'.format(dic_total, (dic_total - self.dic_total)/self.dic_total))
self.dic_total = dic_total.copy()
for var in self.output_variables:
if var in vs.recycled:
recycled_total = np.sum(vs.recycled[var][2:-2, 2:-2, :] * cell_volume)
else:
recycled_total = 0
if var in vs.mortality:
mortality_total = np.sum(vs.mortality[var][2:-2, 2:-2, :] * cell_volume)
else:
mortality_total = 0
if var in vs.net_primary_production:
npp_total = np.sum(vs.net_primary_production[var][2:-2, 2:-2, :] * cell_volume)
else:
npp_total = 0
if var in vs.grazing:
grazing_total = np.sum(vs.grazing[var][2:-2, 2:-2, :] * cell_volume)
else:
grazing_total = 0
logger.diagnostic(' total recycled {}: {}'.format(var, recycled_total))
logger.diagnostic(' total mortality {}: {}'.format(var, mortality_total))
logger.diagnostic(' total npp {}: {}'.format(var, npp_total))
logger.diagnostic(' total grazed {}: {}'.format(var, grazing_total))
for var in self.surface_out:
logger.diagnostic(' mean {} surface concentration: {} mmol/m^3'.format(var, vs.npzd_tracers[var][vs.maskT[:, :, -1]].mean()))
for var in self.bottom_out:
logger.diagnostic(' mean {} bottom concentration: {} mmol/m^3'.format(var, vs.npzd_tracers[var][vs.bottom_mask].mean()))
def read_restart(self, vs, infile):
pass
def write_restart(self, vs, outfile):
pass