Catadromous species such as the eel often present several distinctive featu
res (high plasticity in somatic growth, size-dependent sexual maturation, s
trongly skewed sex ratio, variable recruitment) that make their management
quite complex. Fisheries usually operate in lagoons where effort is exerted
on the sexually mature individuals migrating back to the sea, or on the yo
ung prereproductive individuals feeding in the lagoons, or on both. The aim
of this work is to investigate the possibility of improving the economic r
eturn of the most typical catadromous fish, the European eel Anguilla angui
lla, by exploring which fraction of prereproductive individuals should be c
aught every year, and whether declining recruitments from sea should be sup
plemented by restocking young eels (called elvers). Eel population dynamics
is described through a demographic model based on a multiple classificatio
n of individuals by age and size and which explicitly includes environmenta
l stochasticity and variation in model parameters. The model has been calib
rated on the data from the most important Italian eel fishery located in th
e Comacchio lagoons, where only the migrating, mature eels are traditionall
y harvested. Different policies in terms of fishing effort, net selectivity
, and restocking density have been analyzed by using the simulation model.
Optimal management has been assessed with reference to the average net econ
omic return under different discount rates. A Monte Carlo approach has been
used to evaluate the variability of the mean economic benefit. Our model p
redicts that substantial improvement, relative to the traditional harvestin
g policy, can be achieved by restocking elvers, while catching yellow eels
with fyke nets allows only for minor improvements of economic return. A sen
sitivity analysis shows that these results hold for different discount rate
s and for realistic variations of harvesting costs, but if the present pric
e gap between yellow and silver eels were to increase, yellow eel fishing w
ould be unprofitable. Also, elver restocking should be performed only in lo
w-density populations. Otherwise the density-dependent increment in natural
mortality may over-compensate the benefits of restocking. The conceptual f
ramework used for modeling the specific case of European eels demography is
very flexible and can be easily extended to a variety of other catadromous
species.