Title: A2
1The food web
Death and sedimentation
A2
D
A1
Detritus and associated Microflora
(bacteria/fungi)
Primary producers
inedible
H2
Primary consumers
H1
detritivore
herbivore
P
Productivity
Biomass
Secondary Productivity Primary production
supports a web of consumersa simple example
2Defining some productivity terms
bB/t
Birth (production) term
B
Death (loss) term
mB/t
3 The productivity is a combination of the birth
of new organisms and the growth of the organisms
already present Similarly, the death process is
a combination of death of organisms and weight
loss by existing organisms. If the productivity
(birth term) exceeds the death term the biomass
is increasing, and if the death term is larger,
the biomass is decreasing
Biomass of a consumer
Time (t)
4So if we measure and B at any point in time
and we can estimate the specific birth rate, we
can then obtain the specific death rate by
subtraction.
5A tiny organism like a rotifer is born at full
size, so productivity amounts to measuring the
rate at which new animals are born
Rotifers carrying eggs
- For a tiny consumer like a rotifer the birth rate
is easy to estimate since the adult females carry
their eggs around until they hatch - When they hatch they come out as full sized
rotifers - .
- If we know the fraction of adults carrying egs
and the average time it takes for eggs to hatch,
we can calculate the birth rate. -
- Since the rotifers are born more or less full
size, so there is no need to model or measure the
growth of individuals.
6(No Transcript)
7For a large organism like a fish, biomass
production occurs mostly from individual growth.
New born fish are so tiny that birth of
individuals makes a negligible contribution to
biomass production
The Wt represent the Weights of each year class
- We can calculate the growth rate of biomass
individual fish by weighing fish and determining
their age and then seeing how much weight they
gain each year. - The productivity of each age class is the
Specific growth rate (SGR) of that age class
times the total biomass of that age class in the
population.
8This approach assumes that the size vs age
relationship is relatively constant.
The Wtrepresent the Weights of each year class
9By this method the average SGR for the whole
population can be calculated as the weighted
average over all age classes
However there is a problem with this approach???
10However there is a problem with this approach.
By considering only the gain in weight across age
classes, this method ignores weight gained and
lost within the same year, eg Gonad tissue
Adult fish usually convert a considerable
portion of their body mass to gonads and release
it during spawning every year. Thus it does not
add to next years weight and would not be
recorded as growth We can quite easily correct
for this by factoring in gonad production (add
GSI)
We would of course only make this GSI correction
on adult age classes.
11How can we tell how old a fish is?
Scales of a chum salmon
Measure distances from scale center to
each annulus along a chosen axis
2
3
4
12(No Transcript)
13This allows us to construct a growth curve based
on length and age.
Convert the growth curve based on length to
weights using a length-weight plot for the species
14Many types of bony structures are commonly used
to determine age of fish
Opercular bone
Otolith
Scale
These three structures are all from the same 3
year old 30 cm cutthroat trout
15The specific death rates can also be estimated
from the population structure. This time we
assume that the age structure of the population
is constant, and that the numbers of individuals
of each age within a sample reflects the
proportion of that age group in the
population. Assume we have a sample of 215 pike
from a population.
The Nt / ? Nt represent the proportion of the
population in each year class
16150
The survivorship curve assuming stable age
structure for the population looks like this
110
100
m1 0.69
55
50
30
m2 0.61
20
m3 0.41
0
1
3
5
0
Age
17Summary
Productivity term
loss term
18Ecological efficiency (ln) for a consumer at the
nth trophic level
Ecological efficiency of zooplankton is usually
around 10 of NPP in lakes Variability??
19Zooplankton such as Daphnia filter-feed using
currents generated by their thoracic appendages.
Fecal pellets sediment rapidly to the bottom
2080-95 of energy is lost at each trophic step,
much of it as feces
Assimilation efficiency Herbivores depends on
diet 100 for sugary nectar 40-80 for small
phytoplankton and filamentous algae lt20 for mud
and detritus Carnivores 60-70 for aquatic
insects 70-90 for meat
The undigested material in the zooplankton fecal
pellets was not assimilated. Assimilation
efficiency depends on the digestibility of the
diet Cellulose, chitin, lignin or other
undigestible material makes AE low
Ingested energy - egested energy assimilated
energy Assimilation efficiency (AE, )
assimilated energy/ingested energy x 100
21Exploitation efficiency or Consumption Efficiency
(EE)
Exploitation efficiency is the consumption rate
at a given trophic level divided by the
productivity of the trophic level it feeds
on. Zooplankton will have low EE (CE) when
phytoplankton are sedimenting rapidly to the
bottom before they are being eaten. If EE(CE) is
high then most of the sedimentation will be in
the form of fecal pellets, which sink more
rapidly than individual cells. Zooplankton fecal
pellets are good food for benthic
invertebrates If EE for herbivorous zooplankton
is low then dead (sedimenting) phytoplankton will
be readily available for detritivores
(zoobenthos)
22Activity is energetically expensive and high
Metabolic rate means low Production efficiency
Gross PE Endotherms 5 or less 1 some
birds Ectotherms 10-30 for fish 5-15 insects
Otter swim about rapidly and spend large amounts
of energy looking for fish to eat
Assimilated energy - respiration - excretion
production (growth) Net Production efficiency
(NPE, ) growth/assimilation x 100 Gross PE
()assim/ingest x growth/assim
x100growth/ingest x 100
23Pelagic fish like kokanee salmon expend a huge
amount of energy actively searching for
prey--they have high basal metabolic rates low
conversion efficiencies
The deepwater sculpin sits on the bottom and
ambushes unsuspecting prey. They have very low
basal metabolic rates and high conversion
efficiencies
If these two species were fed the same amount of
food, the sculpin would grow more than twice as
fast as the salmon
24Copepod dominated communites have lower trophic
efficiency than cladoceran dominated
communitiespossible reasons?
Filter-feeding by a calanoid copepod
- Copepods are rapid swmmers and generate feeding
currents as they swim - Copepods filter-feed by generating currents with
their 1st antennae and their thoracic
appendages.. - Water from small eddy currents around the
mouthparts is drawn over the fine setae of the
maxillae, where small algae are collected and
moved to the mouth.
http//www.ucmp.berkeley.edu/arthropoda/crustacea/
images/copepoda03.jpg
25Energy budget for herbivorous zooplantkon
NPP rate of formation of phytoplankton
biomass S rate of production of uneaten
algae, mostly inedible species (sedimentation) F
rate of production of fecal pellets
(sedimentation)
Metabolic costs include basal metabolism,
activity costs and specific dynamic action (costs
of digestion etc) Zooplankton production is the
rate at which biomass (energy) becomes available
for consumption by zooplanktivores
26Energetic losses in the food chain Less than 1
of the incident light energy is captured by
photosynthesis as NPP. Productivity declines
by about 10-fold for each trophic level in the
food chain. Most of the losses are are in the
form of waste heat. Some energy from each
trophic level winds up in the detrital pool, and
some of this remains buried as sediment (or
soil) organic matter (fossilized)
27(No Transcript)
28Productivity at different levels in the food web
500 x (0.1)3 g/m2/yr
500 x (0.1)2 g/m2/yr
500 x 0.1 g/m2/yr
NPP around 500 g/m2/yr
Net productivity at level n the rate of growth
of biomass at that level SGR GSI Biomass
NPP (TE) n-1
29Yields of piscivorous fish are well correlated
with primary productivity but are several orders
of magnitude lower than PP
30- Intensive aquaculture can produce yields that are
orders of magnitude beyond natural ecosystems
How to maximize energy flow to fish Increased
nutrient loadingfertilization ammonia and
anoxia tolerant species Shortening the food
chainprimary consumers (eg carps, tilapia or
mullets) Dont rely on natural recruitment and
managing the life cyclestocking/hatcheries Increa
sing consumption efficiencysmall pens intensive
feeding Increased assimilation efficiencyfeeding
with easy to digest food pellets Increased
production efficiencylow activity species that
dont mind crowding, , highly turbid water
31Many aquaculture proponents argue that
aquaculture reduces harvesting pressure on wild
fisheries. Salmonid aquaculture not very
trophically efficient, food pellets made from
by-catch of wild species Major water quality
issuesnutrientpollution from cages, anti-fouling
paint, antibiotics, habitat destruction Transmit
diseases to wild salmonidsbacteria, viruses,
protozoans, fungi, fish lice parasitic
copepods and other Crustacea Genetic problems
when domestic escapees compete with or
interbreed with wild fish
Lepeophtheirus salmonis
Argulus
32- Summarizing concepts on Secondary production
- The organic matter produced by primary producers
(NPP) is used by - a web of consumers
- NPP is used directly by primary consumers
(herbivores and detritivores), which are in - turn consumed by carnivores.
- Measurement of 2o Production is done by
estimating the rate of growth of individuals - and multiplying by the number of individuals per
unit area in the cohort (age or size group). - The efficiency of secondary production ranges
from 5-20 (Avg 10) - at each trophic level.
- Efficiency depends on several factors--palatabilit
y, digestibility, energy requirements - for feeding (activity costs)(eg homeotherms vs
poikilotherms , other limiting factors - eg water, and nutrient quality of food.
- Trophic efficiency can be represented as the
product of CEAEPE, each of which
33(No Transcript)
34- If the productivity of a phytoplankton population
is 4000 k J (kilo Joules) /yr / m2, If
sedimentation rate of dead cells to the substrate
constitutes 1600 kJ/m2yr, and the phytoplankton
population is dB/dt0. If the rain of
zooplankton fecal pellets to the bottom is 1400
kJ/m2/yr. What is the assimilation efficiency of
the zooplankton trophic level (assume that they
are all feeding on phytoplankton). - 0.42 or 42
- 0.60 or 60
- 0.35 or 35
- 0.25 or 25
- None of these
- What is the exploitation efficiency EE (or
Consumption efficiency CE) of the zooplankton
trophic level - 0.42 or 42
- 0.60 or 60
- 0.35 or 35
- 0.25 or 25
- None of these
35- If the net production efficiency of the
zooplankton trophic level is 0.40 (40) what is
the ecological efficiency (l) of the trophic
level - 0.15 or 15
- 0.05 or 5
- 0.10 or 10
- 0.25 or 25
- None of these
- If the zooplanktivorous fish are consuming
zooplankton at the rate of 400 kJ/yr/m2, their EE
(CE) is - 0.40 or 40
- 0.60 or 60
- 1.00 or 100
- 0.25 or 25
- None of these
36- If the zooplanktivorous fish have an assimilation
efficiency of 0.70 (70) and Net production
efficiency (NPE) of 0.20 (20), the productivity
at this trophic level is - 40 kJ/yr/m2
- 56 kJ/yr/m2
- 100 kJ/yr/m2
- 280 kJ/yr/m2
- None of these
- If in another lake with similar zooplankton
productivity the planktivore fish productivity
was 2 X higher, a possible explanation for this
would be - the AE of the fish in that lake was 2X as high
- the NPE of the fish in that lake was 2X as high
- the EE (CE) in that lake was 2X as high
- the AENPE in that lake was 2X as high
- both b and d are true
- both b and c are true
37Residence time and turnover of energy by trophic
levels
The standing stock of energy in the plankton is
low but it is turned over rapidly, because the
organisms are small, grow rapidly and dont live
long
A
Phytoplankton (0.01mg, life span, few days
H2
H1
Planktonic Herbivore (50mg) life span 1 month
Benthic Detritivore (0.1 g) life span 1yr
P
Carnivorous fish (100g) life span 5-10 yr
Turnover is slower at higher trophic levels,
since larger organisms accumulate energy over a
longer life spanlonger residence time and slower
turnover