Mosquito Protocol

Document Sample
Mosquito Protocol Powered By Docstoc
					                              Mosquito Protocol
Purpose                                             Scientific Inquiry Abilities
To sample, identify and count the number of           Identify answerable questions.
mosquito larvae in each container at your study       Use appropriate mathematics to analyze data
site.                                                 Develop descriptions and explanations using
Overview                                              Recognize and analyze the alternative
Students will collect, sort, identify and count the      explanations
number of mosquito larvae from the                    Communicate the procedures and the
indoor/outdoor containers at your study site.            explanations.

Student Outcomes                                   Time
Students will be able to                           1-2 hours to collect samples, count, identify
   Identify mosquito larvae at your study site     and preserve specimens (excluding travel).
   Understand the importance of representative     Time will vary with the abundance and
      samplings                                    diversity of water containers in each study site.
   Plot a relative abundance graph of the number
      of mosquito larvae                           Level
   Compare the number of mosquito larvae in        Primary and Secondary
      each species in the different containers
   Explore relationships between the larvae        Frequency
      species and climatic factors                 Once a month
   Collaborate with other GLOBE schools (with
      your country or other countries)           Materials
   Share observations by submitting data to the  Mosquito Field Guide
      NBIDS website.                             Mosquito Site Definition Field Guide
                                                 Mosquito data sheets
Science Concepts                                 Equipment used to collect mosquito larvae at
Life Science                                     your site: fishnets
   Organisms have basic needs.                   Many clear plastic bags and elastic bans
   Organisms can just only survive in the Mosquito Larvae identification key
      environments where their needs are met.    GPS Protocol Field Guide
   Earth has many different environments that GPS Protocol Data Sheets
      support     different  combinations     of Markers
      organisms.                                 Microscope
   Humans can change natural environments.       Pens/Pencils
   All organisms must be able to obtain and use
      the resources while living in a constantly Preparation
      changing environment.                      Decide upon study locations and the household
   All populations are living together and the to be selected.
      physical factors with which they interact Practice using fishnet
      constitute an ecosystem.                   Practice identifying mosquito larvae using
   The interaction of organisms has evolved mosquito larvae key.
      together over time.

Walailak University 2007: Mosquito                                                                     1
Mosquito Protocol- Introduction                are willing to work with students. These
                                               people can, for example, help identifying
The mosquito is a member of the family         mosquito up to family level and discussing
Culicidae. These insects have a pair of        the important indicator species, as well as
scaled wings, a pair of haltered, a slender    endemic and introducing to organisms
body, and long legs. The females of most       present in your area. Mosquito Larval keys
mosquito species suck blood from man           are available in printed manuals and
and other warm-blooded animals. Size           books. Select and identification key that is
varies but is rarely greater than 15 mm (0.6   applicable to your locality.
inch). Mosquitoes weigh only about 2 to
2.5 mg. They can fly at about 1.5 to 2.5       Contact local experts in the area to make
km/h (0.9 to 1.6 mph). Mosquito have           sure that you are not sampling at a site
been around for 170 million years.             where other people are conducting
                                               research or where there are endangered
For the Mosquito Protocol, we want to          area. You do not want to inadvertently hurt
estimate the disease risk area. Most often,    a long-term monitoring site or harm
it is impossible to count all individuals of   endangered area.
every species present in the habitat. So we
sample mosquito larvae in the habitat, and     To have the students become familiar with
calculate the Container Index (CI), House      mosquito larval identification before you
Index (HI), and Breteau Index (BI). Each       go to the field, students can bring in
larval index can tell us about the number      mosquito larvae from their neighborhoods
of mosquitoes in our study area, so we can     to identify in class.
use these data to echelon significance in
diseased control.                              Site Definition and Mapping
                                               Students use the stratified random
Background                                     sampling technique to sampling the
You have to read mosquito fact in              households in their study sites. Students
Appendix.                                      collect mosquito larvae from both in the
                                               indoor and outdoor containers. The indoor
Teacher Support                                containers are categorized into two
                                               categories: (1) inside the bathroom such as
                                               small earthen jar, large earthen jar, cement
Advanced Preparation
                                               tank and plastic container and (2) outside
Many teachers and students have a little
                                               the bathroom such as ant-guard, vase,
background in the identification of
                                               refrigerator with plate and water plant pot.
mosquito larvae, and may be reluctant to
                                               For outdoor containers, there are two
begin such a class project. This is not a
                                               categories: (1) artificial containers and (2)
problem, since students find the critters so
                                               natural containers. Artificial containers
fascinating, they will be teaching
                                               consist of small earthen jar, large earthen
themselves and each other.
                                               jar, cement tank and plastic container, old
                                               can, plastic bottle, metal box, plant plate,
There are many local experts to call on.
                                               plant pot, animal pan, preserved areca jar
Often, local mosquito monitoring groups

Walailak University 2007: Mosquito                                                        2
and discarded tire. Natural containers         different tasks. For example, two students
consist of areca husk, banana tree, coconut    can hold the fish nets, one student can hold
shell, tree hole and clump. Mosquito           a plastic bag, one student can read the
larvae are collected from all outdoor          instructions aloud, and etc.
containers within 15 meters around the
house.                                         The most time-consuming tasks are the
                                               sorting and identifying the mosquito
                                               larvae. To save times, we can divide
When to Go Sampling                            students into two teams. Students from
You should collect the mosquito larvae         Team I do the sorting, counting, and
once a month.                                  identifying the mosquito larvae using the
                                               Sorting, Identifying and Counting the
Supporting Protocols                           mosquito larvae Protocol Lab Guide.
Atmosphere Protocol: Students can              Students from team II can be collecting a
explore some relationship between some         second sample.
atmospheric        data         (i.e.  the
maximum/minimum temperature, relative          After the students collect mosquito larvae,
humidity, the number of rainy day, and the     teachers look at the jars of sorted mosquito
amount of rainfalls) and types of mosquito     larvae to verify that all students identify
larvae found at the study site.                mosquito larvae correctly. If not, gather
                                               the students and have them discuss the
Hydrology Protocol: Students can explore       differences and identify mosquito species
some association between some hydrology        correctly.
data (i.e. water pH, temperature,
transparency, the amount of alkalinity, and    After all mosquito larvae are sorted and
dissolved oxygen) and types of mosquito        combined from the teams in separate jars
larvae found at the study site.                for each species, have a committee of
                                               students and yourself look at the larvae to
                                               make sure that you all agree on mosquito
Preparing for the Field
                                               identification. Then, we proceed by
This is a sampling method. It would be a
                                               counting mosquito larvae and report the
good idea to select a study site before the
                                               data on one set of data sheets and
day of sampling.
                                               annihilate all of them.
All students in the field should have a
baseball cap, sneaker, and wear warm suit.     Measurement Procedures
                                               Do not sample containers (container,
If available, you can take folding tables or
                                               puddle, and etc.) that cannot be reached
seat desks for the students to handle and
                                               safely. If your students sample from
count their samples in the field.
                                               multiple habitats, you should determine
                                               which habitats can be sampled safely and
Managing Students in the Field                 evaluate the percentage of coverage of
If you have a large class, you should have     each accessible habitat. Record the
students work in multiple teams. Students      habitats which could not be sampled.
in each team can be responsible for

Walailak University 2007: Mosquito                                                       3
Students should only sort and count the       Helpful Hints
number of mosquito larvae. Tadpole, small     As scientists do, students should keep field
fish and other organism should be             notes of your procedures to report what
removed from the samples and returned to      you did and if there are any deviations
the water.                                    from your plans. Make a photo journal of
                                              your trip, and bring parents or older
We only count the number of live
                                              GLOBE students to mentor. Enjoy
mosquito larvae. To sort mosquito larvae
                                              learning about mosquito species in the
in each genus, we use a small plastic spoon
                                              world around you!!
to collect mosquito larvae, sort them up in
genus level (i.e. Aedes, Anopheles, and       Having the students work in teams will
Culex spp.) and place them in small plastic   make sample collections, sorting and
cups. We sort Aedes mosquito larvae up to     identifying quicker. To work in groups,
a species level in the laboratory by using    though requires more equipment such as
microscope. We discarded all mosquito         fishnet, plastic bags, trays, and mosquito
larvae after we are done.                     larval identical keys, can be more fun.

Voucher specimens are not required, but
                                              The Questions Frequently Asked.
they may help with teaching the students
                                                 1. What is the life cycle of
how to properly identify the mosqito
larvae before going into the field. By
                                                    A: Adult     eggs (2 -3 days)
collecting voucher specimens each time,
                                                    larvae (4 -5 days)    pupae (1- 2
the specimens can be compared to make
sure that identifications are being done
correctly each time.                             2. How do you identify which one is
                                                    the Anopheles, Aedes or Culex
Equipment Use and Maintenance                       larvae (identify with eyes)?
All of the sampling materials are available         A: We can see the characteristics of
commercially, but students can also enjoy              mosquito larvae: Anopheles
making them using the instructions                     larvae cling parallel with water
provided in the Instrument Construction                surface. On the other hand,
section. You can also buy some parts and               Aedes and Culex larvae cling in
make others. For example, one can buy a                angle     of 45° with bank     of
0.5 mm-mesh replacement net for a D-net                container. Aedes larvae have
and make the pole. This is less expressive             short siphon but Culex larvae
than buying the whole device.                          have long siphon.

                                                 3. What does the mosquito male feed
Student Assessments
The student should collect, identify the
                                                    A: Male mosquitoes eat nectar
mosquito larvae and calculate the House
                                                    from flower.
Index, Container Index and Breteau Index.
These indices indicate the risk of DHF

Walailak University 2007: Mosquito                                                      4
    4. At what seasons of the year are            A: We should use a small fishnet to
       greater percentages of mosquito            collect mosquito larvae from
       larvae found?                              discarded tires. If there are few
       A: We found in rainy season more           amount of water in the discarded
       than in winter and summer seasons.         tire, we pour water in large buckets
                                                  and collect mosquito larvae from
    5. What kind of containers that               large buckets.
       female mosquitoes prefer to lay
                                            Questions for Further Investigation
       A: The preferred containers are
                                               1. Are there any relationships among
       water jar, cement tank, and areca
                                                  mosquito larvae and your climatic
    6. How can we collect mosquito             2. Are there seasonal variations to the
       larvae from discarded-tires?               number and type of mosquito
                                                  larvae at your study site? If so,
                                                  suggest some possible reason why?

Walailak University 2007: Mosquito                                                  5
  squito P
Mos      Protocol

    d     e
Field Guide
       the                   o
Count t number of mosquito larvae in e              c
                                         each water container
       y          larvae up to species lev
Identify mosquito l          o           vel

What You Need
   Plast bags
   Rubb bans
      manent mark
   Perm           kers
   Stere          pe
   Mosq           ification Ke
       quito Identi          ey
   Mosq           Sheets
       quito Data S


In the Field
                  water contai
    1. Locate all w                      d          e          ool.
                              iners in and around the house/scho
                 wn           ner
    2. Write dow a contain ID on the water container (students c                    ray
                                                                          can use spr or a
                                        m).         rite      me          er
       permanent marker to mark them Then wr the sam containe ID on yo Data         our
                  mosquito lar
    3. If we find m                       er        rs,                   em
                              rvae in wate container we collect all of the with a fishnet.
           3.1. La                                  c          t          ore       5
                  arge water containers are water containers that can sto water 500 L or
                  reater such as water jar water poo cement tank and etc We samp large
                 gr                       r,        ol,        t           c.        ple
                 water contain
                 w            ners by dip           et                    ng
                                         pping the ne in the water, startin at the top of the
                  ontainer, co
                 co                      o          m
                             ontinuing to the bottom in a swir             n
                                                               rling motion and samp pling all
                  dges of the container (F
                 ed           c          Figure 1).

                   re         c          ampling tec
               Figur 1. Large container sa                               motion.
                                                   chnique in a swirling m

       k             007: Mosquito
Walailak University 20           o                                                          6
                  mall water containers are comp
            3.2. Sm                     s                   f           es,
                                                 posed of flower vase plastic bottles,
                  oconut shell and etc. W empty water out thr
                 co          ls         We      w                      ishnet (Figu 2).
                                                            rough the fi          ure

                        Figure 2. Small ccontainer sam          hnique.
                                                      mpling tech
    4. Place mosq             e          me
                  quito larvae with som small am                w                  ner
                                                     mount of water from the contain in a
                              t          bber ban wit some air in the bag.
       plastic bag and close it with a rub           th

                 n         ner                     g
    5. Write down a Contain ID on a plastic bag with a per    rmanent ma             w
                                                                          arker that we found
                 arvae. Then record the container ID in Data Sheet.
       mosquito la         n                        D         S
    6. Bring mosq           e          fy
                 quito larvae to identif up to sp  pecies level in the lab oratory by using a
                 oscope and Mosquito Id
       stereomicro                                 on                    n
                                        dentificatio Key. Record data on a Data She  eet.

       k             007: Mosquito
Walailak University 20           o                                                         7
Learning Activity
How to Use a Microscope
Purpose                                     Time
Students understand how to use a For 3 hours for practice.
stereomicroscope for mosquito larval
identification.                             Level
Students collect mosquito larvae from study Materials and Tools
sites, and identify mosquito larvae up to Equipment is listed on Activity Sheets
species level by using stereomicroscope and Mosquito Data Sheet
mosquito identification key.                Microscope

Student Outcomes                     Preparation
Students     will learn how  to  use Learn about a stereomicroscope as you can.
stereomicroscope for mosquito larval

Science Concepts
Earth and Space Science
  Mosquitoes lay their eggs in many types of
    water containers both indoor and outdoor
    around a house.
  Mosquito larvae breed in stagnant water
    filled containers.
 Students can identify mosquito larva
    species by themselves.

Background                                     magnification.      However,    compound
Stereomicroscopes     allow      magnified     microscopes of the time suffered from
images of illuminated specimens to be          chromatic and spherical aberrations in the
viewed using 2 lenses (an objective and an     lens, which give single lens microscopes
eyepiece lens). Microscopes that use two       an advantage both in clarity and
lenses are called compound microscopes.        magnifying power. In the 19th century,
Single lens microscopes, of which this         however, these problems with the
antique created by Leeuwenhoek (1632-          compound microscope are successfully
1723) is an example, use only a single lens    resolved. Microscopes have been designed
to magnify the specimen. Compound              based on experience and the design had no
microscopes were first invented in the late    scientific grounds. A German named Ernst
16th century in Holland, when Zacharias        Abbe (1840-1905), however, established a
Janssen and his father discovered that         method to design microscopes utilizing
using two lenses greatly aided in              logical     calculations.   Since    then,

Walailak University 2007: Mosquito                                                     8
microscope design has progressed rapidly      upward). Move it as far as it will go
to the present day, with a wide array of      without touching the slide!
advanced microscopes now available
depending on specimen and research            5. Now, look through the eyepiece and
purpose. (       adjust the illuminator (or mirror) and
tech/ 1-0-1.aspx)                             diaphragm for the greatest amount of light.

Teacher Support                               6. Slowly turn the coarse adjustment so
Advanced Preparation                          that the objective lens goes up (away from
A teacher should be explaining about a        the slide). Continue until the image comes
microscope in class for 2-3 hours.            into focus. Use the fine adjustment, if
                                              available, for fine focusing. If you have a
What to Do and How to Do It                   microscope with a moving stage, then turn
Ask students about their knowledge of         the coarse knob so the stage moves
mosquitoes. Begin with questions such as:     downward or away from the objective
    • What are mosquito species in the        lens.
    • What are diseases caused by             7. Move the microscope slide around so
       mosquitoes as vectors that you         that the image is in the center of the field
       know?                                  of view and readjust the mirror,
Divide the students into two or three         illuminator or diaphragm for the clearest
groups. Each group is composed of 4-5         image.
students. Identify mosquito larvae by using
a microscope as the instrument.               8. Use the fine adjustment, if available. If
                                              you cannot focus on your specimen, repeat
1. When moving your microscope, you           steps 4 through 7 with the higher power
always carry it with both hands, grasp the    objective lens in place.
arm with one hand and place the other
hand under the base for support.              9. The proper way to use a monocular
                                              microscope is to look through the eyepiece
2. Turn the revolving nosepiece so that the   with one eye and keep the other eye open
lowest power objective lens is "clicked"      (this helps avoid eye strain).
into position.
                                              10. Do not touch the glass part of the
3. Your microscope slide should be            lenses with your fingers. Use only special
prepared with a cover glass over the          lens paper to clean the lenses.
                                              11. When finished, lower the stage, click
4. Look at the objective lens and the stage   the low power lens into position and
from the side and turn the coarse focus       remove the slide.
knob so that the objective lens moves
downward (or the stage, if it moves, goes     12. Always keep your microscope covered
                                              when not in use.

Walailak University 2007: Mosquito                                                      9
Example of the number of mosquito larvae in various water containers.
Study site: Muang district, Nakhon Si Thammarat.
House       Container          No. of       Ae.        Ae.       Culex   Anopheles
 ID           type           containers   aegypti   albopictus    spp.     spp.
  1      Flower vase                 1      0           0         0         0            0
         Flower pot
                                     1      0           0         0         0           0
         Pot plants                  1      3           0         0         0          0
         Cement tank                 1      10          0         3         20         12

  2      Pot plants                  1      0           0         0         0          0
         Cement tank                 1      20          0         5         5          15
         Ant guard                   1      0           3         0         0          0
         Ant guard                   2      0           5         0         0          0
         Ant guard                   3      0           0         0         0          0
         Ant guard                   4      0           0         0         0          4

         Flower pot
  3                                  1      0           0         0         0           0
         Water jar                   1      5           0         0         0           3
         Tire                        1      0           3         0         12          0
         Tire                        2      0           5         0         5           3

  4      Cement tank                 1      12          0         8         6           8
         Water jar                   1       8          0         0         0          10
         Tire                        2       0          6         0         10          0

  5      Tire                        1      0           0         0         3           0
         Tire                        2      0           7         0         8           3
         Flower vase                 1      0           0         0         0           1
         Flower vase                 2      0           0         0         0           5
         Water jar                   1      10          0         0         0           8
         Pot plants                  1      2           0         0         0           0

In the table, the numbers that were shown in Italic style represent the number of mosquito
identified in the laboratory.

Walailak University 2007: Mosquito                                                          10
How to Use a Microscope
Activity Sheet
  Plastic cups
  Plastic pools
  Plate dishes
  Data Sheets

What to Do
  1. Place the mosquito larva to be identified into a petri dish or other small tray (not the
      channeled sorting tray as it is too difficult to light the animal correctly).
  2. Cover the mosquito larva with enough water so that the incident light does not cause
      glaring bright patches on the animal (as this makes it difficult to see the key features)
      but not so much liquid that the mosquito larva floats away every time you let go.
  3. Good lighting is imperative, keep the incident light on full unless the view becomes
      too shiny. The transmitted light is useful for discrimination of setae and antennae.
      Some dark beetles will also require side lighting to see the grooves in hard dark
      surfaces of the elytra.
  4. For identification of the mosquito larva, the key generally includes characters of the
      whole animal in dorsal or ventral view.
  5. Be sure the animal is well focused at all times and that the particular part of the
      animal is in focus and being viewed from the correct angle
  6. You MUST interact with the microscope and the mosquito larva. Zoom in and out to
      find the best perspective of the feature in relation to the rest of the mosquito larva.

Walailak University 2007: Mosquito                                                          11
Learning Activity
Practice Identifying Mosquito Larva Activity
Purpose                                           All organisms must be able to obtain and use
Let students learn about collecting and               resources while living in a constantly
identifying the mosquito larvae in water              changing environment.
containers in each study site.                    All populations living together and the
                                                      physical factors with which they interact
Overview                                              constitute an ecosystem.
Students will collect mosquito larvae from        The interaction of organisms has evolved
indoor/outdoor containers at their study sites.   together over time.

Student Outcomes                                  Scientific Inquiry Abilities
Students will be able to                          Identify answerable questions.
- identify the mosquito larvae at their site;     Use appropriate mathematics to analyze data.
- understand the importance of                    Develop descriptions and explanations using
   representative sampling;                          evidence.
- explore relationships between the number        Recognize and analyze alternative
   of mosquito larvae and climatic factors;          explanations.
- collaborate with other GLOBE schools            Communicate procedures and explanations.
   (with your country or other countries);
- share observations by submitting data to        Time
   the NBIDS website.                             6 hours; 3 hours for sample collecting and
                                                  another 3 hours for mosquito larval
Science Concepts                                  identification
Life Science
Organisms have basic needs.                       Level
Organisms can only survive in environments        Varies with the protocol
    where their needs are met.
Earth has many different environments that        Materials
    support different combinations of             Practicing identify the mosquito larvae
    organisms.                                    activity
Humans can change natural environments.           Protocol Field Guides
                                                  Equipment is listed on Activity Sheets for
                                                  specific protocol to be done.

                                                  It would be helpful for the class to have seen
                                                  the collection demonstrated. Teacher can use
                                                  the power point to demonstrate the key

Walailak University 2007: Mosquito                                                             12
Practice Identifying Mosquito Larvae
Activity Sheet
The number of mosquito larvae may indicates the risk level of the vector borne disease.
Students can practice to collect, sort and identify mosquito larvae.

 Clear plastic cup
 Pen and Pencil
 Mosquito larvae in plastic bags
 Plastic spoons

What to Do
  1. Review the characteristics of mosquito larvae.
  2. Pour the mosquito larvae from a plastic bag into a clear plastic cup.
  3. Separate the mosquito larvae from the clear plastic cup into three cups based on the
      type of mosquitoes: cup 1: Anopheles, cup 2: Aedes and cup 3: Culex spp.
  4. Count the number of mosquito larvae in each cup.
  5. Fill the number of mosquito larvae in the table below

The number of mosquito larvae
                                                   The number of
Container type         Ae.              Ae.       Culex    Anopheles   Other mosquito
                     aegypti         albopictus    spp.        spp.       species
Flower vases
Flower pot
Pot plants
Water jars
Cement tanks
Ant guards

Walailak University 2007: Mosquito                                                      13
Learning Activity
Key Breeding Site of Mosquito Larvae
Purpose                                         Time
Let students understand the number of           Field trip time plus 2-3 class periods
mosquito larvae in each water container
and the type of mosquito larvae that            Level
prefer in each water container.                 All

Overview                                  Materials and Tools
Student will study and visit the mosquito Equipment is listed on Activity Sheets
study site, and conduct a questionnaire Mosquito Data Sheet
and larval survey.
Student Outcomes                          Find study sites which are appropriate for
Student will learn:                       mosquito larval survey.
  To use questionnaire and larval survey.
  To analyze data.                        Prerequisites
  To interpret results.                   Simple statistics calculations and results’
Science Concepts
Earth and Space Science
Mosquitoes lay their eggs in many types
   of water containers both indoor and
   outdoor containers around the house.
Mosquito larvae breed in stagnant water
   filled containers.

Background                                              (Ae.), those of rodents, monkeys, and
There are currently 412 mosquito species                humans can only be transmitted by
recognized from Thailand, but most of                   Anopheles (An.). The epidemiology of
these are of little practical significance in           malaria, as with all vector borne diseases,
disease transmission, either because they               is shaped by the habitats of the vector. In
are not biologically susceptible to human               Thailand, the primary vectors are An.
pathogens or, more usually, do not have                 dirus, An. minimus and An. maculates.
habits that bring them into sufficient                  Since An. dirus is a forest dwelling
contact with man. The major mosquito-                   mosquito, it inhabits in areas covered
borne diseases of Thailand are Malaria,                 either with natural forests, orchards, or tree
Dengue, Japanese encephalitis virus and                 plantations. It prefers to bite humans,
Filariasis (Rattanarithikul and Panthusiri              avoids contact with DDT and fenitrothion,
1994).                                                  and is long-lived; all of which
                                                        characteristics make it a very efficient
Although the malaria parasites of birds can             vector in spite of generally low population
be transmitted by Culex (Cx.) and Aedes

Walailak University 2007: Mosquito                                                                 14
densities (Rattanarithikul and Panthusiri      2000, Harrington and Edman 2001,
1994).                                         Thavara et al. 2001, Dieng et al. 2002,
                                               Guzm′an and Kourí 2002, Hoeck et al.
An epidemic of DHF occurred in Southern        2003). Because preventative care is an
Thailand (e.g. Samui Island in 1966 and        increasingly important part of the strategy,
1967 (Winter et al. 1968)) where Ae.           social factors that influence its use must be
aegypti and Ae. albopictus were abundant,      more closely investigated (Benjamins and
and widespread (Gould et al. 1968, Russell     Brown 2004).
et al. 1968, 1969, Thavara et al. 2004). Ae.
albopictus is capable of breeding in a wide    Japanese encephalitis (JE) is less common
range of container types and water-holding     than dengue and is most prevalent in rural
habitats. In Thailand, Ae. albopictus has      areas in the northern regions of Thailand.
been found in forested habitats ranging in     Unlike the epidemiology of dengue,
elevation from 450 to 1,800 m as well as       animal reservoirs, especially domestic
in a variety of other habitats in rural and    pigs, play an important part in the
suburban areas (Scanlon and Esah 1965,         transmission of JE. A number of mosquito
Gould et al. 1970, Thavara et al. 1996,        species, mostly Culex spp., have been
2004). Ubiquitous breeding sites, such as      found naturally infected with JE. Cx.
tree holds, coconut shells, fruit peels,       tritaeniorhynchus, which breeds in great
water jars, unused and discarded tires, and    numbers in the pools left in rice paddy
boats holding water have been found to         fields toward the end of the harvest, is
contain Ae. albopictus larvae (Thavara et      generally considered the principal vector,
al. 2004).                                     but there is also evidence that Cx.
                                               fuscocephala, Cx. gelidus, Cx. vishnui and
There are several factors affecting DHF        Cx. pseudovishnui (Burke and Leake
incidence including water storage, climatic    1988).
and vector factors. Container factors
comprise of shape, type, the size of water     Filariasis is caused by the helminths
surface, purpose for which the water used,     Wuchereria bancrofti and Brugia malayi.
type of materials, lids and water              Adult worms live in the human lymphatic
consumption characteristics (Tinker 1964,      system and produce microfilariae that
O’Meara et al. 1992, Kittayapong and           circulate in the blood, where they can be
Strickman 1993a, Focks et al. 1994,            ingested by mosquitoes. The principal
Luemoh et al. 2003). Climatic factors          vectors of B. malayi in southern Thailand
comprise of the amount of monthly              are the Mansonia species. The vectors’
rainfall, vapour pressure, and maximum,        larval habitats are susceptible to control by
minimum, and mean temperature (Hales et        public sanitation measures. The larvae and
al. 2002). Vector factors comprise of the      pupae of Mansonia remain submerged in
strain of the virus, mosquito density,         large ponds until exclusion occurs.
mosquito      behavior    and     mosquito     Transmission of filariasis along the
competence, food level, duration of            northern border takes place in the forest
development, size at emergence, flight         and the principal vector may be a member
range, survival and biting activity (Rigau-    of the Niveus Subgroup of Aedes, all of
Pérez et al. 1998, McBride and Ohmann

Walailak University 2007: Mosquito                                                       15
which develop in natural containers          Collect the data by using mosquito
(Rattanarithikul and Panthusiri 1994).       protocol
                                             Identify the mosquito larvae by using
Teacher Support                              “Learning Activity: Identify mosquito
Advance Preparation
Discuss with students about the
importance of the disease causes by          Further Investigations
mosquito vectors.                               1. Plot the number of monthly DHF
                                                   incidence at your area. Are there
                                                   any indications of climatic factor
What to Do and How to Do It
                                                   (i.e. the amount of monthly
Ask students about their mosquito
                                                   rainfall, mean, maximum and
knowledge. Begin with the questions such
                                                   minimum temperature) that may
                                                   correlate with the number of DHF
    • Are there mosquitoes in your
                                                   incidences in our area?
       house/school that you live?
                                                2. Compare the number of mosquito
    • What are the diseases cause by
                                                   larvae between seasons, such as
       mosquito that you know?
                                                   between rainy and summer
Divide students into two groups. For each
                                                   seasons. Seasons may have some
group have 1 teachers to take care the
                                                   influence on the number of
                                                   mosquito larvae in your area.
Looking at a local map to identify
                                                3. Think about other factors that may
mosquito sites:
                                                   affect to the number of mosquito
Divide a local map into grids.
                                                   larvae such as water pH,
    Group 1: select 10% of grids to study
                                                   transparency, the amount of
    Group 2: select 10% of grids to study
                                                   dissolved oxygen and etc.
    For each, collect 1-2 houses per grid.

Walailak University 2007: Mosquito                                                 16
Table. The number of mosquito larvae at Muang district, Nakhon Si
                                                                                  Culex     Anopheles
  House      Container type     No. container   Ae. aegypti   Ae. albopictus                            Other
                                                                                   spp.       spp.
    1      Flower vases              1              0               0               0          0         0
           Flower pot plates         1              0               0               0          0         0
           Pot plants                1              3               0               0          0         0
           Cement tanks              1              10              0               3          20        12
    2      Pot plants                1              0               0               0          0         0
           Cement tanks              1              20              0               5          5         15
           Ant guards                1              0               3               0          0         0
           Ant guards                2              0               5               0          0         0
           Ant guards                3              0               0               0          0         0
           Ant guards                4              0               0               0          0         4
    3      Flower pot plates         1              0               0               0          0         0
           Water jars                1              5               0               0          0         3
           Tires                     1              0               3               0          12        0
           Tires                     2              0               5               0          5         3
    4      Cement tanks              1              12              0               8          6         8
           Water jars                1              8               0               0          0         10
           Tires                     2              0               6               0          10        0
    5      Tires                     1              0               0               0          3         0
           Tires                     2              0               7               0          8         3
           Flower vases              1              0               0               0          0         1
           Flower vases              2              0               0               0          0         5
           Water jars                1              10              0               0          0         8
           Pot plants                1              2               0               0          0         0

Table: Mean (± S.D.) numbers of mosquito larvae.
    Container type             Ae. aegypti          Ae. albopictus             Culex spp.   Anopheles spp.      mosquito
Flower vases
Flower pot plates
Pot plants
Water jars
Cement tanks
Ant guards

Walailak University 2007: Mosquito                                                                                     17
Calculating a mean and standard deviation of the number of mosquito
Step 1
First, students should calculate a mean number of mosquito larvae in each water container
using Eq. (1).
                                           n                                      Eq. (1)
                                         ∑ xi
                                     x = i =1

Step 2
Calculate a standard deviation of the number of mosquito larvae in each water container
using Eq. (2).
                                       n                                        Eq. (1)
                                      ∑  ( xi − x )

                               SD = i =1
                                          n −1

Table: Mean (± S.D.) numbers of mosquito larvae.

           Container type     Ae. aegypti   Ae. albopictus   Culex spp.                 mosquito
          Flower vases        0.00±0.00       0.00±0.00      0.00±0.00     0.00±0.00   2.00±0.71
          Flower pot plates   0.00±0.00       0.00±0.00      0.00±0.00     0.00±0.00   0.00±0.00
          Pot plants          1.67±1.53       0.00±0.00      0.00±0.00     0.00±0.00   0.00±0.00
          Water jars          7.67±2.52       0.00±0.00      0.00±0.00     0.00±0.00   7.00±3.61
          Cement tanks        14.00±5.29      0.00±0.00      5.33±2.52    10.33±8.39   11.67±3.51
          Tires               0.00±0.00       4.20±2.77      0.00±0.00     7.60±3.65   1.20±1.64
          Ant guards          0.00±0.00       2.00±2.45      0.00±0.00     0.00±0.00   1.00±2.00

Calculating Larval Indices
Step 1
First calculate larval indices using the following formula:

                                number	of	positive	houses
         House	Index	 HI =                                ×100
                                number	of	selected	houses

                                     number	of	positive	containers
         Container	Index	 CI =                                     ×100
                                     number	of	selected	containers

                                 number	of	positive	containers
         Breteau	Index	 BI =                                   ×100
                                  number	of	selected	houses

Walailak University 2007: Mosquito                                                                  18
Table: Larval indices at Mueng district, Nakhon Si Thammarat.
                                                            Culex   Anopheles   Other mosquito
    Larval indices       Ae. aegypti    Ae. albopictus                                           Total
                                                             spp.     spp.         species
House Index (HI)
Container Index (CI)
Breteau Index (BI)

        The number of selected houses = 5
        The number of positive houses = 5

        The number of selected containers = 23
        The number of positive containers = 17

                               number	of	positive	houses
        House	Index	 HI =                                ×100
                               number	of	selected	houses
                                = ×100
                                = 100

                                   number	of	positive	containers
        Container	Index	 CI =                                    ×100
                                   number	of	selected	containers
                                 =     × 100
                                 = 73.91

                                 number	of	positive	containers
        Breteau	Index	 BI =                                    ×100
                                   number	of	selected	houses
                                 =     × 100
                                 = 340

Table: Larval indices at Muang district, Nakhon Si Thammarat.
                                                                Culex    Anopheles
    Larval indices        Ae. aegypti      Ae. albopictus                             mosquito   Total
                                                                 spp.      spp.
House Index (HI)              100               80               60         100          100      100
Container Index (CI)         34.78             26.09            13.04      34.78        47.83    73.91
Breteau Index (BI)            160               120              60         160          220      340

Walailak University 2007: Mosquito                                                                   19
Key Breeding Sites of Mosquito Larvae
Activity Sheet

   Permanent markers
   Plastic bags
   Rubber bands
   Plastic cups
   Plastic pools
   Plate dishes

What to Do
    1. Review the Mosquito Field Guide.
    2. Move to the study site and collect the data by using questionnaire and larval survey.
    3. Bring mosquito larvae to a Laboratory to identify mosquito larvae up to species level.
    4. Calculate a mean and a standard deviation of the number of mosquito larvae in each
    5. Calculate mosquito larval indices and fill them in a table.

Table 1. Mean (± S.D.) numbers of mosquito larvae
  Container type       Ae. aegypti   Ae. albopictus   Culex spp.    Anopheles spp.     Other mosquito species
Flower vases
Flower pot plates
Pot plants
Water jars
Cement tanks
Ant guards

Table 2. Larval indices at study sites
                             Ae.         Ae.          Culex        Anopheles         Other mosquito
    Larval indices                                                                                       Total
                           aegypti    albopictus       spp.          spp.               species
House Index (HI)
Container Index (CI)
Breteau Index (BI)

Walailak University 2007: Mosquito                                                                         20
Learning Activity
Data Analysis with SPSS Software
Purpose                                    Time
Let students understand the data analysis Field trip time plus 2-3 class periods
with some computer software that is
suitable for data processing.              Level
Students will study and analyze the data Materials and Tools
from a field survey.                       Equipment is listed on Activity Sheets
                                           Mosquito Data Sheet
Student Outcomes
Students will learn:                       Preparation
  To use SPSS software.                    Install SPSS software for data analysis.
  To analyze data with SPSS software.
  To interpret results from SPSS software. Prerequisites
                                           Simple statistics calculations and results
Science Concepts                           interpretation.
  The number of mosquito larvae in each
     region may be different.

Background                                       than 120,000 corporations, academic
Data analysis is the act of transforming         institutions, healthcare providers, market
                                                 research companies and government
data with the aim of extracting useful
                                                 agencies—to better focus their operations
information and facilitating conclusions.        and improve their performance. The
Depending on the type of data and the            software helps organizations optimize
question, this might include application of      interactions    with    their    customers,
statistical methods, curve fitting, selecting    regardless of whether they are patrons,
or discarding certain subsets based on           employees, patients, students, or citizens,
specific criteria, or other techniques. In       and ensure that the actions they are taking
                                                 today will positively affect their ability to
contrast to Data mining, data analysis is
                                                 reach tomorrow's goals.
usually more narrowly intended as not
aiming to the discovery of unforeseen
                                                 Teacher Support
patterns hidden in the data, but to the
verification or disproval of an existing         Advance Preparation
model, or to the extraction of parameters        Discuss with students about the
necessary to adapt a theoretical model to        importance of the data analysis with SPSS
(experimental) reality.                          software.

SPSS is the leader in predictive analytics
technologies. For more than 37 years,
SPSS has enabled its customers—more

Walailak University 2007: Mosquito                                                         21
What to Do and How to Do It                  Looking at the Mosquito datasheet and
Ask students about the number of             follow the activity sheet.
mosquito larvae in each region. Begin with
the questions such as:                       Further Investigations
    • Where did the data come from?             1. Represent the number of mosquito
    • How were the data collected?                 larvae in a different type of water
Design the personal computer for students:         containers with a bar chart.
one PC for 2 students.                          2. Think about other softwares for
                                                   data analysis.

Walailak University 2007: Mosquito                                                  22
            s       PSS ftware
Data Analysis with SP Soft
Part I. Introduction
                                     Start the SPSS program by click
                                     “Start > All Programs > SPSS
                                     Inc > SPSS”.

                                     Wh the SPS program starts, it
                                       hen        SS
                                     will show this dialog.

                                     Select “Type in data”.

                                     Clic “OK” button, the SPSS
                                     program will show as the left

                                     In this figure, the SPSS program
                                        t                     p
                                     sho “Data V             t.
                                                   View” sheet

                                     Bef          ut          a
                                        fore we pu our data in the
                                     Dat View” s   sheet, we have to
                                       fine our var
                                     def                      Vairable
                                                  riables in “V
                                     Vie sheet.
                                     Firs we cl   lick at “V  Variable
                                       ew” (show at the bot
                                     Vie                      ttom on
                                     the left).

       k             007: Mosquito
Walailak University 20           o                                 23
                                     We define the variable as follow  wings:
                                     (1) First vaariable: indo
                                     (2) Name: in_ou
                                     (3) Define Variable T  Type
                                                pe           ble:
                                             Typ of variab Numeric      c
                                             Width: 8
                                             Dec cimals: 0
                                     (4) Label: Indoor/Out   tdoor container
                                         (description of a vvariable)
                                     (5) Values the values of variable that
                                                             s          e
                                                             in         e
                                         we inpu e.g. “1” i the Value box,
                                                             ”          el
                                         “indoor container” in the labe box,
                                         and click “OK”.

                                            Com            data    a
                                               mplete the d entry as shown
                                            in the left hand side.

                                            Clic back to “           w”
                                                            “Data View sheet
                                            for data entry.

       k             0           o
Walailak University 2007: Mosquito                                          24
Part II. Data A
      criptive S
I. Desc        Statistics
                                             e        a
                                     First, we use data weight
                                             y        “
                                     cases by select “Data >
                                     Weight CCases…”

                                     Click at “Weight caases by”
                                             ct         ble(s).
                                     and selec the variab

                                     In this c         elect the
                                             case, we se
                                             of        o
                                     number o mosquito larvae

                                     Select            ze
                                                 “Analyz       >
                                             ive Statisstics   >

       k             007: Mosquito
Walailak University 20           o                             25
                                                           Select va
                                                           Click “O

                               ysis      shown as be
The results of statistical analy will be s         elow.

       k             007: Mosquito
Walailak University 20           o                                              26
II. Chi-Square test
                                             Chi-Square test” by
                                     Select “C
                                     click      “Analyzee      >
                                     Nonparam          >     Chi-

                                     For this dialog, se  elect the
                                               hat       ant
                                     variable th you wa to test
                                     in the “Tes Variable List”.

                                               ple,       ct
                                     For examp we selec the test
                                               s         utdoor
                                     variable as Indoor/Ou

                                     Click “Options”

                                     Select “De

                                              ntinue” and “OK”.
                                     Click “Con         d

       k             007: Mosquito
Walailak University 20           o                                    27
The results will be shown as below.

       k             007: Mosquito
Walailak University 20           o    28
      x2      gency tabl
III. 2x conting        le
         For a table of frequenc data cros             d
                                           ss-classified according to two categ            iables,
                                                                                gorical vari
X and Y each of w              wo
                   which has tw levels or subcategor
                                           r                        ample, we w
                                                        ries. For exa                      t
                                                                               would like to test
       ationship bet
the rela            tween conta            on          a            r                      ner
                               ainer positio (indoor and outdoor container) and contain type
        n           c          ).
(earthen and plastic container) The hypo  othesis is:
         H0: There are no assocciation betw                         ns
                                           ween container position and conta   ainer types (no
         H1: There are some as            between con
                               ssociation b            ntainer posit           ontainer types
                                                                    tions and co

                                                                                      ve         s
                                                           Select “Analyze > Descriptiv Statistics
                                                           > Cros

                                                             Selec the variab ble(s)
                                                             After that, click “Statistics”
                                                                 r                        ”

                                                                 ct        are”
                                                             Selec “Chi-squa
                                                                 n,       ontinue”
                                                             Then click “Con

       k             007: Mosquito
Walailak University 20           o                                                              29
                                                    Click “Ceells…”
                                                            Observed” an “Expected”
                                                    Select “O         and
                                                            ck        ue”
                                                    Then clic “Continu and “OK K”

                                                            lts        hown as below.
                                                    The resul will be sh

                  he                 e          P=0.000). So we reject H 0, it mean that there
The results show th Pearson Chi-Square = 4.029 (P          o                      ns
      me          ion      n
was som associati between container p            nd
                                      positions an container types.

       k             007: Mosquito
Walailak University 20           o                                                       30
IV. Da Visualization
       We can plo graphs between the n             m          rvae =from indoor/outd
                                         number of mosquito lar                    door
                  then/plastic containers.
containers and eart                      .
                                                   S           phs       cy
                                                   Select “Grap > Legac > Bar…”    ”

                                                  S          stered” and “Define”
                                                  Select “Clus

                                                  I           on
                                                  In the sectio of Bars Represent, select
                                                  “Other stati            mean)”
                                                              istic (e.g. m

                                                  S                       o
                                                  Select the variables into the box:
                                                  Category Ax xes:
                                                  Container tyype
                                                  D          ster
                                                  Define Clus by: Mos                ae
                                                                          squito larva
                                                  And then click “OK”

       k             007: Mosquito
Walailak University 20           o                                                          31
                                     T                    hown as th left
                                     The results will be sh        he

                                     W          it      h         e
                                     We can edi the graph by double click
                                     a          h.
                                     at the graph

                                               k                      ng
                                           Click at the bar for changin a bar

                                           If yo finish ch           r
                                                          hanging bar colors,
                                           click close “Cha Editor”.

                                                         om changing bar
                                           The results fro
                                               rs        n           t
                                           color are shown on the left side.

       k             0           o
Walailak University 2007: Mosquito                                          32
     mple Linear Regres
V. Sim                        alysis
                      ssion Ana
                                       Prepare data as described
                                       above. In this exammple,
                                       we will examine th he
                                             ation betwee the
                                       associa            en
                                       minimu and max     ximum
                                       tempera            mount
                                              ature, the am
                                       of monnthly rainfall and
                                             mber of DHF
                                       the num

                                              “Analyze >
                                       Select “
                                       Regress          ear”.
                                              sion > Line

                                       The dis splay will be shown
                                       as in th left figuree.

       k             007: Mosquito
Walailak University 20           o                              33
                                                               Ch           ndependent
                                                                 hoose the in
                                                                ariables into Independe
                                                               va           o         ent(s)
                                                               bo and the d           v
                                                                            dependent variables
                                                                nto         ent
                                                               in Depende box.

                                                               Click “OK”.

                   ow       w.
The results will sho as below

                                                                his      shows the variables
                                                               Th figure s
                                                               which are the inputs of the

                                                               Th relations            n
                                                                            ship between the
                                                                ndependent a the dep
                                                               in            and      pendent
                                                                ariables wer shown as in the
                                                               va           re        s
                                                               left figure.

        From the re                       ficients of variables and the variab that are suitable
                   esults, we get the coeff                       d          ble        s
in the m            is
       model. In thi case, the maximum t               e                      del
                                           temperature is suitable for the mod (t = 2.3333,
        2).        ear         on                     96X
P=0.022 The line regressio model is Ydhf = 5.09 max temp -342.070, F3 = 6.152, P =
                                                                              3,80      ,
0.001, R = 0.187.

       k             007: Mosquito
Walailak University 20           o                                                            34
                                          dition, we can plot a sc
                                     In add          c                       am
                                                                 catter diagra that
                                                     onship betw
                                     shows the relatio          ween the max ximum
                                     tempe                       r
                                          erature and the number of DHF

                                         ct                  ialogs >
                                     Selec “Graphs > Legacy Di

                                     Then the display will be sho
                                                    y                      d
                                                                own in the dialog.

                                         ct        ple
                                     Selec the “Simp Scatter” and click on

                                         ct         bles.
                                     Selec the variab
                                         dhf” variable into Y Ax
                                        “d                     xis
                                         maxt” into X Axis

                                      Clic “OK” button.

       k             0           o
Walailak University 2007: Mosquito                                                   35
                                              re        e           ot        t
                                     This figur shows the scatter plo between the
                                             m          re
                                     maximum temperatur and the n             D
                                                                    number of DHF

                                            e           aph     ble      n
                                     We can edit this gra by doub click on the

                                              t           f            aight line.
                                     Click on the dot for fitting a stra

                                     Click at “A Fit Lin at Total”..
                                               Add     ne

                                      Click on “Fit Line” and select “
                                             n                       “Linear”.

                                      And then click “Apply”.

                                      This figu shows the regressi line wit R2
                                              ure        t          ion     th
                                      = 0.143, that mean this model c predict data

       k             0           o
Walailak University 2007: Mosquito                                                   36
Learning Activity
Time Series Analysis with Mathematica
Purpose                                    Time
Let students understand series data using Field trip time plus 2-3 class periods
a time series analysis.
Overview                                   All
Students will study series data and
construct a time series model and forecast Materials and Tools
the data.                                  Equipment is listed on Activity Sheets
                                           Mosquito Data Sheet
Student Outcomes
Student will learn:                        Preparation
  To use Mathematica software.             Install Mathematica software with Time
  To develop a time series model with Series application for a time series
Mathematica software.                      analysis.

Science Concepts                         Prerequisites
 Series data of the number of mosquito Time series analysis calculations and
 larvae in each region may be different. results’ interpretation.

Background                                         , ,…         is called a scalar or
A discrete time series is a set of time-        univariate time series. If at each time t
ordered     data         , ,…, ,…,              several related quantities are observed,
                                                is a real vector and             , ,…
obtained from observations of some
                                                corresponds to a vector or multivariate
phenomenon over time. Throughout this
                                                time series.
manual, we will assume, as is commonly
done, that the observations are made at
                                                The fundamental aim of time series
equally spaced time intervals. This
                                                analysis is to understand the underlying
assumption enables us to use the interval
                                                mechanism that generates the observed
between two successive observations as
                                                data and, in turn, to forecast future values
the unit of time and, without any loss of
                                                of the series. Given the unknowns that
generality, we will denote the time series
                                                affect the observed values in time series, it
by     , ,… ,…         . The subscript t can
                                                is natural to suppose that the generating
now be referred to as time, so         is the
                                                mechanism is probabilistic and to model
observed value of the time series at time t.
                                                time series as stochastic processes. By this
The total number of observations in a time
                                                we mean that the observation is presumed
series (represent as n) is called the length
                                                to be a realized value of some random
of the time series (or the length of the
                                                variable ; the time series { }, a single
data). We will also assume that the
                                                realization of a stochastic process (i.e., a
observations result in real numbers. So if a
                                                sequence of random variables) { }. In the
single quantity is observed at each time t,
                                                following we will use the term time series
the resulting      is a real number and

Walailak University 2007: Mosquito                                                        37
to refer both to the observed data and to                 • Where did the data come from?
the stochastic process; however, X will                   • How were the data collected?
denote a random variable and x a                  Assign the personal computer for students:
particular realization of X.
                                                  one PC for 2 students.
                                                  Looking at the number of DHF incidences
Teacher Support                                   in the activity sheet.

Advance Preparation
Discuss with students about the                   Further Investigations
importance of the time series analysis with            1. Plot the number of mosquito larvae
Mathematica software with Time Series                     in different types of water
application.                                              containers with bar charts.
                                                       2. Think about other computer
                                                          software for data analysis.
What to Do and How to Do It
Ask students about the number of DHF
incidence in each region. Begin with the
questions such as:

Time Series Analysis with Mathematica software
Stationary Time Series Models
       In this section, the commonly used linear time series models (AR, MA, and ARMA
models) are defined and the objects that represent them in this package are introduced.
Functions that check for stationarity and invertibility of a given ARMA model and that
expand a stationary model as an approximate MA model and an invertible model as an
approximate AR model are then defined.

Autoregressive Moving Average Models
        The fundamental assumption of time series modeling is that the value of the series at
time t,    , depends only on its previous values (deterministic part) and on a random
disturbance (stochastic part). Furthermore, if this dependence of  on the previous p values
is assumed to be linear, we can write

                          =∅         +∅       + ⋯+ ∅        +   ,                       (1)

       Where ∅ , ∅ , … , ∅ are real constants.     is the disturbance at time t, and it is
usually modeled as a linear combination of zero-mean, uncorrelated random variables or a
zero-mean white noise process

                           =     +        +      + ⋯+                                   (2)

        (    is a white noise process with mean 0 and variance    if and only if    = 0,
     =     for all t, and        = 0 if s≠t, where E denotes the expectation.)   is often
referred to as the random error or noise at time t. The constants ∅ , ∅ , … , ∅ and
    , ,…,       are called autoregressive (AR) coefficients and moving average (MA)

Walailak University 2007: Mosquito                                                        38
coefficients, respectively, for the obvious reason that (1) resembles a regression model and
(2) a moving average. Combining (1) and (2) we get

                                     −∅      −∅        − ⋯− ∅      =             +                          +       + ⋯+                    (3)

        This defines a zero-mean autoregressive moving average (ARMA) process of orders p
and q, or ARMA(p, q). In general, a constant term can occur on the right-hand side of (3)
signaling a nonzero mean process. However, any stationary ARMA process with a nonzero
mean μ can be transformed into one with mean zero simply by subtracting the mean from the

Start the program
       “Start > Program > Mathematica”

We load the package first.
      Import Data:
      Import ["D:\PhD_Study\DataDHF\DHF4Regions_update.csv"];

        The first thing to do in analyzing time series data is to plot them since visual
inspection of the graph can provide the first clues to the nature of the series: we can "spot"
trends, seasonality, and no stationary effects. Often the data are stored in a file and we need to
read in the data from the file and put them in the appropriate format for plotting using
Mathematica. We provide several examples below.
For example:

Transformation of Data
        In order to fit a time series model to data, we often need to first transform the data to
render them "well-behaved". By this we mean that the transformed data can be modeled by a
zero-mean, stationary ARMA type of process. We can usually decide if a particular time
series is stationary by looking at its time plot. Intuitively, a time series "looks" stationary if
the time plot of the series appears "similar" at different points along the time axis. Any no
constant mean or variability should be removed before modeling.
                                                                          Seasonally Differencing H1-b12L

The number of DHF incidence


                              2000                                                        500


                              500                                                       -500

                                      2003   2004    2005   2006   2007                                         5    10           15   20
                                                    Year                                                                  Month
Fig. 1 The number of DHF incidences in Northern Thailand                        Fig. 2 Seasonally differencing 2nd order with 12 months of
from January 2003-September 2007.                                               DHF incidence in Northern Thailand from January 2003-
                                                                                December 2006.

Estimation of Correlation Function and Model Identification
       As stated in the beginning, given a set of time series data we would like to determine
the underlying mechanism that generated the series. In other words, our goal is to identify a

Walailak University 2007: Mosquito                                                                                                            39
model that can "explain" the observed properties of the series. If we assume that after
appropriate transformations the series is governed by an ARMA type of model, model
identification amounts to selecting the orders of an ARMA model.
        In general, selecting a model (model identification), estimating the parameters of the
selected model (parameter estimation), and checking the validity of the estimated model
(diagnostic checking) are closely related and interdependent steps in modeling a time series.
For example, some order selection criteria use the estimated noise variance obtained in the
step of parameter estimation, and to estimate model parameters we must first know the
model. Other parameter estimation methods combine the order selection and parameter
estimation. Often we may need to first choose a preliminary model, and then estimate the
parameters and do some diagnostic checks to see if the selected model is in fact appropriate.
If not, the model has to be modified and the whole procedure repeated. We may need to
iterate a few times to obtain a satisfactory model. None of the criteria and procedures is
guaranteed to lead to the "correct" model for finite data sets. Experience and judgment form
necessary ingredients in the recipe for time series modeling.
        In this section we concentrate on model identification. Since the correlation function
is the most telling property of a time series, we first look at how to estimate it and then use
the estimated correlation function to deduce the possible models for the series. Other order
selection methods will also be introduced.

        1                                                                   1

       0.8                                                                 0.8

       0.6                                                                 0.6
                                                            Partial ACF

       0.4                                                                 0.4

       0.2                                                                 0.2

        0                                                                   0

      -0.2                                                                -0.2

                   5          10          15           20                        5       10          15          20
                           Lag Number                                                 Lag Number
Fig. 3. ACF of DHF incidence in Northern Thailand between   Fig. 4. PACF of DHF incidence in Northern Thailand
January 2003-December 2006 (--- represented 95% upper       between January 2003-December 2006 (--- represented 95%
and lower confidence intervals).                            upper and lower confidence intervals).
Parameter Estimation
        We first introduce some commonly used methods of estimating the parameters of the
ARMA types of models. Each method has its own advantages and limitations. Apart from the
theoretical properties of the estimators (e.g., consistency, efficiency, etc.), practical issues
like the speed of computation and the size of the data must also be taken into account in
choosing an appropriate method for a given problem. Often, we may want to use one method
in conjunction with others to obtain the best result. These estimation methods, in general,
require that the data be stationary and zero-mean. Failure to satisfy these requirements may
result in nonsensical results or a breakdown of the numerical computation. In the following
discussion we give brief descriptions of each estimation method in the time series package;
for more details the reader is urged to consult a standard time series text.

             For example we get these models from the data:
             Model 1: MAModel[{0.757804},194948]
             Model 2: ARModel[{0.574041},198217]
             Model 3: ARMAModel[{0.142215},{0.608612},193154]
             Model 4: ARModel[{0.698411,-0.224706},196561]

Walailak University 2007: Mosquito                                                                              40
        Model 5: M                  0.0886237},200856]
        Model 6: A        el[{-0.0105 696,0.0931435},{0.753489},1925551]
        Model 7: A                  322},{0.663
                          el[{0.07283                    251},200812
                                               322,0.04282          2]
        Model 8: A                                                 64999},2003
                          el[{-0.0455 539,0.0779581},{0.775963,0.036         373]

Diagnoostic Check king
                   g                     et         t          ss           he         s
       After fitting a model to a given se of data, the goodnes of fit of th model is usually
examined to see if it is ind  deed an ap            m           he          is
                                        ppropriate model. If th model i not satis      sfactory,
      cations are made to th model an the whole process of model se
modific                       he         nd                                            arameter
                                                                            election, pa
       ion, and dia
estimati                      ecking must be repeated until a sat
                   agnostic che          t                                 model is foun
                                                                tisfactory m            nd.

Residua Testing
                   various way of check
         There are v          ys                    odel is satisfactory. Th common used
                                         king if a mo                      he       nly
approac to diagn  nostic check           examine the residuals. There are several alt
                              king is to e          e                               ternative
definitio of the r residuals an here we define the residuals to be the noi calculat from
                              nd                               o           ise      ted
        mated mode
the estim          el.

The res                       d
       siduals are calculated first.
        Since the re           e            red                      at                   es.
                    esiduals are also order in time, we can trea them as a time serie As in
the anal                       s                        ostic test in examining residuals is to plot
        lysis of the time series itself, the first diagno
       s                       s
them as a function of time to see if it app ears to be a stationary random seq quence.
        The correlaation functio is plotted along with the bounds 2/√
                               on           d           h

                                               on           f            om
                      Fig. 5. (b) The correlatio function of residuals fro SARIMA
                      (2,0,1)(0,2,0)12 model (-- represented 95% upper and lower
                                               --          d
                      confidence inntervals).

                   looking at the correlati function of the resi
        Instead of l                      ion       n            iduals                 ,
                                                                              at each k, we can
       ok          rst
also loo at the fir h correla             s         a            he         orrelations are zero
                              ation values together and test if th first h co
       he                                he
(H0). Th portmanteau test is based on th statistic

                                     =      +2 ∑               ,                                (4)

        which has an asympto
        w                    otic      stribution with h-p-q degrees of freedom. If
                                     dis          w            d
    ∝              dequacy of the model i rejected at level α.
            , the ad                    is         a

       k             007: Mosquito
Walailak University 20           o                                                               41
                 e,                  c
      In this case Portmante statistic
                           eau                             8644 and χ
                                                      = 8.98                .   ;
                                                                                       7.5871 (P>0
                                                                                    = 27         0.05). So
we acce H0;

                  There was no different b
              H0: T          o                     siduals and zero
                                         between res
                  There was di
              H1: T                      ween residu and zero
                             ifferent betw         uals        o

                  c          o          s          hat
        The graphic analysis of residuals showed th the resid  duals in the model appeared to
       te        y            ro
fluctuat randomly around zer with no o   obvious trend in variattion as the p           ncidence
                                                                             predicted in
                 (Fig. 5). Th indicate s that the most suitab model f predictin DHF
values increased (           his                    m          ble          for         ng
incidence in North          nd
                 hern Thailan was the S SARIMA(2,   ,0,1)(0,2,0)12 model

        Now that w have exp
        N        we           plored meth
                                        hods to esti
                                                   imate the parameters o an appro
                                                              p          of         opriately
                              e         ain        es
chosen model we turn to one of the ma purpose of time series analy                  asting or
                                                                         ysis, foreca
predicti the futu values of a series. I this secti we disc
       ing       ure         o          In         ion                  forecasting methods
                                                              cuss some fo           m
                 n            es                              e         ar          r
commonly used in time serie analysis . We first present the best linea predictor and its
derivati in the in
       ion                    ple
                  nfinite samp limit. TThen we der            proximate b
                                                   rive the app                     p
                                                                        best linear predictor
      used to spee up the calculation. We show how to write a prog
often u           ed                    .         w          w          gram to update the
       ion       a          w
predicti formula when new data are a               nd
                                        available an also intrroduce the simple exp ponential
smoothing forecast procedure.

       Suppose tha the statio
                   at         onary time s
                                         series model that is fitte to the da
                                                                  ed         ata   , ,…,     is
known and we wou like to predict the f
                              p                     es
                                         future value of the ser ries      ,     ,…,     based
                             me          p          n                        d
on the realization of the tim series up to time n. The time n is called the origin of the
forecast and h the l
                   lead time.

      Here we dissplay the ne 8 values of the series predicted from the e
                            ext      s                       d          estimated SARIMA
      along with t last 48 observed da points.
model a          the        o        ata

                    ig.       mber of DHF in
                   Fi 6 The num                           N              and
                                             ncidences in Northern Thaila from Janua    ary
                   20 -September 2007. ─ repr
                    003        r                                          esented predictted
                                            resented actual data, --- repre

       k             007: Mosquito
Walailak University 20           o                                                                      42
Examples of Student Research
Students at Walailak University investigated the effect of seasons, topographical areas, and
mosquito species on the number of mosquito larvae in different types of water containers in
Nakhon Si Thammarat province, Thailand. Data were collected by using a stratified simple
random sampling technique with a total sample size of 300 households in dry season and re-
sampled again in wet season in 2006.

Forming a hypothesis
A number of mosquito larvae, mostly Culex spp., have been found naturally infected with
Japanese encephalitis (JE) which causes substantial human diseases. Various factors affect
mosquito abundance and their species distribution including climatic factors, %vegetation
cover, breeding sites, season, topography and faith-based communities.

Colleting and Data Analysis
A structured questionnaire and larval survey were conducted in Nakhon Si Thammarat
province in March-November 2006 covering 300 households in three topographical areas (i.e.
mangrove, rice paddy and mountainous areas with 100 households per topographical area):
300 households in dry season. These 300 households were re-sampled again in wet season.
Data were collected by using a stratified simple random sampling technique. Data were
analyzed using t-test and three-way ANOVA tests.

Communicating Results
The results showed that in wet season, there were higher numbers of mosquito larvae in
indoor earthen jars, outdoor cement tanks, outdoor plastic containers, and outdoor metal
containers than in dry season. In mangrove area, there were higher numbers of mosquito
larvae in outdoor earthen jars than in rice paddy area. Culex larvae were found higher in
outdoor earthen jars than Aedes larvae (Table 1 and Figure 1).

Walailak University 2007: Mosquito                                                       43
                                 8                                                                                                     8                                                                                                     8

                                                                                                                                                                                                             The number of mosquito larvae
 The number of mosquito larvae

                                                                                                       The number of mosquito larvae
                                 6                                                                                                     6                                                                                                     6

                                 4                                                                                                     4                                                                                                     4

                                 2                                                                                                     2                                                                                                     2

                                     EJI   CTI   PCI   MCI     EJO     CTO     PCO   MCO   ECO   NC                                        EJI   CTI   PCI   MCI     EJO     CTO     PCO   MCO   ECO   NC                                        EJI   CTI   PCI   MCI     EJO     CTO     PCO   MCO   ECO    NC
                                                             Container types                                                                                       Container types                                                                                       Container types
(a) mangrove area in wet season                                                                       (b) rice paddy area in wet season                                                                     (c) mountainous area in wet season
                                 8                                                                                                     8                                                                                                     8

                                                                                                                                                                                                             The number of mosquito larvae
 The number of mosquito larvae

                                                                                                       The number of mosquito larvae
                                 6                                                                                                     6                                                                                                     6

                                 4                                                                                                     4                                                                                                     4

                                 2                                                                                                     2                                                                                                     2

                                     EJI   CTI   PCI   MCI     EJO     CTO     PCO   MCO   ECO   NC                                        EJI   CTI   PCI   MCI     EJO     CTO     PCO   MCO   ECO   NC                                        EJI   CTI   PCI   MCI      EJO    CTO     PCO   MCO    ECO    NC
                                                             Container types                                                                                       Container types                                                                                       Container types
(d) mangrove area in dry season                                                                       (e) rice paddy area in dry season                                                                     (f) mountainous area in dry season

Figure 1. Mosquito larva abundance (□: Aedes, : Culex) in three topographical areas. Indoor containers: EJI = earthen jars, CTI = cement
tanks, PCI = plastic containers, MCI = metal containers. Outdoor containers: EJO = earthen jars, CTO = cement tanks, PCO = plastic containers,
MCO = metal containers, ECO = earthen containers, NC = natural containers.

Walailak University 2007: Mosquito                                                                                                                                                                                                                                                                                  44
Table 1. The number of mosquito larvae in water containers in wet and dry seasons at Nakhon Si Thammarat in 2006.
                                                                            Statistical test
  Container type
                             S                   T                  M                 SxT            SxM            TxM          SxTxM
                                                              Indoor container
Earthen jar           F1;1788=4.685       F2;1788=1.625      F1;1788= 1.346      F2;1788=0.554  F1;1788=0.504  F2;1788=2.046   F2;1788=1.040
Cement tank           F1;1788=0.499       F2;1788=0.499      F1;1788= 0.499      F2;1788=0.499  F1;1788=0.499  F2;1788=0.499   F2;1788=0.499
Plastic container
Metal container       F1;1788=1.455       F2;1788=1.455      F1;1788= 1.455      F2;1788=1.455  F1;1788=1.455  F2;1788=1.455   F2;1788=1.455
                                                             Outdoor container
Earthen jar           F1;3588=0.276       F2;3588=1.199      F1;3588=0.155       F2;3588=0.590  F1;3588=1.858  F2;3588=0.742   F2;3588=1.175
Cement tank        F1;3588=12.506***      F2;3588=3.251     F1;3588=8.580** F2;3588=3.420* F1;3588=9.389** F2;3588=2.943       F2;3588=2.562
Plastic container  F1;7188=10.934***      F2;7188=0.815      F1;7188=0.024       F2;7188=0.455  F1;7188=0.182  F2;7188=0.189   F2;7188=0.068
Metal container      F1;5388=4.011*       F2;5388=1.868      F1;5388=3.227       F2;5388=2.070  F1;5388=0.087  F2;5388=1.438   F2;5388=0.061
Earthen container     F1;5388=3.964     F2;5388= 5.477** F1;5388=4.952*          F2;5388=1.465  F1;5388=2.906  F2;5388=2.723   F2;5388=2.021
Natural container     F1;5388=0.000       F2;5388=0.525      F1;5388=0.471       F2;5388=0.417  F1;5388=0.915  F2;5388=0.100   F2;5388=0.075
Container types, season (S), topographical area (T), and mosquito larva species (M) factors.
*P<0.05, **P<0.01, ***P<0.001.

Walailak University 2007: Mosquito                                                                                                        45

Shared By: