Skip to main content

Novel automatic and stand-alone integrated pest management tool for remote count and bioacoustic identification of the Olive Fly (Bactrocera oleae) in the field

Final Report Summary - ENTOMATIC (Novel automatic and stand-alone integrated pest management tool for remote count and bioacoustic identification of the Olive Fly (Bactrocera oleae) in the field)

Executive Summary:
Olive cultivation is widespread throughout the Mediterranean and is vital for the rural economy, local heritage and environment. The area under olive groves in the EU is approximately 4.376.000 hectares, 4 % of the agricultural area. Approximately 2.5 million producers (one third of EU farmers) are involved. Olive production is the main source of employment and economic activity in many regions, and the sector is basically dominated by SMEs united in associations. The average holding size is as low as 1 ha in Italy, though olive holdings in Spain are larger (6 ha on average) and the annual net income per hectare can reach €1.8974.

ENTOMATIC addresses a major problem faced by EU Associations of Olive growing SMEs: the Olive fruit fly (Bactrocera oleae). This insect pest causes yearly economical losses estimated to be almost €600/ha. ENTOMATIC has developed a novel stand-alone field monitoring system comprising: a fully autonomous trap with integrated insect bioacoustic recognition embedded in a wireless sensor network and supported by a spatial decision support system.

The ability to quantify and make a precise control of Olive fly populations in a cost- effective way, has been a long-desired goal in the Olive sector. Via ENTOMATIC, olive producers will be able to track pest population and geographical status and receive advice on precision pesticide application. The expected benefits of the system developed at ENTOMATIC are the reduction of damage to olive fruit and oil production and to promote the sustainable use of pesticides.

The ENTOMATIC system offers EU Olive SME-AGs an advanced IPM system for Bactrocera oleae, based on an innovative trap capable of automatically counting each insect trapped, identifying the species based on bioacoustic analysis and send the data wirelessly. The proposed system is able to function autonomously during the whole fruiting season, have an operative life-time of at least 5 years and be adapted to handle other fruit flies.

We have developed an easily accessible system, which enables automatic and cost-effective IPM for Bactrocera oleae for all end users. In addition, the proposed system is a powerful tool in the hands of regional/national authorities to enforce legislative requirements, which need data collection on a regular basis to achieve a sustainable use of pesticides. The ENTOMATIC system will enable the Olive SME-AGs and their members to improve their production, reduce the amount of pesticides used and reduce the labour cost associated with spraying activities and inspection of traps.

Using ENTOMATIC, a reduction of the Olive fruit fly damage and of spraying costs and an increase of productivity is expected to be achieved, reporting benefits to the olive growers.

Project Context and Objectives:
Nowadays, the European Union is leading the world producer and consumer of olive fruits and olive oil. This market represents an important of the economy for different regions and, in the last years, this market is increasing in EU and third countries.
Moreover, there exists an important plague that it still affecting olive trees. The olive fruit fly, Bactrocera oleae, is a serious Olive pest in all the countries around the Mediterranean Sea, and the most destructive pest of olives worldwide. To control this and other pests, EU countries primarily use pesticides to control the populations of insects that attack cultures.
In order to fight against this plague, more than 5 billion euros are spent in the EU. Although, the regulation on pesticides are tighter in the EU, their consumption is still increasing. The expenditures on them varies depending on the country, but the Plan Protection Plans costs estimated is around 200€/ha per year, including the labour to spray and the control of the residues of these pesticides in the fruits, mainly by an over-spraying.
The main reason of this over-spraying performed by olive growers is due to the lack of an efficient monitoring system. In order to fight against the plague, farmers are usually over-spraying in an uncontrolled manner. Nonetheless, it is studied that over-spraying is not completely useful.
Furthermore, the cost of the infestation is estimated in a reduction of the revenues by up to 50% in the case for superior extra virgin olive oil, up to 80% in total oil value when the harvest is not completely affected. The damage caused by tunnelling of larvae in the olive fruit results in about 30% loss of the olive crop in Mediterranean countries.
It is shown in some studies that acquiring more frequent data on insect activity can decrease dramatically these losses by a more accurate control of the infestation. The main objective of ENTOMATIC has been based on this requirement.
The new ENTOMATIC system is able to automatically detect the number of flies entering to the trap, hence we can increase the precision of the counting

The successful deployment of the ENTOMATIC solution is expected to benefit the olive production sector. The results obtained by the consortium offer an innovative solution to the growers’ sector.
The ENTOMATIC solution is offering a new Integrated Pest Management (IPM) tool, not only based on an automatic trap. The major innovations performed along these three years are summarized in three main parts:
• a new bioacustic sensor that is able to recognize and automatically count the olive fruit fly entering at the trap.
• a new communicated trap designed that, thanks to a set of novel communication protocols, is able to distribute the information along a mesh network to aggregate the information in a gateway node.
• an IPM tool based on a Spatial Decision Support System algorithm that with the information collected by the traps estimates the propagation of the plague and offers the recommendations that a grower should follow to diminish its effects on the orchards.
The SME-AGs and SMEs participant in ENTOMATIC consortium have the opportunity of taking advantage of being the first association offering a novel solution to the market. The expected benefits are not only related to economic growth, as there will also contribute to employment, expansion of markets and incursion in new markets.
The ENTOMATIC project has worked to offer the best solution to the market requirements and has accomplished the major objectives established at the beginning of the project with the ENTOMATIC system developed.
Project Results:

In the context of Integrated Pest Management (IPM) insect pest population monitoring is crucial [1-3]. The decision to take action against pests using chemical or biological measures is based on insect population measurements. These measurements define the Economic Injury Level, that is, the landmark point in time after which an economic damage appears. The simplest method to monitor the population of insects is by using insect traps that are commercially available for all common pests. Insect traps are usually plastic, low-cost boxes coming at different configurations and carrying a pheromone or food attractant. The cost of applying population monitoring through a network of traps is mainly due to expenses for manual practices (wages for placement of traps, scouters to report counts, zone-managers to oversee scouters etc.). As reported in [4], the California Department of Food and Agriculture operates a network of roughly 63000 attractant-based traps to monitor Diptera: Tephriditae; likewise, in Israel approximately 2600 traps monitor 20,000 ha of citrus orchards for Diptera: Tephritidae. Therefore, large networks of traps are deployed to monitor their presence and population dynamics in an effort to keep an eye on the situation or the effectiveness of a treatment. Due to the enormous cost of manual monitoring and the compromises that often appear in practice, there is a pending need to automatize the monitoring process of the presence of insects by using electronic means, that is, smart-traps also referenced as e-traps.
The concept of e-traps has sporadically appeared in the past [5-7] with a focus on upgrading typical plastic traps with a device that senses the incoming insects and a communication capability to log these counts. Early efforts, though inspirational, were at the time necessarily fragmentary and limited to small scale paradigms that neither communicated with one another nor could be seamlessly integrated into a universal view of insect fauna monitoring. Recent approaches have advanced to the point of transmitting insect counts by using the GPRS functionality [8-12]. The emergence of the Internet of Things concept (IoT) [13] that allows the networking of physical devices to exchange and report data along with the possibility of providing internet services with global coverage on the planet creates new opportunities for communication and cooperation between e-traps and a central agency.
It is the fortunate coincidence of the simultaneous maturation of diverse technologies such as low power electronics, the IoT and artificial intelligence that allows insect monitoring to be viewed through the prism of cyber-physical systems and possibly reach a global scale of application.
We have valid reasons to believe that there is a market for large networks of automatic insect traps that report on counts and identity of insects. To mention but a few, contractual agreements for manual surveillance of large networks of insect traps are already in force as an indispensable stage of IPM for many countries. In addition, services that now do not exist because of the manpower constraint are bound to emerge: on-time infestation/outbreak prediction, time-stamping of insect captures and their correlation with the efficiency of attractants, assessing the nocturnal activity of insects and many more. Moreover, the first commercially available optoelectronic counters of insects have either made an entrance to the market [14] or are about to [15]. Likewise with image-based e-traps [16-18]. The main advantage of our approach as compared to image-based systems is that it does not require human experts to interpret images [19-20]. Wingbeat classification is performed during the entrance of the insect into the trap, thus making the process easier than analyzing an image filled to excess with insects.
Our electronics are designed to meet multiple types of constraints mainly in terms of efficiency and compactness and, as regards the e-trap, power-consumption. A top priority and a hard, non-negotiable constraint we need to meet is their cost-effectiveness so that they can be integrated into practical and affordable insect traps, thus allowing new services to emerge.
It is our belief that science should be alert to people's needs. Research is, as a rule, supported by public money, so it is only ethical to return to taxpayers their money's worth. The emphasis of this applied paper is on the production of a low-cost device that can prove invaluable in protecting people's agricultural lands against fruit flies worldwide.

I.a. Optical light guides
The benefit of a large receiving aperture introduced in the final version of the McPhail trap compared to a single LED or 1D array of diodes is that fast flying insects spend more time in front of the receiving surface and, therefore, offer more information on the flight process. The FOV in [21] is a volume of 70mmx 59mmx 11mm and has the advantage of having a smooth, compact, surface; there are no gaps in the receiving surface of both sensors that could potentially lead to false frequencies when a fast insect passes through the FOV. Moreover, the construction of the light-guide [25] is a 4-mm slim configuration.
I.b. The electronic McPhail trap.
After completing the sensor, we proceed into transferring a low-power version of this technology into a useful product, that is, an automatic electronic McPhail trap. The McPhail trap is a commonly used trap to monitor fruit flies in different parts of the world [26]. McPhail-type traps, in general, are either fully transparent or semi-transparent because light is a strong attractant for fruit flies. The insects enter the trap following the odor of the chemical attractants placed inside the trap and also because of the light, the round shape of the trap and the yellow color sometimes applied to the lower part of the funnel. In our experiments we used a gel-type (non-liquid) of food attractants to avoid problems with the electronics. As a means for verification data needed to validate the automatic counting module, we have developed a quick process on applying entomological glue active for more than 2 months on a transparent plastic sheet (Folex BG-72) that is subsequently inserted to cover the walls of the trap and a part of the interior funnel. The insects must not have a safe place to land as free flight patterns for a long time may lead to double counts. After thorough experimentation we integrated the optical sensor (see Fig. 1).
The triggering process of the e-trap results into a recording such as the one in Figure 2. In this work we present only the differences encountered in the last version of the e-trap. Once the recording has been auto-triggered, we receive a time-domain input of 1024 samples. The sampling rate is low at 4KHz as fruit flies beat their wings at around 200-250 Hz. The embedded algorithm performs a periodogram of 7x256 overlapping windows using a Hanning window of 256 samples and an overlap of 128 samples. The fundamental frequency of the wingbeat of B. oleae, our targeted pest, is expected at 200 Hz [19, 21]. Because the insect is ectothermic the fundamental frequency can vary around 200 Hz, typically from 160 to 230 Hz [21]. The e-trap has an embedded algorithm that decides in situ if the recorded wingbeat snippet belongs to the targeted insect. Our approach is simple and quite efficient. We calculate the power of the adjacent frequency bins around 200 Hz and its 3 higher harmonics (i.e. 400, 600, 800 Hz). The higher harmonics are at integer multiples of the fundamental frequency. We do not search for the fundamental frequency for several reasons. We know what we are searching for, this being a fundamental frequency around 200 Hz and a harmonic structure at multiples of it. Then, depending on the orientation of the insect as it enters the field of view, the fundamental may or may not be larger than the first harmonic. That is, the fundamental may not be the highest spectral component in the signal and, therefore, one should avoid estimating it. The function only estimates a signal power (Ps) from the regions of 200Hz and its three subsequent harmonics. The DC components are excluded from the calculation. The noise power, Pn, is then estimated as a subtraction of the signal level from the values of the total power Ptotal. That is Pn=Ptotal-Ps. If a recording has an SNR below zero then it is rejected as not originating from B. oleae.
Since SNR=10log10(Ps/Pn) we just need to compare Ps to Pn and if Ps-Pn then we accept that the recording originates from a fruit fly (i.e. SNR>0). We have validated this rule from recordings taken in the lab from known cases of B. oleae. This rule is correct 99.9% when confronted with signals originating from B. oleae. The algorithm has also rejected a corpus of 500 recordings of various noise types and insects’ wing-flap with f0<150 Hz. Note that this rule cannot discern between two different species of fruit flies and it is left to the cultivator to use species specific attractants. However, if the snippet is transmitted to a server instead of being classified in situ by simple spectrum-based rules, then we can use top-tier classifiers running on the server and, as shown in the results, the classification is expected to be accurate even for different species of fruit flies. Finally, safety measures must be taken to insulate the trap against electromagnetic interference stemming from RF devices and the power line. Our insulation was based on covering the analogue parts with copper sheets connected to the ground of the system. Moreover, the cables had double insulation for transferring the signal from the sensor to the recorder.
II.a.Experiments with the McPhail trap in the field
In Fig. 8 we give some examples for the reader to get a better insight into the internal procedures of the McPhail trap while operating in the field. The top picture is typical of a B. oleae recording taken from the SD card of the trap. We were positive that this particular insect was of the B. oleae species because we released a number of them below the trap and this one was visually confirmed to fly into it. In the second figure we see the spectrum of the wingbeat (i.e. the frequencies that constitute the ‘signature’ of the wingbeat in the frequency domain). The harmonic structure of the spectrum is typical of an oscillatory movement. The first peak is the wingbeat frequency corresponding to the so-called fundamental frequency (f0). One can see that it is located at 200 Hz as expected, a typical situation of the spectral pattern originating from a B. oleae. The ‘peaks’ numbered from 2-6 are the so- called harmonics f1-f5 approximately at integer multiples of f0. One can see that the detection algorithm in Section C attributes a high SNR value to this recording, much higher than 0. The zero threshold is the one under which a recording is classified as non-B. oleae (i.e. it is rejected as not being B. oleae).
In Fig. 3, the third in row figure is a recording of an insect (not B. oleae) flying in the trap. One can again see the structure of a wingbeat (i.e. multiple peaks in the frequency domain at integer multiples of a fundamental frequency). Note that in the 4th in row figure the fundamental frequency is around 130 Hz and this is impossible for B. oleae. The detection algorithm attributes SNR<0 to this recording and, therefore, rejects the signal as not originating from B. oleae, although it is a perfectly valid wingbeat signal. Last, in the two figures at the bottom we have the case of an interference. We know that, as there is no wingbeat structure in the signal. The recording cannot be originating out of any insect as there is no oscillation. Instead we see a shock-pulse. Note that the algorithm attributes a small SNR below zero and confidently rejects the signal as not originating from B. oleae.
The evaluation for the in situ experiment is based on counting the number of insects trapped on the sticky surface and comparing them to the counts that the trap uploads to the remote server by using the GPRS functionality. We hereby include some observations after deploying e-traps in the field for several months in Greece and in Spain:
A) The traps sustained several abrupt showers of light rain without a problem. No malfunction was observed, and the traps were not triggered (that is, they did not register false alarms for the presence of insects). The humidity sensor correctly reported the events. Due to this event we modified the software that switches off the trap when the humidity sensor exceeds 99%. We do not believe that this action effects a compromise in the utility of the traps as in such conditions the flight activity of insects is doubtful.
B) During a thunderstorm the traps registered false alarms. We attributed them to thunders that cause large-scale electromagnetic interferences and to very strong winds that possibly effected vibration shocks to the trap.
C) The traps were initially fastened tightly on the trees in order to prevent them from hitting the trunk or branches. We later relaxed this constraint so that the traps swung freely in the air to make their use more convenient for cultivators. An abrupt hit to the device will produce a vibration that is propagated through the trap to the sensor. The slight displacement of the emitter with respect to the receiver due to this shock can trigger the recording process of the trap.
Note that triggering does not necessarily imply that an insect count is registered as we further examine the spectral content of the recording after triggering. However, we tried to avoid false triggering as much as possible (see Section B, on the acceleration sensor).
D) After each triggering we applied a delay of 30 seconds to the next permissible triggering. This constraint was added to avoid the relatively rare event of a double count. A double count is observed when an insect lands abruptly on the sensor and flies off again.
E) A minimum duration constraint was applied to a recorded event to qualify as valid wingbeat recording. In this way we discarded events that we attributed to the thermal expansion of the sensor due to external temperature variations. Wherever transparent material was used as a holder for the sensor this was made by polycarbonate material resistant to thermal expansion. F) We report zero false alarms due to the sun and slowly changing light variations (because of passing clouds) or fast changing light variations (because of the movement of tree branches and leaves).
The following results concern some unpublished work on some late modifications to the McPhail trap. The top one is the version delivered to the partners. We improve the device continuously and this will be continued after the end of the project as well. The bottom is the modified one (change of optics).
In this project we delivered a functional e-trap for fruit flies that is validated in the real-field for the task of detecting B. oleae. The point of this device is to replace human monitoring services. There are many problems with manual practices, the most important being the high cost, followed by feasibility and reliability issues when thousands of traps are involved. Note that monitoring is not a new service that we introduce as a novelty but an established process that is currently part of the Integrated Pest Management regulations.
Because of manual practices, large contracts are granted every year to each interested country. The experimental results from field observation support the following conclusions:
a) The trap does not report false alarms due to sun or other environmental reasons although it is transparent and functioning in the open field. Triggering from non-insect sources occurs at very low rates, the recordings produced by false alarms are successfully rejected by the frequency analysis of its content.
b) The trap can sustain bad weather conditions including rain and strong winds without malfunctioning and without false alarms.
c) The detector of the trap discerns the wingbeat of insects and can lock on a specific wingbeat pattern. A temperature dependent SNR calculation will be included in the future to follow the ectothermic nature of insects that change their wingbeat frequency rate according to temperature variations.
d) There is a close correlation between insects found trapped inside and insects counted automatically. A small divergence between counts of captured insects and reported numbers does not affect the decisions of the qualified entomologists responsible for taking decisions on the initiation of treatments.
e) Based on the SNR algorithm (that is, in the case of in situ decisions), the trap can attract and count flies quite reliably. It relies on attractants specific to the target pest to be able to count the targeted pest among similar pests.
f) The device can also transmit the wingbeat snippet to be classified to a server and, therefore, a general purpose food-bait can be used. Transmitting the waveform proves very convenient in practice and allows the composition of historical data. In the server top-tier machine learning techniques can classify insect wingbeats even among fruit flies.
Actuation and remote control take place to a degree but can be extended in the future. The trap can be currently activated and deactivated automatically according to humidity levels and GPS/acceleration readings. Moreover, activation and deactivation of the trap takes place for temperatures below 16 and above 35 oC where B. oleae is inactive. Sampling of environmental conditions require very low power and actuate re-activation of the e-trap when back in the proper range of environmental conditions. The battery levels are also transmitted to the server and are monitored, and the actuation process can be extended in the future by allowing automatic, remote update of the firmware.

We expect e-traps to change the way IPM is applied as the dynamics of insect population and the environmental parameters associated with them are delivered almost in real time. The surveillance of insects through e-traps that record their wingbeat is now ubiquitous thanks to the proliferation of wireless sensors and the GPRS as well as the IoT functionality.
We envision that low cost plastic traps upgraded with low-budget optical sensors and wireless communication will form a network of uniquely addressable objects transmitting insect counts, wingbeat snippets and environmental parameters to a cloud server over vast expanses of land. This information will be visualized in GIS (Geographical Information System) compatible format so that monitoring will be a continuous process that will keep an eye on infestations, perform post-treatment analysis and establish a link with predictive analytics [27].

Design & Development of the ENTOMATIC Trap

The Design & Development of the ENTOMATIC Trap posed three main objectives to be achieved during the project:
1. Preparation of a conceptual design of a prototype trap and the selection of the most appropriate one.
2. Design and construction of the prototype trap including outer shell, fly passage module and suita-ble enclosures for the bioacoustic ID system and the embedded electronics.
3. Manufacturing and testing of prototype trap including its communications and power modules in laboratory and field.
The basis for die investigations in this project was the McPhail-trap, Figure 5. This trap is the most widespread in southern Europe and with this kind of trap, most studies on the spread of the olive fly and the use of pesticides have been performed in the past. The McPhail-trap is almost a standard and based on the studies with it the guidelines for farmers were established.
The focus of the activities in this work package was the development of concepts for the ENTOMATIC trap for the investigations and tests during the project and a final concept for the utilization at the end of the project. The concepts consider the requirements acquired in work package 1 “System spec-ification” for the following main components
-the outer shell and the insect passage module
-the enclosures for the bioacoustic-ID-system
-the enclosures for the embedded electronics (data acquisition and data processing modules as well as the communication modules)
-the power supply (battery pack, if necessary a solar panel).

II. Further development of McPhail trap
The investigations and the further development of the trap based on preliminary investigations of ENTOMATIC partner TEIC. With the feedback from the previous tests, several improvements were in-troduced to both the sensors and electronics. This led to the development of a fully customized EN-TOMATIC trap, including a custom compartment on top to house the also fully customized electron-ics. An illustration of this appears in Figure 6.
Based on the further development of the sensor system, described in chapter of WP2, and the further development of the ENTOMATIC Wireless Sensor Network (WSN), described in chapter of WP4, two new housings / cases were developed.
The first electronic case, installed on top of the trap, figure 5, contains all the control and processing electronics of the opto-electronic sensor system (OESS), the Zolertia Remote gateway and the batter-ies. This case was manufactured of white Plexiglas® XT to protect the electronic components against direct sunlight.
The second housing, mounted lateral at the first electronic case, contains the environmental sensors for the WSN. On the small BCB are temperature, humidity and luminance sensors. The sensor housing has been manufactured by using 3D-printing technology with Duraform ProX material. It is designed as a separate housing in order to be able to detect the measured data independently of the trap - to avoid the influence of the higher temperatures inside the trap - and allow a better acquisition of these environmental magnitudes.
The design of the housings involved some additional challenges and constraints. The placement of OESS and WSN inside the trap is not possible, because
-A change in the internal life of the trap would not have permitted a comparability of the results of the measurements, the number of caught olive flies, between a standard McPhail trap and the electronic McPhail trap.
-The transparent trap heats up in the sunlight; the high heat can damage the electronics and the batteries.
It was not possible to find a suitable commercially available housing. Therefore, the housing of the prototype trap for the field tests during the project period was designed using a cylindrical tube, an upper and a bottom plate as well as a laterally attached sensor housing. Some more detailed information about the housings:
-The housing parts are white and absorb visible light up to 56%, UV completely.
-The upper and the bottom plate of the electronic case were produced using laser-cutting technology including the manufacturing of groove and drill holes.
-Electrically non-conductive spacers were designed for receiving the OESS-PCB to prevent error current.
The grooves of the spacers and the recesses in the circuit board are designed in such a way that they allow an easily assembly.
-The subassemblies battery holder, Zolertia remote und antenna are mounted at the cover plate of the electronic case.
-The antenna is fixed by using a moisture chamber feedthrough.
-The sensor housing is mounted on the side of the electronic case, between both parts are silicone O-rings for sealing.
-The sensor housing is laterally partially open, it is a weather-protected structure for the interior life, so that air can flow through it; light can enter, but no rain.
-The sensor case is closed by 4 knurled screws. The screws can be opened by hand without tools, so that a battery change can be carried out without tools. For sealing purposes, a silicone seal is lo-cated in a groove.
-The opto-electronic sensors hang at two tubes inside the trap. The cables are guided through the tubes to the OSS-PCB.
-The electronic case is made of Plexiglas® XT and all the sensor-housing parts and the battery hold-er were made by additive manufacturing from Duraform ProX.
-The suspension for the trap is a hexagon head screw with a drilling. Thus, the trap can be suspend-ed on a tree as a conventional trap using a wire.
In the project mechanical parts for 25 traps were manufactured, assembled and put into operation, see Figure 8. In addition, 25 OSS electronic boards (OESS-PCBs), were produced, Figure 9.

III. Design Variations
different concepts for the new MacPhail design were specified, optimized and at least a preferred ver-sion were developed.
The design variations are based on the following basic assumptions:
-The shape and color of the trap should be substantially the same as the traditional MacPhail. This concerns the upper and lower part of the housing
-In the vicinity of the entry hole / the funnel, the integration of the sensor system for detecting the Bactrocera oleae is arranged. This has proven itself during laboratory and field tests.
-The integration of the battery should, if possible, be done in a location with convenient accessibil-ity for changing.
-In the uppermost position of the assembly should be provided a possibility to hang the trap.

III.a. Increase of the lower part of the trap to increase the volume of the attractant
If a larger amount of attractant is to be taken up by the trap, e.g. in order to extend the maintenance intervals, it is possible to produce an enlarged the lower part of the trap. For this purpose, the pro-duction of new injection molds is necessary. With a doubling of the attractant volume, which can be absorbed by the trap, approximately a doubling of the height is necessary, Figure 10.
The appearance of the trap can be preserved by using a clear poly¬carbonate.
If it is not necessary to increase the volume of the attractant, it is possible to reuse the existing trap for cost reasons and for reasons of recognizing the traps for the Bactrocera oleae. This is e.g. the case, by using solid attractants.
In any case, the substitution of the upper part requires the use of the exact geometry of the coupling point between the upper and lower part, so that compatibility is guaranteed.
At the lower part, no further components are mounted to ensure an unhindered change of the attractant. When cleaning the vessel can proceed in the usual way.

III.b. Modification of upper part of the trap
The modification of the trap shell is based on the reuse of the existing MacPhail shell. Only minor changes are made, similar to the machining of the ENTOMATIC traps used for the 2017 field test. They limit themselves to the introduction of holes and the milling of the suspension.
The advantage of this solution is that it can be used on the existing tops. The high cost of producing a new injection mold for the relatively large part would be eliminated. The electronics are mounted above the trap in a separate electronics housing.

III.c. Mounting the sensors to detect the Bactrocera oleae
The sensor system is integrated into the upper part. Through the sensor system the detection of Bactrocera oleae takes place at the entrance, the funnel. The installation of the sensor system can be realized like in traps for the field test of 2017. The assembly can be carried out hanging in the upper part on 2 or, to increase the stability, on 3 pipes. For this purpose, the production and installation of sup-port tubes is necessary, Figure 11.
Another solution that is simple and cost-effective to install is the new form of sensor carrier shown in Figure 12, which is clicked into 3 holes in the outer wall of the upper part.
In this solution, both the production of the connecting pipes and the complex attachment of the pipes are no longer necessary. The connecting cables for the sensors is combined in one thicker cable and this can be attached to the inner wall of the upper part or led directly upwards into the electron-ics box.

III.d. Electronic case
The electronics box of the traps from the 2017 field test is a completely closed box, on the side of which the sensor housing for the environmental sensors is mounted. The parts of the box are made of Plexiglas XT; they were made by laser cutting. The sidewall and the floor are glued together. The cover contains a silicone seal, which prevents the ingress of rainwater into the housing interior. The lid is closed with 4 knurled screws. The connection of the side-mounted sensor housing was sealed with silicone cord. In the electronics box, the PCB and a battery pack are mounted. The antenna could be omitted.
For the design of the shell several variants were developed. All variants are based on the perspective of the production of injection-molded parts, analogous to the upper and lower part of the MacPhail.

III.e. Assignment of the outdoor sensors to the electronics case
The electronics include two environmental sensors that measure the brightness for day / night activation and humidity in the environment. These sensors should be in direct contact with the environment, but protected from rain. Complete ventilation, as seen in Figure 9, is relatively easy to implement. However, it has the disadvantage that the entire electronics is exposed to the temporarily humid environment. In addition, the protection against dirt is insufficient, which in total makes this variant does not make sense.

III.f. Assignment of the outdoor sensors to the electronics case
The electronics include two environmental sensors that measure the brightness for day / night activation and humidity in the environment. These sensors should be in direct contact with the environment, but protected from rain. Complete ventilation, as seen in Figure 9, is relatively easy to implement. However, it has the disadvantage that the entire electronics is exposed to the temporarily humid environment. In addition, the protection against dirt is insufficient, which in total makes this variant does not make sense.
The variants according to Figure 13 and Figure 14 realize a sealed electronics box and a vented sensor area. The main difference between the two variants is that the tool for producing the box according to Figure 14 is easier to realize. On the other hand, the manufacture of the weatherproof housing for the sensors requires a complicated injection-molding tool. The production of the separate sensor housing according to Figure 14 is not possible by vacuum casting; it can be produced only by STL pro-cess or by injection molding.
If the overhanging edge of the electronic box is not sufficient for the sealing, a sealing of the electronics box can be realized by a silicone gasket produced in an additive manufacturing process.

III.g. Newly designed upper part with integrated electronic case
Another variant is the production of a raised top that can accommodate all the electronics. The upper part corresponds to the current one, but is executed higher.
This variant presents several challenges, but should be considered for the sake of completeness. The accessibility of the batteries is only possible by removing the sensitive sensors, which is not recommended in field use. This would be necessary, for example, for every battery change. The attachment of the separate weather protection housing with lamellar protection for environmental sensors is technically only technically solvable. The cabling of the sensors must be realized via a plug connection so that the cable can be disconnected from the electronics when removing it.

III.h. Attached electronic case
When designing the case, different solution ideas were considered. This case basically consists of a component carrier and a hood. On the component carrier, which is mounted on the McPhail upper part, the electronics and the battery pack are mounted. The hood has the function of protection against environmental influences and can be opened for maintenance purposes or the replacement of the battery.
A variant based on the McPhail from the 2017 test and is a mounting of the lid with screws. The use of tamper-proof screws, the heads are sunk and can only be solved with a special key. Disadvantage of this variant is the space required for the screws and possibly screw domes that restrict the space requirements of components on the PCB or the arrangement of the battery.
As a closure mechanism, a retractable closure according to the pattern of the connection of the lower and upper part is also conceivable. The integrated overhanging edge of the lid simultaneously protects the interior of the box against dirt and moisture.
This solution has the advantage that there is no space in the interior for fasteners such as dome for screws o.a. must be reserved. At the top there is a fastening eye for hanging the trap

See a comparative of the novelties introduced at the table attached at the document.

III.i. Research on production methods for prototypes and for small, medium and large series
The housing parts are the main components to plastic parts. In the case of the field test 2017, a very cost-effective production variant was chosen in which the parts were manufactured and mounted by hand in the IMMS.
-The Plexiglas parts of the housing were made by laser cutting.
-The parts of the sensor housing were manufactured using the rapid manufacturing process.
-For a total of 25 traps, this procedure was practicable.
For larger quantities in small series or large series, there are other, more cost-effective production options. The newly developed plastic parts are designed to be manufactured in durable materials and by injection molding techniques that also meet the environmental conditions and other as well as user requirements.
For small and medium series vacuum casting is suitable. For this purpose, a judgment is made e.g. produced in the STL process and can then be molded several times. In large series plastic injection parts are produced, which are only then produced inexpensively due to the expensive tools.

Design & Development of the ENTOMATIC Wireless Sensor Network
Work Package 4 (WP4) “Design & Development of WSN” posed three main objectives to be achieved during the project:
-Evaluation of the network and communication requirements of the ENTOMATIC system.
-Design, evaluation and development of the ENTOMATIC Wireless Sensor Network (conceptual design, network architecture and communication protocols at the traps and gateway).
-Testing the ENTOMATIC WSN in laboratory and field.

The network designed and developed for the ENTOMATIC project within WP4 can be split into two differentiated parts (Figure 20):
1.An acquisition network, based on wireless sensor network (WSN) technologies and used for the communication between the traps and the gateway. While the gateway (GW) is placed in a central position, a series of rings formed by scattered communication devices are deployed over the monitored area.
2.A transport network, using cellular technologies (more specifically, GPRS) and responsible of transmitting the gathered information from the gateway to a data receiver server. In this case, the single requirement for the gateway is being provided with cellular coverage.

The ENTOMATIC network is clearly aligned with typical low-power wide area networks (LPWAN) applications and circumscribes its suitability to those scenarios with special concern for energy efficiency, where device batteries are so limited that the establishment over time of a direct connection to the base station, or gateway, would greatly affect their lifetime.
Assuming that the employed hardware provides good signal penetration, a single GW could serve up to thousand low-traffic devices within its given coverage range. Applications executed by stations (STAs), in turn, follow a continuous data delivery model for their sensed information, periodically delivering small amounts of non-delay sensitive data.
As sensor nodes are scattered over large areas, sometimes with problematic access, self-maintenance of the system shall be a priority, capable of giving response to the following challenges:
1.Node initiated network connection: once installed for the first time or relocated, any node shall initiate its association process through a simple action (for instance, pressing a button).
2.Self-configuration and management: with the aim of building a robust network, it shall adapt itself to environmental and/or topology changes without human intervention.
3.Battery lifetime maximization: LPWANs replace old monitoring systems consisting of assigning human resources to study in situ the behavior of one or more physical parameters. Therefore, maximizing battery lifetime in such systems is vital in order to justify their usage ahead of other methods.
4.Firmware distribution: any change in the network configuration or in the application purpose shall be remotely and easily distributed by the GW.
In addition, although the operating system of embedded sensor nodes is typically less complex than general-purpose operating systems, it shall cope with the high variety of resources to manage in this kind of devices (processors, memories, clocks, network interfaces, and so forth) and the demand of support for concurrent execution of processes (time synchronization, data acquisition, task scheduling, channel access, routing parameters, and so forth).

II.a.Acquisition network
The communication protocol stack for the acquisition network has been designed and developed as a new technology flexible enough to adopt uplink multi-hop communications when proving energetically more efficient than single-hop. A full set of advanced techniques belonging to different communication layers has been designed for this purpose, while ensuring data transmission reliability:
-Physical layer: The developed protocol stack is intended to be used over any wireless PHY layer fulfilling a minimum set of functions.
-Link layer: beaconing system, wakeup patterns, data transmission, aggregation and segmentation, power regulation mechanism.
-Network layer: addressing system, association, routing.
-Transport layer: end-to-end acknowledgement (e2e ACK), poisoning mechanism (packet loss detection), transmission windows, distributed caching.

II.b.Transport network
The network topology chosen for the transport network between the GW and the data receiver server is a point-to-point topology over a GPRS link. The GPRS technology was chosen due to its simplicity and its availability in rural environments worldwide.
The client–server model is used between ENTOMATIC gateways (clients) and the data receiver server (server), where a distributed application structure partitions tasks or workloads. Due to its simplicity for sending and receiving information, the GET method of the HTTP/1.1 specification is used in the transport network.
Lastly, the communication protocol used in the transport network does not disturb the normal operation of the GW as part of the WSN, since the gateway waits until the end of the last transmission window of each beacon period to initiate any kind of communication with the server.

The ENTOMATIC network conceives end devices as elements controlled by the GW by means of beacons. This centralized approach allows STAs to remain asleep the majority of the time, so that their single concern is to be awake enough in advance to listen to the next beacon. Network synchronization is thus achieved and allows the GW to ask for specific data and/or distribute configuration changes easily.
The GW is considered to be appropriately placed close to a power source or an energy harvesting solution. Therefore, it may always stay in an active state and is provided with the ability to directly communicate (i.e. via single-hop communications) with any node of the network through unicast and/or broadcast messages as well as to redirect gathered data to the data receiver server via GPRS messages.
Conversely, STAs can take advantage of their neighbors to create multi-hop paths over which data is transmitted to the GW by means of lower transmission power levels. During the association process, and depending on their position within these paths, STAs are ideally organized into rings, so that the number of hops to reach the GW determines the ring number (i.e. STAs belonging to ring #2 require two hops to reach the GW).
Each uplink data transmission phase begins with a beacon signal from the GW and is split into as many TDMA slots as network rings, so that STAs are only active during their own slot (for transmitting data) and the previous one belonging to their children (for receiving data).
The first slot is then allocated to the highest ring and the rest are scheduled consecutively. Data received by STAs is aggregated to that generated by themselves, and finally sent to the corresponding parent at the minimum power level that ensures reliable communications. This process is repeated as many times as rings that the network has.
The correct reception of data transmissions at the GW is acknowledged at the end of every transmission window with a broadcast message, so that STAs are not only aware of their own end-to-end reliability, but also of those STAs in the same path to the GW. These acknowledgment beacons, together with the information obtained from their adjacent nodes, allow STAs to decide whether they should remain awake to perform retransmissions of lost network packets in successive transmission windows.
Lastly, network association (also started by a beacon) remains stable until a change in the topology is detected or the mechanism is reset by the GW. Nevertheless, the agreed transmission power between adjacent nodes in the association phase is constantly monitored and adjusted in a decentralized way in order to reduce the energy consumption.

The core element of every device in the acquisition network is the Zolertia RE-Mote, a device with processing and communication abilities. In addition, an integrated board has been designed and manufactured to connect other elements to the Zolertia RE-Mote.
-Gateway unit: It is made up of a Zolertia RE-Mote, an integrated board, a GPRS SIM 900 module and, optionally, a solar panel.(Figure 21)
-Trap unit: It is made up of a Zolertia RE-Mote, an integrated board, the opto-eletronic sensor, a temperature/humidity sensor and a luminance sensor.(Figure 22)

IV.b. Software
The communication protocol stack from both the acquisition and the transport network was fully programmed as a new hardware independent module for Contiki 3.0 OS in C programming language, adding all the described functionalities to the already available upper communication layers (link and network), and including a novel transport layer. Specific interactions with PHY layer of the Zolertia RE-Mote hardware were programmed separately.

V. Testing
To prove the performance of the ENTOMATIC network and its suitability in typical olive fields, different test stages have been carried out, ranging from accurate evaluation of specific network mechanisms, testbeds emulating real conditions and tests performed in real olive orchards.

-Testbed #1, executed on the 2nd floor, right wing of the Tanger building at Universitat Pompeu Fabra (UPF) facilities, was used to analyze the performance of the main network functionalities in a controlled area with easy access to devices (figure 23).
-Testbed #2, performed in the localities of Prats i Sampsor, and Savanastre (Catalan Pyrenees), established the maximum range coverage of the system in line-of-sight condition.
V.b.Real tests on olive fields
1.Technology integration testing was performed on 24th and 25th April in a selected oliveyard from the Falset-Marçà cooperative, located in the town of Falset (Tarragona) (figure 24).
-7 devices running the ENTOMATIC system were deployed in the area under study: one gateway located in a raised position to cover the maximum coverage range, 3 trap units to perform end-to-end communication between them and the web-server, and 3 additional “support” units collecting environmental data.
-In addition, another test was performed to determine the maximum range coverage of the Zolertia RE-Mote nodes in an area where trees may affect the line-of-sight propagation.
2.Different real tests performed by Nutesca and UPF in July, August and September 2017 on selected oliveyards located in the surroundings of the town of Baeza (Jaén) (figure 25).
-In this case, 3 devices running the ENTOMATIC system were deployed in the area under study: a full gateway unit (Zolertia RE-Mote + integrated board + GPRS SIM 900 module + solar panel) and two trap units (McPhail trap + Zolertia RE-Mote + integrated board + opto-electronic sensor + temperature/humidity sensor + luminance sensor).
-The full experiment lasted more than 11 days, during which traps gathered and transmitted fly and environmental information every hour, feeding the web-platform.

Design & Development of the ENTOMATIC Spatial Support Decision System
The Entomatic pest management tool relies on olive fly counts made by a tailored trap that includes a bioacoustic sensor. Much effort was put into energy management and making the sensor and communication module as energy efficient as possible. The low power consumption reduces maintenance requirements, especially because batteries should be replaced a limited number of times. During some months, no olive flies are expected and maintenance can be reduced even further by not powering on the sensors. The SDSS includes an indication when sensors should be switched on again to capture the onset of olive fly counts and to perform the first olive fly control measures of the year. The period in which olive flies are active is strongly correlated with olive growth for which altitude and temperature are important environmental predictor variables. Based on data captured in Italy and Spain, a model was established that provides insight in when to expect olive flies to become active.

The SDSS can be considered in three parts: (i) The B. oleae spring onset model (ii) The control decision tree and (iii) Additional analysis on the ENTOMATIC Web app. The B. oleae spring onset model serves as an early warning system to notify farmers when to turn on the traps in early spring. It runs from the 1st of January and constantly record the change in daily temperature in each of the orchards. When a critical, threshold temperature (TT) is reached the model calculates the growing degree days (GDD) for the orchard and gives an estimate of when the first peak of B. oleae is expected. The model uses the link between spring flowering of the olive trees and the onset of the first generation of B. oleae [34] to provide a nearly warning to the farmers. Air temperature was used as the main factor for estimating the change in olive phenology because this link has been widely demonstrated [32,33,31].

Once the traps are on and the B. oleae population increases, a control decision tree (CDT) model is run in parallel with the wireless sensor network (WSN) of traps that provides data on air temperature (T), relative humidity (RH) and olive fruit fly count. Each branch of the decision tree represents a particular pest management strategy whilst the nodes of each branch indicate the decision points. The leaf node is then the final decision to be taken by the orchard manager. Up till now, olive fruit fly control was done on an expert basis only. Now, the CDT allows to optimise control strategies in an ecological sustainable and cost-efficient manner by taking into account objective criteria and harmonized recommendations for spraying decisions against the olive fruit fly. However, there will always be a certain level of uncertainty as it is rarely possible to fully predict the effects of major interventions.

I.Spring Onset Model
The Spring Onset model was developed to assess the early spring risk of the emergence of B. oleae. The model is based on the relationship between the spring peak in the population density of B. oleae and the flowering of the olive tree [34]. It is integrated in the web ENTOMATIC app and serves as an early warning system to help the growers make an administrative decision on when to start spraying when there is still no count data from the traps. Once all the traps are turned on, the WSN will be sending data on regular basis to the centralized database. Each node of the WSN will send: device ID, GPS coordinates, temperature, humidity, wind and counts of the olive fruit fly. The model relies on three levels for obtaining the environmental variables: (i) WSN data, (ii) Weather API data, (iii) Climate Normals. The most accurate information would be obtained from the WSN system directly from the orchard. If the traps are not yet turned on, environmental data would be pulled automatically from a weather API (Application Programming Interface) from a reliable meteorological data provider. In case this is not available for the coordinates of the olive orchard, Climate Normals will be used as input variables for that region, and will be taken from a prepared database. This constitutes three-decade averages of observed climatological variables.

II.Control decision Tree
Once there is information coming from the field a spatial decision support calculates when and where to conduct control using relevant criteria. The relevant criteria for the control decision tree model (CDT) can be divided into three types: continuous, discrete and Boolean. Continuous criteria included (i) the number of flies in the nearest trap; (ii) the population dynamics (especially in between spraying activities); (iii) the history of trapping; and (iv) the condition of the population in the nearby traps. Discrete criteria included the slope and or soil type of the plot. Boolean constraints included (v) the presence of fresh specimens in the nearest trap; (vi) status of the plot—whether it was harvested or not; and (vii) the overall condition of the Olive fly population in predefined regions.

The outline of the model is given on Figure 26. The main part of the CDT is constructing a certainty factor (CF) using SCFA (Stanford Certainty Factor Analysis) from a set of criteria which are described in Table 3.

The final output of the CDT model will classify all plots according to different types of recommendations, which will be: (i) ‘Spraying is not required’ for all plots that gained an overall CF lower than 31%; (ii) ‘Spraying is not recommended’ for all plots that gained an overall CF between 31% and 50%; (iii) ‘Spraying is recommended’ for all plots that gained an overall CF between 51% and 70%, (iv) ‘Spraying is required’ for all plots that gained an overall CF higher than 70%

The CF‘s in principal express a ‘belief’ in an event, which can be a fact or hypothesis, and are relative, and thereby not absolute, measures. In general, the CF(x) is a measure of how confident we are in x. This certainty ranges from -1 to +1 with CF = - 1 = very uncertain; CF = 0 = neutral and CF = +1 = very certain. Several rules can be added to the calculation; these can be conjunctive or disjunctive. These rules have also a certainty factor. The certainty of the rule will be combined with the initial event certainty to obtain a final certainty factor. The order of adding CFs does not change the overall CF, it’s commutative, and they always add up to a value between +1 and -1. Moreover, if one of the CFs equals -1 or +1, providing unambiguous evidence for or against a certain conclusion, the overall CF would be -1 or +1, no matter what the values of the remaining factors are. This means that both soft and robust rules can be integrated by using the same algebra. Adding a CF of 0 will not change the overall CF. Finally, the combined CF value is a monotonically increasing or decreasing function such as one would expect when combining criteria. The Stanford Certainty Theory or Stanford Certainty Factor Analysis uses human expert estimates not based on probabilities but which are heuristically derived from experience in reasoning in a particular domain. This is very relevant in situations for which no previous results are available to derive those probabilities. The spraying decisions for the Olive fly given environmental and biological conditions falls within this category.

The CF of each piece of evidence ranges between (−1) and (+1). As the CF approaches (+1) or (−1), the strength of the evidence for or against the conclusion, respectively, increases. A CF around 0 indicates that there is little evidence for or against the conclusion. Combination of the CFs is calculated in a modular mode. An initial CF is taken to be one of the CFs ascribed to the criteria and it is updated x times (depending on the number of criteria included) by adding the CF of each of the other criteria in succession.
The final CDT of the Olive fly SDSS will be made of the Boolean constraints which govern the overall CF calculation based on the SCFA. These constraints will either avoid spraying actions or mandate a spraying action. One of the Boolean constraints is imposed to avoid spraying actions: if a plot is young, harvested, dry or uprooted. Two others are imposed to mandate spraying action: (i) the presence of fresh adults and (ii) more than 3 olive fruit flies (although the latter is different between table olives and olives for oil). Another Boolean constraint to consider is the identification of the plots for which no reliable recommendations can be given: if a plot is too far from a trap and if information from the nearest trap was not correct/unreliable.

III. Additional Analysis
Apart from automatic alerts being raised by the SDSS, the system also allows to analyse the raw data and look for additional indicators for the necessity of spraying or other control activities. To this end, the web-app provides both maps and box plots that visualise B. oleae counts. Which data is included for visualisation is determined by a spatiotemporal query (see figure 27):

-To select the traps from which data is extracted, the user can select an orchard or a specific trap. The list of orchards and traps to choose from is restricted to orchards and traps owned by the user or by users that are member of the umbrella organisation (when analysis is done by a user with administrator rights for that umbrella organisation) for privacy reasons. In the box plots, the user can compare this data with data from others by specifying a distance to the own traps. Comparison data will be taken from the traps that are within this distance. Also a fixed number of nearest traps can be used to select the comparison data.
-The selected data, including the comparison data, can be restricted in time by setting a start and end date.

Box plots (see figure 28) show the distribution of measurements in the own traps giving the quartiles: minimum, Q1, median, Q2 and maximum. The average counts are given as a line. For analysis of data coming from long periods of time, data is aggregated. When moving the mouse pointer over the boxplot, the range of dates for which data is aggregated is shown. In the same box, also the exact values can be found. This way, users can compare B. oleae counts over time and between traps of their own and neighbouring traps. While analysis over time reveals a trend of increasing counts one may decide to already start control actions even when decision tree thresholds have not been exceeded yet. The same goes for counts that are higher in traps from neighbouring orchards as compared to the own traps and that may indicate the advance of B. oleae.

This does not necessarily mean that spraying activities need to be undertaken but paying more attention in specific areas could be appropriate. This includes checking the web-app more regularly to follow up on the B. oleae counts from traps in the own and neighbouring orchards.
Alternatively, analyses may be performed in third party software after extracting the sensor readings as a .csv file.

IV. Conclusions
We devised a pest management tool for the olive fly that is based on a tailored trap with a bioacoustic sensor that registers fly counts. The trap’s sensor and communication module was made as energy efficient as possible, thereby significantly reducing maintenance requirements. The SDSS consists of three parts: (i) a B. oleae spring onset model, (ii) a control decision tree and (iii) the ENTOMATIC Web app to perform additional analyses. The B. oleae spring onset model notifies farmers when to turn on the traps in early spring. Once the traps are on and the B. oleae population increases, the control decision tree (CDT) is run in parallel with the wireless sensor network (WSN) of traps that provides data on air temperature (T), relative humidity (RH) and olive fruit fly count. Up till now, olive fruit fly control was based on an expert basis only. Now, the CDT will provide objective guidelines and recommendations for spraying of B. oleae taking into account olive fly counts and certainty criteria. The Web app allows the user to view and summarize the raw data to provide further support in the decision making around control activities against the olive fruit fly.
Potential Impact:
The new ENTOMATIC solution is expected to, not only provide economic impact to the consortium beneficiaries, as explained at the deliverable 11.2. ENTOMATIC is also expected to provide impact at different levels.
The system designed will be important to define policies, not only at a regional or country level. As olive production is an important market along the Mediterranean countries of the European Union, a common approach for all the olive producers it will benefit the common market of the EU.
Moreover, the olive producers will also benefit from ENTOMATIC. Better procedures to control the important plague of Bactrocera Oleae is an old demanding from them. This new system, that provides real-time information of the infestation by means of using a remote system will increase the prompt reaction to the appearance of the plague and will help producers to fight against it with a better knowledge and with enough advance in order to reduce loses.
Last but not least, one of our more important expected benefits is the environmental benefits. The regulation on the use of pesticides is also an important feature in which European Union is putting more and more efforts. In that sense, a better control of the plague will lead to olive growers to a better application of pesticides at their orchards due to a more precise and instantaneous information on the infestation of their trees.
A more in-depth explanation of the expected impacts are provided below regarding each of the levels where ENTOMATIC is expected to provide benefits.

Impact on Standards and Policies
EU regulations for the Olive sector introduced are designed to improve oil quality, decrease environmental impacts and implementation of more effective monitoring methodologies. ENTOMATIC is an advanced solution to help SMEs comply with such regulations, and contains elements which could advance existing regulations. The EU Olive standards of Protected Designation of Origin (PDO) can also benefit from ENTOMATIC. Despite the stringency of these standards, there is great variability in quality within the same PDO. This situation can be reversed through the use of ENTOMATIC.
Finally, ENTOMATIC contributes to environmentally responsible practices via the use of precision agriculture to better control the use of pesticides. With the precise field information provided by ENTOMATIC, crop management becomes more efficient, allowing for limited interventions (where before blanket practices were used), thus promoting sustainable agriculture in-line with Common Agricultural Policy (CAP) standards and REACH objectives. ENTOMATIC will contribute to a stricter use of PPPs making growers comply with the new Plant Protection Products (PPPs) regulation.
Community Impacts
ENTOMATIC is expected to aid supporting a traditional EU activity that shapes regional communities: The European Olive Associations and their SME members have been severely hit by job and revenue losses for the past six years. Additionally, EU regulations impose an ongoing restructuring and modernisation of Olive orchards, threatening jobs in a universe of 2.5 million workers. The ability to improve field management and produce quality olives for high quality extra-virgin oil-making will improve the revenues of Associations and growers.
ENTOMATIC will help preserving and promoting employment: the technology can help to preserve and expand Olive growers’ businesses, thereby supporting employment in EU rural areas, which face particular challenges related to growth, jobs, lower income and technological gaps.
Europe’s rural communities are under threat. With farmers’ incomes only about half the average EU wage, it’s no surprise that over the last decade agricultural employment fell by 25 percent. Every year Europe has 2 percent fewer farmers. Around 60 percent of the EU population lives in the countryside which covers 90 percent of the Union’s territory.
In the Mediterranean countries of the EU Olive growing and Olive oil production is an important source of employment. If we take the example of Spain, the olive sector is considered to be strategic in a country with an unemployment rate of practically 25%. In recent declarations, the Spanish minister of agriculture stated that the olive sector has been greatly promoted by Spanish SME-AGs, and the growth in exports make this sector one of the keys for reducing unemployment in the country. A major factor to achieve this goal is the investment on R&D, to boost the production of high quality olive fruits and olive oil, which represent almost 15.0% of all agro-food exports of the country.
Evidently the Spanish situation can be extrapolated to all olive producing countries. However, job destruction in the olive oil sector has been a continuous threat to this activity, as in the EU the job destruction rate (16.6%) has exceeded the job creation rate (13.0%) by 3.6%.
ENTOMATIC is a cost-effective and user friendly and will allow to increase the knowledge base of Olive growers in these regions. A clear contribution will be made to employment as the European Olive and Olive oil industry becomes more value added and competitive within the EU and globally, protecting the 2.5 million direct jobs generated by the olive sector
The use of the ENTOMATIC system will reinforce the production of high quality olive fruits and olive oil, strengthening potential exports. This will help tackling the job destruction rate and reduce it by 2.6 percentage points in a five-year frame, and increase the job creation rate by at least 1.0%. ENTOMATIC will also help creating new jobs through the manufacture and sale of the system and through the development of a standard Integrated Pest Management system.
As for indirect employment, ENTOMATIC will generate employment in companies responsible to represent and distribute the system in different regions, as well as Precision Farming consulting companies who will be able to create specialized services to assist their clients. At least one trained specialist per company will be required to perform sales and services with the ENTOMATIC system.
Environmental Impacts
ENTOMATIC will contribute to environmental protection and sustainable production: ENTOMATIC will reduce the environmental impact of pesticides, complying with CAP standards and PPP regulations. The proposed project will facilitate the implementation of agro-environment schemes that address environmental problems of the sector by encouraging improved and sustainable procedures. Olive growing also has positive impact on the environment and the conservation of the landscape. It is an essential factor in combating desertification, which is one of the greatest environmental problems in the Mediterranean regions. National and transnational authorities will benefit from ENTOMATIC when monitoring the evolution and occurrence of the pest infestation, with significant associated costs.
ENTOMATIC will promote health and safety to consumers and EU citizens in general will benefit from reduced need to use pesticides, which will have a significant impact on the health of EU citizens. The proposed ENTOMATIC technology will also reduce the exposure of workers to extreme conditions, since while collecting data in standard traps they face exposure to pesticide residues under high temperatures and direct sun. Toxic reactions may be worse for those suffering from dehydration and warmer temperatures also may increase the toxic effects. These factors mean that field labourers working in the heat may be more susceptible to poisoning.
ENTOMATIC will support rural development: Rural areas, representing more than 90% of EU territory and containing more than half of the EU population, face challenges related to growth, jobs and sustainability. Most Olive growing and Olive oil production is concentrated in less-developed rural regions of the Community: Spain, France, Italy, Portugal, Greece, Cyprus, Slovenia and Malta are among the Member States which benefit from the cohesion funds. The proposed project will facilitate the implementation of recent European rural development programmes.
ENTOMATIC will enhance knowledge and quality of work and life: This project will also help to increase the quality of work and producers’ satisfaction, providing them with new knowledge and technology that will benefit them by improving the management of their olive orchards.
Transnational Approach
Olive growing is a sector of economic importance in 8 of the EU-27 member states, as well as in Turkey, Israel, etc. In the sector, SME-AGs play a central role in performing oil production, bottling and distribution tasks for their members, who typically have small orchards and who could not do these tasks alone. Consortium SME-AGs recognize the need for EU-wide implementation of the ENTOMATIC technology as the Olive fruit fly affects all producers and that improvements in field management and crop quality must be performed at transnational level to be effective. EU-wide implementation of the ENTOMATIC technology will only be obtained if its benefits for users (including cost-effectiveness) are properly disseminated. Therefore, training programs and dissemination activities in several countries have been planned and will be accomplished by the SME-AGs. ENTOMATIC will also contribute to the European Research Area (ERA) by joining the know-how of entities from different countries to create a new transnational IPM system.

List of Websites:
The web page of the project is the following:
All relevant contacts can be found in the web page.