Month: June 2020

We have the pleasure to release the first SWAMP Newsletter, to spread the word about the concepts, efforts, practical experiences, and lessons learned during the development of the project. As a multi-partner, multi-national, and bi-continental research collaboration, SWAMP has been collecting valuable experiences regarding the smart management of water resources, particularly for irrigation purposes in agriculture. Agriculture is the biggest consumer of freshwater in the world, amounting up to 70%, which makes a case for smart water management to guarantee water and food security to the world’s population.
In the SWAMP Newsletter #1, we expect to convey valuable information to different interested stakeholders regarding impact creation measures of communication, dissemination, and exploitation. Notably, this issue is featuring:
• Summary of SWAMP Communication Activities
• Summary of SWAMP Dissemination Activities
• Summary of SWAMP Exploitation Activities
• SWAMP Interest Group – Call for Members
• News from the SWAMP Pilots
• Summary of recent research papers
If you are willing to obtain additional information regarding the SWAMP Project’s findings, please do not hesitate to contact us by any means, including the official project email at contact@swamp-project.org. Also, if you want to be a member of the SWAMP Interest Group, which promotes the exploitation of SWAMP project results firsthand, please sign up here.

Sincerely,

Juha-Pekka Soininen & Carlos Kamienski
(on behalf of the SWAMP partners)

 

During the Covid emergency SWAMP activity on the Reggio Emilia pilot progressed mostly on line.

The lockdown started in Italy on March 9 and ended on May 4.

Last SWAMP wide-ranging mission on the fields, was planned at light-speed on February 27, when the fear for the lockdown was increasing. And the first happy missions after the lockdown are dated May 12 (ground mission) and May 22 (drone flights mission).

On February 27, the first Drone mission over the 3 pilot farms was completed, and this was the only February held mission during the three years  project span. The mission was complemented by local LAI estimation with ceptometer-based measurements, the first of the year, with very little vegetation at that time (see picture).

On the same day a piezometer, that had its cable cut from a tractor some time before was put back in place after being repaired at UNIBO.

The piezometer is located at -4.5 m depth, in a vineyard, in the area where the soil moisture sensor from Andrè is installed underground. This sensor sends its data, as a pulsing heart, every hour through a Lora connection, since its installation occurred almost a year before (June 7 2019), and it acts also as a vital sign of the swamp platform in Italy.

In the same place there are three Meter soil moisture, conductivity and temperature muti-sensors installed at three different depths. Data were previously collected on a data logger and on the same mission the data logger was replaced by a LoRa multisensor node entirely designed and developed within SWAMP.

On the same mission a faulty data logger was found in another farm and promptly replaced with the data logger just removed from the vineyard.

On the same day the Dugarolo App was tested in its ability to support the operator in the identification of the channels in the irrigation area.

On Feb 29 Jeferson Cotrim, from UFABC who spent 6 months working on the setup of the SWAMP LoRaWAN servers at UNIBO left Bologna, to go back home, just in time to avoid being catched by the lock down in Italy.

Few days after the 27^th feb mission, after the lockdown start,  the multi-sensor node stopped sending data. At the same time, in Bologna, we were discussing about LoRa connections reliability and coverage. So we were envisioning to increase the connectivity reliability adding public-private redundancy.
In fact, after our first installation, the local authority LEPIDA started the set up of an experimental LoRa network intended to support also the agricultural areas.
Nowadays the network has three levels of redundancy:

• A private server using a public infrastructure (TheThingsNetwork),
• A private server using a private infrastructure (SWAMP gateway and SWAMP LoRaServer) thanks to the work of Jeferson Cotrim for the development and to CBEC for the installation,
• A public server using public infrastructure (LEPIDA’s platform), thanks to the new collaboration started between UNIBO and LEPIDA.

During the lockdown, the SWAMP Multisensory node has been updated and several features for node remote control and diagnostics are now available.

The Smart Water Management Platform is now showing its potential. Data about sensors, weather, irrigation, crops and fields are now collected by  the platform and available through its user interface.

The SWAMP team will report about the project achievements and results at a Special Session of the IEEE MetroAgriFor conference (Metrology for Agriculture and Forestry), that will be held in Trento  in November 2020.

All the interested readers  are invited to participate to MetroAgriFor and share experiences submitting a paper to the special session on Sustainable Sensing for Smart WAter Management in Agriculture (S3WAMA2020).

Simulations with Criteria 1D model were continued offline, during the lockdown period, in order to set the parameters which better represent the real conditions of the fields.

Coming soon:

A new release of the “Dugarolo App”, which is supporting the Gate Keepers in managing the water requests on a daily basis  is being tested and a final release will be available for download soon.

Testing of the deployment of 12 water level sensors are ongoing , in order to verify their  correct configuration, Lora connectivity and operation within the SWAMP platform.

The MATOPIBA pilot is a Center Pivot of 100ha out of seven installed in Rio de Pedras farm, located about 40 Km from Luis Eduardo Magalhães municipality, state of Bahia in Brazil. The city is known as the capital of Brazilian agribusiness. This on-farm research pilot is expected to be a showcase for the neighbor farmers. The Cerrado, a savannah climate subtype, prevails in the region and can help the irrigated soybean production to reach 80 sacks per hectare. However, severe droughts can reduce this number to 30 sacks per hectare, as one occurred during the 2011/2012 season,  and at a high electrical energy cost to pump irrigation water. The main MATOPIBA pilot objective is to reduce energy costs by implementing a smart irrigation system based on Variable Rate Irrigation, VRI. Four management zones were identified in the pilot area, and a site-specific irrigation model based on soil moisture data is in development to produce the irrigation prescription maps to be downloaded to the VRI controller in a daily basis. A soil sensor array is a crucial instrument for that approach.

Therefore, the first step to implement the MATOPIBA pilot was to develop a multi-depth and multiparameter soil probe solution based on the capacitive principle and LoRaWAN technology. At first, seven prototype units were installed in the pilot area to evaluate the probe behavior in operation. So far, we have obtained raw data from each management zone for eight months. Several operational issues with the probe and the LoRaWAN infrastructure (Gateway and Server) faced during the data gathering help us improve the data streaming. We have recently obtained by the gravimetric method the calibration curve for the soil collected in three depths of interest from each management zone of the pilot. It allowed us to run a post-calibration process over the raw data to obtain the soil water content necessary to feed our water need estimation model under development.

We are looking forward to seeing some results soon. The large amount of soil data collected at short sampling intervals (as short as 10 min) provides us with detailed spatial and temporal variability of soil water content. The model we are proposing is innovative by making use of recent technics of data analysis and, at the same time combining well-established models like the WOFOST crop growth model developed in Wageningen.

A question to be answered is about the advantages and possibilities of using a not-so-accuracy but low-cost capacitive soil probe in more significant numbers. The expectation is that a large volume of data can give more useful information than a small volume provided by a limited number of accurate and relatively expensive probes. We are working on an ARM architecture-based circuitry to make our IoT probe even more feasible to help us answer that question.

Last but not least, an essential next step is to turn the conventional center pivot of MATOPIBA pilot into a variable rate capable pivot, the so-called Variable Rate Irrigation, VRI. This transformation is going to be accomplished by a cooperation effort between the SWAMP project team and two relevant suppliers of center pivot industries. The first one is an innovative Brazilian maker that supplied the center pivot for the pilot farm. The second one is a well-known multinational that is suppling and partially investing in the kit of valves, the positioning system, and the controller to allow the variable-rate irrigation. The two project partner companies have a great interest in the SWAMP results.

The SWAMP Cartagena pilot is a baby-leaf spinach field farmed by Intercrop Iberica. The pilot plan for the spring crops was to continue data collection of soil moisture, the data collection of weather data using local Libelium weather station, the drone-based imaging and crop analysis, and to test our IoT-driven irrigation system. In the irrigation system, the pump and valves are implemented as IoT nodes that are controlled directly by virtual entity models in FIWARE-based SWAMP platform. The data model, also implemented as virtual entities in FIWARE describes the structure of the system.

The COVID-19 and the lockdown in Spain and in rest of Europe caused major problems. Even though the access to the farm was allowed only to the farm employees, the farmer managed to install the monitoring system and to collect the soil moisture and weather data. A short break occurred, when heavy wind collapsed the solar panel system, but replacement was found and installed be the farmer. With the irrigation system, this was supposed to be the first on-site testing opportunity with first time on-site installation of the system. Without the presence of trained staff, it was impossible to do it. Drone flights had to be cancelled as flight permission procedures seem to take a lot of time. It would not have been possible to fly at the field anyway due to the lockdown.

Despite these problems, we managed to collect data from two growth periods. While waiting for the final piloting opportunity in Cartagena, the plan is to test data collection and irrigation units in Finland. We are also extending the SWAMP platform so that the season and alternative irrigation scenarios could be simulated with real Cartagena soil model and weather data.

Due to the very nature of the project, which allies key ICT technologies with critical societal challenges in agriculture, the SWAMP project has been causing a positive repercussion within the different stakeholders interested in the project, as well as receiving significant coverage by the traditional media channels. The critical purpose of communication actions in the project follows three main lines:
1. Awareness: the creation of general awareness regarding the project target area among the general public, its importance to society and the approaches and techniques to be employed
2. Pilots: spreading the word about the experience with the deployment of pilots, awareness about their main results and understanding of the key lessons learned
3. Platform: spreading the word about the experience with the development, deployment, and utilization of the SWAMP platform, as well as awareness about its main results and understanding of the key lessons learned

In the first two years of the SWAMP Project, a broad range of communication activities have been developed, reported on the website, in the 1st year Activity Report, and the 2nd year Activity Report.

Activity	2017	2018	2019
Blog		1
Exposition			1
Interview	1	2	1
Lecture / Speech		12	2
Meeting			4
News in media channels	12	2	2
News in web portals		6	1
Panel or round table		3	5
Pitch Session		1
Workshop (participation)	1	5	2
Exhibition		5	2

 

The SWAMP Website has been the most used and accessed as the leading online communication channels, but also Twitter and ResearchGate are of greater relevance.

Dissemination refers to the spreading of technical and scientific knowledge generated within a project, an essential endeavor for its outcomes to take up that is also a measure of project impact. SWAMP includes four very relevant pilots, around which the interplay between research, public, and private partners is well shaped since the early stages of the project. Each of these facts inherently has been offering extraordinary opportunities to reach an unusually large community of stakeholders.

The target users benefited from the SWAMP dissemination involve relevant stakeholders from different sections, such as scientific communities in ICT, agriculture, and water management; IoT community; ICT community and water management system developers; agriculture community; water service providers; developers of irrigation systems and water management services; infrastructure systems developers; policymakers.

In the first two years of the SWAMP Project, a broad range of dissemination activities have been developed, reported on the website, in the 1^st year Activity Report, and the 2^nd year Activity Report.

Activity	2017	2018	2019	2020
Journal paper		2	3	1
Conference paper		8	10	2
Short paper	1	4	5
Magazine article			1
Workshop (organization)		2		1
Cluster/Community Meeting	1	7	4

An example of a dissemination activity that has been generating a significant impact in the research community is the publication of the paper Smart Water Management Platform: IoT-Based Precision Irrigation for Agriculture by the Sensors Journal in January 2019. In the journal website, this paper has 4730 views and 6886 downloads, being the most viewed paper of Sensors in 2019.

This paper has been cited by 50, according to Google Scholar.

Exploitation efforts aim to guarantee that significant project results survive after the project, taking concrete measures to exploit project results in three ways:

1. Using project results in further research activities not covered within the project
2. Developing and providing a product, process or service, which have a clear focus on the market
3. Using project results in standardization activities and policy-making or advocacy actions

SWAMP conducted an internal workshop on innovation and exploitation in 2018, where all partners showed their perspectives of possibilities, opportunities, and barriers for the exploitation of SWAMP results, always followed by a series of questions and comments. Given the profile of SWAMP consortium members, both exploitation way #1 (research, by academic partners) and #2 (market, by business partners) have been explicitly pursued. An international project derived from SWAMP has already been proposed, and it is now waiting for the final results. SWAMP concepts and technologies are aimed at a technology readiness level (TRL) of 7, which may vary for different platform components. Business partners also have a desire to transform the SWAMP technologies into services and products.

Also, SWAMP has been putting forward the SWAMP Interest Group (SIG) for bringing together stakeholders with special interests in the approaches, technologies, experiences, findings, results, and exploitation opportunities of the SWAMP project firsthand. SIG members may be companies, individuals, government agencies, third sector organizations, or any interested stakeholder. Being a member of SIG is non-binding as SIG members do not have responsibilities and may allocate effort at their discretion.  If you (or your organization) are interested in being a member of the SIG, please register here.

 

The SWAMP project recently published a paper in the Sensors Journal summarizing theory and practice with designing IoT Architectures, developing IoT Platforms, and deploying IoT Smart Applications. Architecting and Deploying IoT Smart Applications: A Performance-Oriented Approach makes use of the experience with smart irrigation gained since the beginning of the project and broadens it to different IoT Smart Applications. We recognize that diverse smart applications (such as cities, agriculture, healthcare, and industry) share common characteristics influenced by the design of the IoT Architecture and by the inherently distributed nature of the IoT deployment process. Thus, valuable lessons learning with agriculture can be used in other areas, and that is confirmed with a performance analysis study.

IoT architectures provide almost no hints on where components should be deployed. In such a complex environment, a one-size-fits-all approach does not adapt well to varying demands and may hinder the adoption of IoT Smart Applications. We propose a 5-layer IoT Architecture (IoTecture) and a 5-stage IoT Computing Continuum (IoTinuum), as well as provide insights on the mapping of software components of the former into physical locations of the latter.

Also, we conduct a performance analysis study with six configurations where components are deployed into different stages.

Our results show that different deployment configurations of layered components into staged locations generate bottlenecks that affect system performance and scalability.

The SWAMP project recently published a paper in a special issue of the IEEE Internet of Things Magazine focused on IoT-based smart agriculture. Advancing IoT-Based Smart Irrigation is the first SWAMP paper to specifically address the combination of IoT and artificial intelligence, together with physical agronomical models, to provide better irrigation prescription plans that at the same time save water and energy and improve productivity.

We developed an IoT-based platform for smart irrigation, with a flexible architecture to easily connect IoT and Machine Learning (ML) components to build application solutions. Our architecture enables multiple and customizable analytical approaches to precision irrigation, giving room for the improvement of ML approaches. Impacts to different stakeholders can be anticipated, as IoT professionals, by facilitating system deployment, and farmers, by providing cost reduction and safer crop yields.

The water need estimation process is divided up into two key activities. Soil water content and dynamics estimation consists in estimating soil water content and dynamics through the direct measurement of soil water content, rainfall, and irrigation, as well as physical models of soil water dynamics applied over the collected data (weather data mainly) and soil and crop characteristics. Soil water need forecast consists in calculating soil water content forecasts and water need forecasts for each moment of a planning horizon, using techniques such as simulation and machine learning algorithms.

We developed a smartphone Farmer App where farmers are informed of immediate water needs, water balance time series, and the current soil moisture information for a 3-depth sensor probe. With this real-time status of the farm at hand, and equipped with the optimized irrigation plans computed by the system, the farmer can achieve better use of the water resources without harming productivity.

SWAMP project pilots have just been deployed, they are operating properly, and data is being collected. The next step, expected by the end of 2020, is to analyze the data and disseminate quantitative impact results.

The SWAMP project recently published a short paper in the 9^th Latin-American Symposium on Dependable Computing (LADC 2019) presenting a secure end-to-end data flow for an inherently distributed IoT smart application. End-to-End Security in the IoT Computing Continuum: Perspectives in the SWAMP Project discusses how the SWAMP project is dealing with end-to-end security and technologies for the use of IoT in agriculture. The security of IoT systems is a challenging task, spanning data generated in the sensors until the user application passing through different stages of intermediate computing elements such as mist, fog, and cloud.

For the FIWARE-based security, Segment 1 (Sensor ↔ LoRa App Server) depicts the use of the standard LoRaWAN security features provided by the LoRa Server Project. Segment 2 (LoRa App Server ↔ Orion) describes the LoRa App Server sending packets to an MQTT broker connected to an IoT Agent. Segment 3 (Orion ↔ Applications) depicts the FIWARE domain, where the setup can be configured to require HTTPS among all GEs.

The same idea applies for SEPA based security, but with four segments. Segment 1 (Sensor↔ LoRa App Server) is similar to FIWARE. Segment 2 (LoRa App Server ↔ SEPA Broker) is similar to FIWARE, but the MQTT Adapter of the SEPA Architecture plays the role of the IoT Agent. In Segment 3 (SEPA ↔ SEPA), the communication is performed via HTTPS and WSS (Web Service Security) and using OAuth 2.0. The same applies to Segment 4 (SEPA Broker ↔ Applications), where the communication between Applications and SEPA Broker is performed via HTTPS, WSS, and OAuth.