Content uploaded by Viviana Piermattei
Author content
All content in this area was uploaded by Viviana Piermattei on Jun 16, 2021
Content may be subject to copyright.
255
Medit. Mar. Sci., 22/2 2021, 255-269
Mediterranean Marine Science
Indexed in WoS (Web of Science, ISI Thomson) and SCOPUS
The journal is available on line at http://www.medit-mar-sc.net
DOI: http://dx.doi.org/10.12681/mms.25060
Research Article
Toward the widespread application of low-cost technologies in coastal ocean observing
(Internet of Things for the Ocean)
Marco MARCELLI1,2, Viviana PIERMATTEI1,2, Riccardo GERIN5, Fabio BRUNETTI5, Ermanno
PIETROSEMOLI8, Samuel ADDO3, Lobna BOUDAYA4, Richard COLEMAN6, Nubi OLUBUNMI7,
Johannes RICK9, Subrata SARKER10, Zacharie SOHOU11, Marco ZENNARO8, Karen Helen WHILTSHIRE9,
The OPENMODS Crew and Alessandro CRISE5
1 Laboratory of Experimental Oceanology and Marine Ecology, DEB, University of Tuscia, Civitavecchia (RM), Italy
2 CMCC Foundation - Euro-Mediterranean Center on Climate Change, Ocean Predictions and Applications, Bologna, Italy
3 Department of Marine and Fisheries Sciences University of Ghana
4 Laboratory of biodiversity and aquatic ecosystems, Faculty of Sciences of Sfax, Sfax University, Tunisia
5 OGS (National Institute of Oceanography and Applied Geophysics - OGS), Trieste, Italy
6 Institute for Marine and Antarctic Studies (IMAS), University of Tasmania, Hobart, TAS, Australia
7 Nigerian Institute for Oceanography and Marine Research, Victoria Island, Lagos, Nigeria
8 UNESCO Abdus Salam International Centre for Theoretical Physics (ICTP), Trieste Italy
9 Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research-AWI
10 Department of Oceanography, Shahjalal University of Science and Technology, Bangladesh
11 Institut de Recherches Halieutiques et Océanologiques du Bénin, Cotonou, Benin
Corresponding author: v.piermattei@unitus.it
Contributing Editor: Sarantis SOFIANOS
Received: 19 October 2020; Accepted: 12 March 2021; Published online: 8 April 2021
Abstract
The ability to access user-friendly, low-cost instrumentation remains a limiting factor in coastal ocean observing. The majority
of currently available marine observation equipment is difficult to deploy, costly to operate, and requires specific technical skills.
Moreover, a harmonized observation program for the world’s coastal waters has not yet been established despite the efforts of the
global ocean organizations. Global observational systems are mainly focused on open ocean waters and do not include coastal
and shelf areas, where models and satellites require large data sets for their calibration and validation. Fortunately, recent techno-
logical advances have created opportunities to improve sensors, platforms, and communications that will enable a step-change in
coastal ocean observing, which will be driven by a decreasing cost of the components, the availability of cheap housing, low-cost
controller/data loggers based on embedded systems, and low/no subscription costs for LPWAN communication systems. Con-
sidering the above necessities and opportunities, POGO’s OpenMODs project identified a series of general needs/requirements
to be met in an Open science development framework. In order to satisfy monitoring and research necessities, the sensors to be
implemented must be easily interfaced with the data acquisition and transmission system, as well as compliant with accuracy
and stability requirements. Here we propose an approach to co-design cost-effective observing modular instrument architecture
based on available low-cost measurement and data transmission technologies, able to be mounted/operated on various platforms.
This instrument can fit the needs of a large community that includes scientific research (including those in developing countries),
non-scientific stakeholders, and educators.
Keywords: Internet of things; low-cost technologies; ocean observations.
Acronyms
ARGO-Global array of free-drifting profiling floats; AUV-Autonomous Underwater Vehicle; DBCP-Data Buoy Cooperation
Panel; EOV-GOOS Essential Ocean Variable; EV -GOOS Essential Variable; GEOSS-Global Earth Observation System of Sys-
tems; GLOSS-Global Sea Level Observing System; GOOS-Global Ocean Observing System; GO-SHIP-Global Ocean Shipbased
Hydrographic Investigations Program; GSM-Global System for Mobile; LoRa-Long Range; LPWAN-Low Power Wide Area
Network; LoRaWAN-LoRa Wide Area Network; MPPRCA-Marine Plastic Pollution Research and Control Act; MSFD-Marine
256 Medit. Mar. Sci., 22/2 2021, 255-269
Strategy Framework Directive; NB-IoT-Narrow Band Internet of Things; NTC-Negative Temperature Coefficient (thermistor);
OceanSITES-A worldwide deep-water reference stations; OECD-Organization for Economic Co-operation and Development;
OpenMODs-Open Access Marine Observation Devices; POGO -Partnership for the Observation of the Global Ocean; ROV-Re-
motely Operated Vehicle; SOOP-Ship of Opportunity Program; SOT-Ship Observations Team; TUV-Tower Underwater Vehicle;
UNEP-UN Environment Programme; USV-Unmanned Surface Vehicle.
Introduction
Most global coastal areas are located in countries
with low/medium GDP per capita. Although most hu-
man marine activities take place in the coastal zone, this
area is seldom regularly observed. On the other hand,
UNEP (2016) predicted that in the next 50 to 100 years,
up to 70% of the world`s population will be living in the
coastal zone, affecting and being affected by it. The con-
comitant growth of ocean-based economic activities will
produce 2.6 trillion euros by 2030 according to OECD
estimates (OECD 2017). Over past decades, the global
ocean environment is changing rapidly because of nat-
ural and anthropogenic pressures. For all these reasons
continuous monitoring of coastal and marine systems is
becoming more and more important.
Within the Sustainable Development Goal 14 (life be-
low water), in the Revised Roadmap for the UN Decade
of Ocean Science for Sustainable Development (2021-
2030) (Ryabinin et al. 2019) the need is felt to improve
the observation capabilities in coastal areas, especially
in developing countries. Similar recommendations can
be found in Tsukuba ( in 2016) and in Turin (in 2017)
documents approved by G7 Ministers of Science and Re-
search, as in the Charlevoix blueprint for healthy oceans,
seas, and resilient coastal communities (Kirton, 2018).
To foster the ocean monitoring in the developing
countries, the development of user-friendly and low-cost
instruments is essential. However, the ability to access
user-friendly and low-cost instrumentation is still a lim-
iting factor in ocean sciences because the majority of
marine observation equipment is difficult to deploy and
costly to operate since it requires specific technical skills.
A shortage of local skills in ocean observation and in-
terfacing capacities with stakeholders further exacer-
bates the problem (Miloslavich et al., 2018). Moreover,
a harmonized observation program for the world coastal
waters has not yet been established despite the efforts of
the GOOS organization, which is coordinating the assess-
ment of ocean observing requirements, observing system
implementation, and innovation through GOOS Projects
(Tanhua et al., 2019). As highlighted by Tanhua et al.,
(2019) and reported by the Panel for Integrated Coast-
al Observation of GOOS (Digiacomo et al., 2012) many
areas have too infrequent, sparse, inadequate, or impre-
cise ocean observations; moreover, many new technol-
ogies are under development or their implementation on
regional to global scales is very limited. A large part of
the GOOS existing programs (ARGO, DBCP, GO-SHIP,
OceanSITES, SOOP) are focused on open ocean waters
and do not cover at all the coastal and shelf areas while
models and satellites require reliable and rich data sets
for their calibration and validation in coastal zones – es-
pecially for biogeochemical variables. Yet even as the
importance of coastal oceans continues to increase, our
knowledge of these areas remains limited with obvious
limitations in implementing informed management (e.g.,
Maritime Spatial Planning), fulfilling environmental reg-
ulations (e.g., European MSFD, Canada’s Oceans Act,
United States of America Shore Protection ACT, Japan’s
Basic Environment Law), and responsibly exploiting the
marine resources. The lack of data and knowledge has an
impact even more evident in the least developed coun-
tries and in small island developing states (SIDS) where
the direct and indirect dependency on marine resources is
often crucial for their survival.
Fortunately, technological advancements have recent-
ly led to novel improvements in sensors, platforms and
communications systems that will enable a step-change
in coastal ocean observations because of the lower costs
of the components (while maintaining precision and ac-
curacy sufficient for many applications), cheap housing
availability, low-cost controller/data loggers based on
embedded systems and low-cost LPWAN communication
systems. The growing data availability in heterogeneous
sources (e.g., citizen science), from sectors traditionally
reluctant to share environmental data (e.g., fishing, fish
farming and non-renewable energy) and the above-men-
tioned technological improvements will transform the
observing philosophy in the coastal area. This will lead
to the implementation of the Internet of Things paradigm
in surface (Yang et al., 2018; Wright et al., 2016). and
underwater applications (Kao et al., 2017, Abdillah et
al., 2017). Previous reviews address in situ autonomous
ocean observing methods and sensors both generic (Mills
et al., 2012; Crise et al., 2018) and application specific
(Danovaro et al., 2016; Kumari et al., 2019).
Moreover, many important projects, such as NEXOS
(Delory et al., 2014; http://www.nexosproject.eu/),
SCHeMA (http://www.schema-ocean.eu/), COMMON
SENSE (https://www.commonsenseproject.eu/) and
Sense Ocean (http://www.senseocean.eu/) contributed in
the last years to implement ocean observing capabilities
and to support policies also through the development of
low-cost technologies.
Here, we report some examples of the existing avail-
able technologies for sensors, platforms and data trans-
mission suitable for sustained coastal observing systems
with specific attention paid to cost-effective fit-for-pur-
pose technologies. We also propose a modular multi-plat-
form architecture based on available low-cost measure-
ments and data transmission technologies that can fit
the needs of a large community that goes from scientific
research in developing countries to operational use by
non-scientists for educational purposes. However, its po-
tential is not limited to the implementation of observing
257
Medit. Mar. Sci., 22/2 2021, 255-269
infrastructures for developing countries but also for ap-
plications in remote and poorly observed regions.
Many of the ideas and the analysis presented here stem
from major achievements from the POGO’s OpenMODs
project (https://pogo-ocean.org/innovation-in-ocean-ob-
serving/activities/openmods-open-access-marine-obser-
vation-devices/) that produced the motivation and the
background for this paper.
Needs and requirements for coastal observation for
science and society
Concerning the priorities of variables to be observed,
the GOOS Expert Panel has identified the Essential
Ocean Variables (EOVs) based on the following criteria
(Miloslavich et al., 2018):
• Relevance: the variable is effective in addressing the
general GOOS issues of climate, ocean operative
services and ocean health.
• Feasibility: the observation or calculation of the vari-
able on a global scale is technically feasible using
proven and scientifically valid methods.
• Cost-effectiveness: generating and storing variable
data is convenient and relies mainly on coordinated
observation systems that use proven technology, ex-
ploiting, where possible, historical datasets.
Owing to the OpenMODs contexts and scope, a subset
of the EOVs will be considered here based on the appli-
cation requirements (sea temperature, salinity, pressure,
chlorophyll a, turbidity and dissolved oxygen). However,
an improving process of the EOVs led to include biogeo-
chemical and biological variables (Muller-Karger et al.,
2018), so a general overview will be done since a small
variety of cost-effective sensors are available (Wang et
al., 2019), as shown in Table 1.
The focus of the observational efforts on a limited
number of EOVs (and companions EVs for the climate
and the biodiversity) has also been recently proposed by
Reyers et al., (2017) to meet similar (but more ambitious)
socio-economical goals.
OpenMODs proposes a way to respond to different
operational scenarios and, according to the operational
needs, to easily integrate different sensors in different
platforms.
The general issue is to involve scientific institutes and
universities from developing countries interested in im-
plementing the OpenMODs infrastructure. This approach
needs to co-design the functionalities and the operation-
al mode of a coastal observing network working closely
with the potential users to meet their requirements.
Another fundamental aspect is the production of a
blueprint of the architecture of a modular platform ca-
pable of hosting the basic sensors to conceive/identify
an easy-to-use, flexible, and affordable core set of ocean
sensors and platforms, leveraging recent advances in
telecommunications technologies particularly suited for
coastal areas.
Advancement steps are to revise the requirements/
progress in the preparation of the pilot studies that imple-
ment the OpenMODs philosophy in terms of education,
science, and services and to pave the way for future ini-
tiatives.
Considering the above criteria, OpenMODs identified
a series of general needs/requirements to meet the general
objectives of the project. Here we report those related to
the technological development:
• choice of essential ocean variables that meet so-
cio-economic priorities (temperature, salinity, chlo-
rophyll a, turbidity, currents) via a comparative mar-
ket analysis of relevant low-cost sensors;
• definition of a simple modular design of autonomous
platforms hosting multiple sensors for coastal ocean
observations with cost-effective telecommunication
capabilities.
• open science approach to remove the barriers for
sharing/reuse any kind of output, resources, meth-
ods, or tools at any stage of the development process.
State of the art of cost-effective instruments
Many low-cost instruments and sensors have been
developed with different characteristics in terms of mea-
surement performance (accuracy and sensitivity) and use
(Albaladejo et al., 2010; Piermattei et al., 2019). Some of
them have been installed on buoys or fixed installations
(i.e., on different measurement platforms); others have
remained in an experimental laboratory phase. There are
many shared initiatives towards the use of low-cost, mod-
ular, flexible and open source systems for marine-moni-
toring networks (Jiang et al., 2009).
This requirement arises from the cost of commercial
sensors and probes that limits the creation of extended
monitoring observatories (Crise et al., 2018; Beddows
et al., 2018). Thus, the scientific community has created
cost-effective and open-source components to increase
data availability. We acknowledge that the purchase cost
of equipment is only one, yet in many cases significant,
part of the overall cost incurred in operating in situ ma-
rine instruments. Deployment/installation, operation
and data communication/handling are items which can
have a substantial impact on the overall cost. The final
impact of these items is however largely site and instal-
lation dependent. The installation costs for the same in-
strument changes dramatically if it is installed in front
of your laboratory with a wi-fi connection or in an Ant-
arctic base connected via Iridium. We therefore will not
consider these costs since they are too much application
- and site-dependent. The cost of calibration can be safely
disregarded since the most effective way to maintain a
sensor within the expected accuracy and precision is to
regularly substitute it according with a carefully designed
strategy of preventive maintenance (that can progressive-
ly be trimmed on the basis of the acquired experience and
the impacts of local conditions.
In this section we will report some examples of sen-
sors (commercial or non-commercial) and components,
especially employed in marine environment, such as the
Arduino microcontroller, which is a customizable, low-
258 Medit. Mar. Sci., 22/2 2021, 255-269
cost, and user-friendly platform for data collection (Lock-
ridge et al., 2016). This kind of technology supports the
interface of a wide range of sensors to increase the ob-
servation potentiality and the resolution of the existing
sampling efforts (Marcelli et al., 2014; Lockridge et al.,
2016). This approach can be applied to a broad spectrum
of purposes.
Sensors
Among the wide variety of sensors available, we pri-
oritized the basic physical and chemical variables that are
fundamental to describing the oceanographic processes.
Marine instruments for both physical and bio-opti-
cal measurements are extremely varied. A detailed and
continuously updated catalogue of the available instru-
mentation applied to marine observation is reported by
the Alliance for Coastal Technology (ACT: www.act-us.
info), which is composed of research institutions, man-
agers and private companies promoting the development
of new effective and reliable sensors and platforms to be
used in the ocean environment.
Some examples of commercial instruments and their
specifications are reported in the appendix and sum-
marized in different tables organized by measure type
(Appendix Table 1). The tables report a cost catego-
ry classified as follows: very low € (€0 - 200), low €€
(€200 - 1,000); medium €€€ (€1,000 – 5,000); high €€€€
(€5,000 - 10,000). The cost of commercial high-end in-
struments may exceed €100,000 (Davis et al., 2016), but
this can be justified by the need for accurate and reliable
observations for scientific purposes. On the other hand,
extremely cheap commercial sensors can also be found,
but these often have very low resolution and poor stabil-
ity, which are inadequate for most marine applications.
Cost-effective sensors that still meet performance re-
quirements often have small and low-cost components
as well as modular technology; however, they are often
reliable with an adequate resolution for marine applica-
tions. Examples can be found in the Internet platform of
“Oceanography for everyone” where you can exchange
ideas and propose projects and where you can find de-
signs of a 100 m max depth CTD with components costs
of about 260 € including a low-cost Niskin bottle (about
130 €) (Thaler et al., 2013). Some examples of cost-ef-
fective marine sensors and probes combined with their
main specifications are reported in Table 2.
Temperature sensors
Temperature is a fundamental parameter because it
can affect other measurements such as salinity and den-
sity. It also influences pH, dissolved oxygen, and biogeo-
chemical processes.
The choice of the sensing element is important for
modularity, stability, and accuracy: The PT100 or PT1000
RTDs (platinum resistance temperature detector) offers
good linearity and stability and can have different reso-
lutions and increasing accuracy, meeting the modularity
concept.
Commercial circuits with adequate quality can cost a
few euros; sensing elements have variable costs depend-
ing on quality.
Thermistors are cheap and guarantee a fast acquisi-
tion, but they are not linear in their measurement per-
formance and therefore require complicated electronics.
Suitably conditioned and well calibrated thermistors can
reach the level of accuracy and stability required. Sensor
calibration and conditioning are essential to correct mea-
surements.
Thermistors can be chosen at a cost ranging from
about 15 - 30 € to hundreds of €. To achieve the high-lev-
el performances, NTC-type thermistors can be integrated
into a Wien bridge circuit. If the circuit has very high
technical characteristics, then high accuracy temperature
measurements can be obtained. The cost obviously in-
creases with the quality of the measuring circuit.
The market offers many low-cost (all the prices
should be considered as purely indicative) and low-reso-
lution sensor solutions, which seem to be sufficiently re-
liable for use in many applications. For example, ATLAS
Scientific proposes a PT1000 temperature sensor with an
accuracy of 0.15° C, and Adafruit offers both a pre-wired
and waterproofed version of the DS18B20 digital sensor
(Méndez -Barroso et al., 2020) with an accuracy of ± 0.5°
C (2€) as well as a PT100 AD converter and amplifier
(13 €), and a three-wire PT100 sensor with an accuracy
of ± 0.5° C (10 €) (Faustine et al., 2014). These sensors
require data management and transmission boards, but
the vendors provide Arduino sample codes. Regrettably,
almost none of the sensors on sale provide information
related to the measurement stability, which is a funda-
mental parameter needed for long-term observations.
Salinity sensors
Salinity is the second fundamental variable for ocean-
ographic and environmental applications. In regions of
river influence, it can be also inversely correlated with
other biogeochemical variables more difficult to be di-
rectly observed with automated instruments (e.g., nutri-
ents, contaminants). A sufficient level of accuracy is sup-
posed to be 0.1 PSU for some basic purposes. Currently,
traditional methods based on conductivity and inductive
cells seem to still be the best methods, but they require
both a very good quality circuit and a system that pre-
vents fouling. Although alternative approaches to salini-
ty measurement have been trialed (tests based on optical
properties), these approaches do not currently meet mea-
surement and monitoring needs.
The measurement of conductivity by means of induc-
tive cell seems to be a simpler and more robust solution,
less subject to fouling problems since the sensing element
is not in direct contact with sea water. Even in this case,
the calibration is a critical factor. Many researchers ap-
proach the salinity measurement by experimenting with
alternative methods or trying to integrate traditional sen-
259
Medit. Mar. Sci., 22/2 2021, 255-269
sors into low-cost technologies.
Some have proposed a salinity measurement based on
the optical properties of seawater (Kumari et al., 2016);
other authors have proposed to measure salinity by induc-
tive/conductivity cell sensors (Pham et al., 2007), a very
low-cost conductivity cell, a micro-USB cable (Carmina-
ti et al., 2016), and even using a device to be applied to
a smartphone to measure refractive index and absorption
(Hussain et al., 2017). Although these low-cost technol-
ogies can have a great potential, their sensitivities and
accuracy are not yet sufficient for use in marine observa-
tions. Another progress is the conductivity sensor CTD-
SRDL which is a novel device that can be deployed on
marine mammals: it represents the first animal-borne in-
strument that provides temperature and conductivity pro-
files along the water column to 2000 m of depth (Boehme
et al., 2009).
Besides oceanographic data acquisition, these in-
struments can also record the environmental conditions
where the tagged animals live and assess the impact of
the environment on animal behavior.
Turbidity and Chlorophyll a fluorescence sensors
Different low-cost instruments for the measurement
of both turbidity and fluorescence of chlorophyll a have
been developed and tested but not yet field deployed.
Due to the low resolution and sensitivity to low concen-
trations, not all the existing sensors are suitable for scien-
tific applications. The resolution required for chlorophyll
a measurement should be 0.05 µg/l, and 0.1 NTU for
turbidity in coastal waters. The use of these sensors will
require calibration against on-site samples in addition to
laboratory evaluation.
In this section, we consider optical sensors to study
light, turbidity, and photosynthetic pigments (mainly
chlorophyll a). Several chemical variables can also be
measured by optical sensors, but we will analyze these in
the next paragraph together with traditional sensors.
In 2015, Murphy et al., (2015) developed a low-cost
optical sensor to continuously monitor turbidity in aquat-
ic environments. The sensor was composed of an LED
array source, two photodiodes, and a robust and modu-
lar electronic system, all controlled by a customized data
logger. From the commercial point of view an example of
low-cost sensors is the Atlas Scientific set, tested on both
an onboard drifter and in moored applications. It was also
compared with other commercial probes. These sensors
are a valid option for marine observations despite not be-
ing fully reliable and accurate as top commercial probes
(Lockridge et al., 2016).
Thus, a turbidity sensor based on low-cost compo-
nents (LEDs and photoresistors) was developed to moni-
tor the different water inputs in aquaculture facilities; this
is also suitable for many other applications (Parra et al.,
2018) as in the case of the fisheries management where
the increase of turbidity can decrease fish catches.
ArLoc (Arctic Low-cost probe) is a cost-effective
technology developed to acquire pressure, temperature,
and fluorescence of chlorophyll a along the water col-
umn. This probe was developed to be modular and easily
integrated onboard of different platforms with particular
attention to Arctic applications (Piermattei et al., 2019,
Marcelli et al., 2014).
Dissolved Oxygen sensors
Electrochemical sensors with a variety of sizes and
qualities and relatively low costs are commonly avail-
able. The main concern is the need for periodic calibra-
tion and maintenance.
DFRobot offers a galvanic oxygen sensor and related
Arduino shield at about 140 €, with a precision of 0.05
mg/L, but there is no data about the accuracy (Glud et al.,
2000). The Atlas Scientific oxygen kit (shield, replace-
ment membranes, calibration solutions, and circuit isola-
tor) reports interesting characteristics and a cost of about
260 € (Table 2) (Demetillo et al., 2019).
Commercial optical sensors for dissolved oxygen are
based on luminescence properties of some complex mol-
ecules (DFRobot). These sensors could be integrated into
cost-effective systems but unfortunately, they are still too
expensive. Currently there are many experimental works
and few commercial products, but these cannot currently
be integrated into automatic measurement systems.
Nutrients
The measurement of seawater nutrients represents a
big challenge because of the strong diversity in marine
environments which reflects large concentration differ-
ences. This concern affects the in-situ measurement of
nutrients that still need harmonization and standardiza-
tion procedures (Daniel et al., 2019). A variety of labo-
ratory methods are commonly used to determine the con-
centration of most nutrients with a high accuracy, while
in situ sensors have still some limitations connected to
different issues such as local concentrations, type of de-
ployment and monitoring target.
Many commercial sensors are available, as reported
by ACT database, but many of them are characterized by
high consumption, high costs, limits in the detection of
low concentrations, low autonomy for long-term deploy-
ment. No low-cost nutrient sensors are still available but
are under development.
pH and pCO2
As already seen for the oxygen sensor, marine chem-
ical sensing is very important, and some parameters can
be easily measured with low-cost sensors. Regarding pH
measurements, there are pH electrodes (DFrobot, Gravi-
ty: Analog pH Sensor / Meter Pro Kit for Arduino, sensor
H-101) and related Arduino shields at a price of around
45 € with an accuracy of 0.1. Atlas also offers different
pH tools at different costs depending on the construc-
260 Medit. Mar. Sci., 22/2 2021, 255-269
tion and stability requirements (Méndez -Barroso et al.,
2020). An experimental lab-on-a-chip pH sensor for sea-
water measuring based on spectrophotometry was pre-
sented by Pinto et. al. (2019); it is characterized by a very
low-cost (approximately 100 €) a strong miniaturization
and low consumption.
In any case, most of low-cost sensors are not suitable
for marine applications and the principal issue remains
calibration and maintenance procedures for pH drift,
which can be considerable after long-term time mea-
surement. Alternative pH sensors are based on optical
measurements, which use fiber optics based on a coat-
ing substance (in contact with seawater) via fluorescence
(Martín et al., 2006), as well as pressure compensating
semiconductor (ISFET) electrodes (Johnson et al., 2016),
however most of these sensors still have high costs.
The pCO2 sensors are now moderately mature; some
sensors can be deployed in marine environments but are
limited in depth range and low acquisition rates. A lot of
these sensors use membrane permeable to dissolved CO2
molecules whose partial pressure is determined by means
of IR absorption or colorimetric spectrometry (Wang et
al., 2019). Many components are decreasing in costs;
however, these sensors are still not typically cost-effec-
tive.
Low-Cost Technologies and Citizen Science
Citizen science can allow data acquisition in an effi-
cient and cost-effective way, and many programs have
been developed to study the marine environment via vol-
unteers. In this model, scientists train lay people to apply
the scientific method to study environmental processes
and phenomena by collecting data (Lauro et al., 2014).
The connection with citizen science is two-fold: on one
hand, the availability of cheap technology for ocean data
acquisition can favor further engagement in a larger com-
munity of non-professional data producers; on the other
hand, additional observations are made to complement
scientific applications. In developing countries, this en-
gagement is crucial, especially when fisher communities
need to be involved because they can act as a source and
destination of the data produced. This strategy has been
already applied in some pilot endeavors such as the Fish-
ery Ocean Observing System (Falco et al., 2007).
These projects must be supported by the sustainabili-
ty of available technology, and cost-effective sensors are
key to this approach. A remarkable example of citizen
science is the Smartfin realized to acquire temperature
data through common surfboards. This can offer details
on the time and space gap in surf zones (Bresnahan et al.,
2017). Aquatic recreational sports can be used as environ-
mental platforms to host low-cost sensors; for example,
sailing vessels can continuously cruise along the oceans
and collect oceanographic data (Brewin et al., 2017).
Many citizen science projects collect data via miniatur-
ized and low-cost sensors, especially temperature profiles
in coastal areas (CMFRI 2018). Moreover, the usage of
a smartphone applied to citizen science is challenging:
an example is an easy-to-use, low-cost (approximately
50 EUR excluding smartphone) affordable fluorescence
sensor (SmartFluo) based on smartphone elements, de-
veloped within the EU FP7 project CITCLOPS (Citizens’
Observatory for Coast and Ocean Optical Monitoring
(Friedrichs et al., 2017).
Sensors and sampling can both make important data
available. The NCCOS (National Centers for Coastal
Ocean Science) of NOAA created a Phytoplankton Mon-
itoring Network (PMN) for a better understanding of
harmful algal blooms through volunteer monitoring. This
initiative enhances the capacity to respond to and man-
age the growing threat posed by harmful algal blooms
by collecting important data for species composition and
distribution in coastal waters and creating working re-
lationships between volunteers and professional marine
biotoxin researchers’ (Morton et al., 2015).
Various citizen science initiatives were launched to
monitor the jellyfish abundance (e.g., the Jellywatch Pro-
gramme; Boero et al., 2009) to preserve the coral reefs
(e.g. Reef Check) (Hodgson 2001) or the coastal marine
environment (e.g. Sea Search Koss) (Koss et al., 2009). A
recent report on the state and the advancement of citizen
science in Europe has been published by the European
Marine Board (Garcia-Soto et al., 2017).
Observing platforms
The platforms for the ocean and marine observation
cover a wide range of applications being operated at sur-
face, on the bottom, onboard vessels, freely drifting, au-
tonomous on planned tracks, sliding, towed by a ship, or
remotely operated (Albaladejo et al., 2012). The platform
selection is based on the objectives of the research and
monitoring activity and by the specific site characteris-
tics.
Many commercial platforms currently measure pa-
rameters in an autonomous and continuous way; howev-
er, most of them are expensive and difficult to use.
Thus, many projects use low-cost monitoring plat-
forms for marine monitoring. One example is the SEMAT
(Smart Environmental Monitoring and Analysis Technol-
ogies) project, that promotes the creation of a low-cost in-
telligent sensor network via different measuring platform
prototypes, underwater communications, and alternative
power management schemes with a cost of approximate-
ly 3500 € (Trevathan et al., 2012). Albaladejo (2012) de-
veloped a cost-effective wireless sensor buoy system for
shallow marine environment monitoring. This tool could
acquire data at the surface, along the water column, and
at the bottom. The miniaturized buoy acquired pressure
and temperature at different depths. It could also collect
meteorological data depending on the sensor availability
and mission objectives. However, the buoy (without in-
struments) costs only €340.
Floats and drifters are the most cost-effective plat-
forms. They can be equipped with many kinds of sen-
sors enabling continuous and real-time monitoring of
the surface and deep ocean. Argo floats primarily mea-
261
Medit. Mar. Sci., 22/2 2021, 255-269
sures temperature and salinity of the upper 2000 m of the
ocean. The Argo system has a wide spatial and tempo-
ral coverage and is a key element of the Global Ocean
Observing System (GOOS), the World Climate Research
Program (WCRP) Climate Variability and Predictability
(CLIVAR) project, the Global Ocean Data Assimilation
Experiment (GODAE), and Global Earth Observation
System of Systems (GEOSS) (Roemmich et al., 2009)
allowing to acquire and make freely available a lot of data
in a long period and in a cost-effective way (Rudnick,
2018), despite the cost which is about €17000. Similar-
ly, drifters are used in many marine research fields such
as climate studies, oil spill tracking, weather and marine
forecasting, and search and rescue operations. Recent-
ly, the Global Drifter Program (GDP) was funded by
the National Oceanic and Atmospheric Administration
(NOAA) Ocean Observing and Monitoring Division of
the Climate Program Office and mainly managed by the
Lagrangian Drifter Laboratory (LDL) at Scripps Institu-
tion of Oceanography (SIO). Within this program, a se-
ries of drifters were developed to acquire currents, waves,
temperatures, and conductivity at different depths, winds,
and atmospheric pressures (Lumpkin et al., 2017), and
the price can vary between 1500€ and 5000€ depending
on the additional sensors. Other examples can be found
on the market.
Open source electronic boards and programming
language
In the last decade, different technologies based on
open-source electronic platforms have been developed
and are characterized by common points including easy-
to-use hardware and software for beginners. Among
these, Arduino (started in 2003) is by far the most com-
mon. It has simple and open-source software (IDE) lead-
ing to a large open access community of Arduino users.
A worldwide community of makers including students,
hobbyists, artists, programmers, and professionals has
joined this open-source platform helping novices and ex-
perts alike. In this way, thousands of projects have been
developed from everyday objects to complex scientific
instruments (https://create.arduino.cc/projecthub).
There are currently many small online companies that
offer low-cost components. These can be combined in a
modular way to obtain measuring instruments.
The strengths that today’s technology allowing us to
be able to develop instruments at low cost and good accu-
racy can be summarized as follows:
powerful embedded systems, cheap, widely used,
easy to program and with low development costs, allow
realizing the control, data storage and interface of the in-
strument. We can choose between very low power con-
sumption systems without an operating system such as
Arduino, programmable in C or Pyboard programmable
in Micropython, or use real embedded computers with an
operating system, such as Raspberry or Beaglebone that
support Linux.
integrated circuits, developed for example in the med-
ical field for portable systems, offer at very low cost as
they are widely used, but with all the features that allow
realizing the conditioning of the analog signal coming
from the sensor and its conversion to digital with cost/
accuracy performance unthinkable until some time ago
bringing the computation toward the sensors (edge com-
puting).
Data transmission technologies and Internet of Things
of the Ocean (IoT-O)
Internet of Things allows objects to communicate
with each other and to be remotely controlled. It com-
bines existing internet infrastructures with optimized
wireless communication systems to directly integrate the
physical and digital worlds.
Several elements are fundamental to implement sus-
tainable cost-effective IoT solutions: (i) low cost trans-
mitting devices; (ii) low power consumption; (iii) avail-
ability of a network that can support a large number of
connected objects; and (iv) a very wide geographic cov-
erage reachable by as many elements as possible. The
main impediments are related to: (i) network manage-
ment costs; (ii) scalability and network organization; (iii)
edge-nodes dimensioning and power efficiency; and (iv)
coverage (Sanchez-Iborra et al., 2016).
Sensors to be managed via an IoT approach must also
be low cost with low power consumption. Therefore,
an observing platform that can send data (hourly, dai-
ly, weekly or monthly) and that can be battery powered
should consume as little as possible both in transmission
and in reception. Network technologies that satisfy these
needs of low power, good coverage, and network scal-
ability are called Low Power Wide Area Networks, LP-
WAN.
Traditional cellular technologies have been widely
deployed especially in densely populated spaces and are
very good at meeting the needs of voice and high-speed
data communications. Nevertheless they fall short at
meeting the requirements of IoT for the following rea-
sons: (i) the end device must periodically communicate
with the base station even when it has no data to be trans-
mitted – this can consume significant amounts of energy
in the idle state: (ii) coverage is often lacking in areas
of low population density that cannot normally offer an
attractive business case for the operator; (iii) the recur-
ring monthly costs for each device, while adequate for
human users, are comparatively too high for objects with
limited communication requirements; (iv) the coverage
of a base station is limited to a few kilometers and cannot
reach basements and cellars. These shortcomings have
been addressed by several vendors who offer proprietary
solutions specifically focused on the requirements of IoT
in terms of low cost, low power consumption, and long
range; these are achieved by limiting the transmission
rate thus precluding its use for voice transmission.
Here, the best representatives are Sigfox and LoRa.
They both use unlicensed frequencies, and thus do not
incur spectrum usage fees, but are not protected from in-
262 Medit. Mar. Sci., 22/2 2021, 255-269
terference from other users and face limitations in trans-
mission times. In Europe, a given device cannot transmit
more than 1% of the time, while each transmission cannot
exceed 400 ms in the USA. The maximum radio frequen-
cy power transmission is also specified by the regulator;
it is therefore the same for both vendors.
Sigfox uses a transmission bandwidth of only 100
Hz, which is an order of magnitude less than that of cell
phones. It can reach very long distances, but with very
short and sparse messages. The Sigfox business case is
oriented at partnership with cellular service providers,
and their modules are quite inexpensive offering also low
recurring costs per device.
LoRa uses a completely different strategy to achieve
similar goals; it transmits modestly longer messages us-
ing a much wider bandwidth employing spread spectrum
modulation. This type of modulation allows for the de-
coding of very weak signals so that they can also reach
very long distances and achieve deep penetration in
buildings. The main advantage of LoRa is that it can be
installed by any interested party, and it can be deployed in
areas not reached by cellular operators. The LoRaWAN
alliance has produced an open-source protocol stack that
can be freely used to provide a complete communication
infrastructure for IoT.
Furthermore, a crowd-sourcing initiative, “The
Things Network (TTN)”, has succeeded in installing
freely usable “Gateways” that are connected to Internet
application servers in many countries. Thus, anyone can
register in the TTN web site, write an application for a
specific purpose, buy a LoRa module, configure it to ac-
cess TTN, and then leverage any reachable TTN gateway
to transport the data to the specific application. Many us-
ers can connect to a given gateway, but the data streams
are encrypted end-to-end so only the application owner
can decode the data.
It is also possible to buy a LoRaWAN gateway and
install it wherever it might be needed. The only require-
ments are Internet connectivity and a power source since
the gateway must be always on and cannot rely only on
batteries; a small photovoltaic panel can fulfill this need.
In addition, the 3GPP alliance – the organization that
establishes the protocols for the cellular industry – has
also addressed the needs of IoT in its most recent releas-
es by means of two different protocols: Narrow Band
IoT (NB-IoT) and LTE-M. Both considerably reduce the
power consumption of the end device by allowing much
longer “sleeping times” and increase the range by reduc-
ing the bandwidth and therefore the throughput. NB-IoT
employs a technique of sending several replicas of the
same message which are combined in the receiver allow-
ing deep penetration into buildings and very long trans-
mission distances. The big advantages of the 3GPP solu-
tions are that they use a protected spectrum and are better
sheltered from interference. The disadvantage is its de-
pendence on an operator to deploy service in a given area.
Another interesting approach applies a LoRaWAN
communication system to a low-cost CODE-like drifter
(Gerin et al., 2018). This drifter represents a cost-effec-
tive solution for coastal use due to the current limited dis-
tance range of alternative communications technologies.
This outperforms the standard CODE in terms of sam-
pling frequency and reusability (recovering the drifter at
the end of its mission and replacing the battery pack when
needed).
Feasibility of a new multipurpose platform
For the sake of clarity, we will hereafter use the hierar-
chy sensor-instrument-platform-observing infrastructure
proposed by Bermudez et al., (2009). They describe the
sensor or transducer as an entity capable of observing a
phenomenon and returning an observed value, the instru-
ment as a device hosting multiple heterogeneous sensors,
the platform as a vehicle carrying multiple instruments,
the observation infrastructure as the in situ multi-plat-
form infrastructure. Not all the hierarchy is always ap-
plicable: simple surface drifters equipped with GPS only
can be considered both as instruments or platforms.
In OpenMODs, we proposed the architecture of a
platform that meets a series of requirements such as: the
needs identified during the Mindelo OpenMODs meeting
(https://pogo-ocean.org/about/pogo-meeting/pogo-20/);
the use of components with sufficient technical features,
low cost, and readily available; deployable from small
boats (i.e. fishing boats); to be supported by an adequate
training and information programs also via communica-
tion tools available and accessible to all (e.g. YouTube,
Instagram).
Architecture of Low-cost Effective Ocean-observing
(LEO) instrument
The OPenMODs proposed solution is a Low-cost Ef-
fective Ocean-observing (LEO, Fig.1) instrument as a
single instrument that can be easily integrated into mul-
tiple platforms. Its technical specifications should meet
the requirements of strong miniaturization, low consump-
tion, and easy integration into all types of oceanographic
platforms, and could be a valid example.
The system architecture involves the construction of a
modular apparatus (Fig.2) able to be used individually or in-
tegrated into the greatest number of measurement platforms:
• the sensors must be easily interchangeable (i.e., bet-
ter to change old sensors with new ones than re-cal-
ibrate them).
• the control electronics must be based on “open” sys-
tems such as Arduino or other embedded systems
with different analog and digital inputs.
• LEO can operate in an on-line mode as a cabled in-
strument (direct acquisition of data in a PC) or in de-
layed mode by recording data internally and sending
data (possibly after a preprocessing step) by using
different data transmission methods (mainly LoRa
and NB-IoT but also GSM).
• LEO will guarantee the proper interface of the sen-
sors with the network, e.g., LoRa can provide access
to an observational network including sensors tagged
263
Medit. Mar. Sci., 22/2 2021, 255-269
as IoT objects.
• an interface will be provided for a GPS receiver.
• LEO will be equipped with batteries and internal
memory so that it can operate in stand-alone mode.
• the housing must be made with low-cost components
adapted from other uses such as to withstand the cor-
rosion of seawater.
Additional requirements concern the operational de-
tails:
• the launching and recovery must be as simple as pos-
sible.
• LEO will operate between the surface and depths of
200 meters (extendable in case of need);
• LEO will be made as compact as possible and with
particular attention to the weight out of the water to
make it easy to manage during transportation.
• LEO will be assembled in different ways depending
on both the use and the different platforms to which
it could be interconnected; so that the weight, buoy-
ancy, and electrical interfaces can be managed.
• the assembly of the sensors into the electronic man-
agement and power supply system, the data trans-
mission system, and the assembly inside the pressure
hull or the protection box must be easy and manage-
able even without special equipment.
The LEO flexibility can facilitate its use in different
autonomous operational modes (Fig.3) such as:
Fig. 1: LEO architecture showing the main electronic components of the instrument and their specifications (left) and its integra-
tion in different platforms/operational modes (brown: examples of different data transmission; green: second stage improvements).
Fig. 2: LEO modularity reflects internal main electronic components distribution and assemblage.
264 Medit. Mar. Sci., 22/2 2021, 255-269
• Self-recording/RT vertical profiler. It is operated
by a boat with both electromechanical cables (con-
nection with a PC) and mechanical cables (acquisi-
tion on internal memory).
• Stand-alone fixed installation. This would use a
FerryBox, fixed station of coastal measurement, and/
or be on a platform. This would have autonomous
data acquisition, storage, transmission, and power
supply (solar panels or battery packs).
• Hosted onboard different platforms (AUV, USV,
Glider, ARGO, Lagrangian Buoys, TUV and ROV).
In this configuration, the instrumentation may be in-
tegrated and adapted to different existing measure-
ment platforms and may include a serial interface
(RS485-RS232).
As a side comment, we suggest that a sustainable ob-
servational network based on low-cost technology (such
LEO) should include (when available) top-quality instru-
ments and sensors to consolidate the operational confi-
dence and roughly estimate the uncertainties of instru-
ments of LEO class. The optimal use of LEO instrument
network simultaneously with top-class instruments meets
the IoT heterogeneity and represents the state-of-the-art
in observational oceanography. Involving as many actors
as possible, and operating on different time scales at dif-
ferent resolutions, IoT heterogeneity will allow us to add
value to all the observation by verifying and certifying
the process of implementation, calibration, and effective-
ness of the whole network.
Operational guidelines and capacity creation
The calibration and standardization processes of
different technologies are important. The possibility of
using different sensors coming from different origins is
a great advantage but also a considerable risk. Besides
the accuracy and resolution of each sensor, one should
also follow the guidelines that allow the comparability
of observations obtained from different instruments and
sensors.
Another essential aspect related to the calibration pro-
cess is the long-term stability of the sensors which will
have to be addressed and verified by an experimental
phase. Stability is often a missing component in low-cost
sensors.
Additional considerations to be considered in the ap-
plications follow:
It is more convenient to change sensors, platforms,
and ancillary materials before their performance degrada-
tion occurs instead of using a costly and time-consuming
in loco calibration phase.
An in-house accurate comparison with high-quality
instruments will help to identify possible malfunctions or
evident drifts in the low-cost sensors. This in turn implies
that some of the high-end equipment and related exper-
tise should be made available locally.
Thanks to the simple design and system modularity,
assembly, sensor testing, and inter-comparison can bring
students (at various levels) closer to the scientific method
(Solomon, 1980).
It is worth noting that the training is one of the most
promising LEO application fields. The educational value
of the OpenMODs approach goes well beyond the ocean-
ographic domain. It can lead to curious, rigorous, and
quantitatively oriented minds (Hodson, 1988).
Conclusions
This paper aims to call for additional interest toward
sustained observations in the global coastal ocean (in-
cluding the waters belonging to developing countries),
define the architecture that will lead to LEO, the first pro-
totype of a new generation of cost-effective multifunc-
tion instruments. LEO takes advantage of the sensors and
components on the market that can be easily integrated
into open architectures that meet the objectives of Open-
MODs. These modular measurement platforms will be
adaptable to different operating modes. In addition, the
availability of low-cost transmission methods, such as
LoRa, increases the possibility of managing these plat-
forms by inserting them in shared systems such as IoT-O.
These can be integrated via an optimized data architec-
ture ready to be explored, (re)used, and exploited to help
leverage the novel opportunities offered by AI technolo-
gies.
The use of open sensors and electronic platforms
will certainly require more calibration and maintenance
work on the instrumentation. The community must de-
velop new protocols and adopt best practices to verify
the system functioning; local staff is needed to operate
Fig. 3: LEO Operating scheme. These operating modes require a modular development capable of being adapted to different
needs, to different communication systems, as well as to different supply possibilities.
265
Medit. Mar. Sci., 22/2 2021, 255-269
them. Capacity-building initiatives in this field will re-
ceive a big boost by modern communication platforms
(e.g., YouTube videos on basic training courses). These
also involve international organizations committed to ca-
pacity creation in the marine sector (e.g., UNESCO IOC).
Framing this technological development in the per-
spective of the IoT-O, a step-change in the way we obtain
information from ordinary objects offers an alternative
way to use instrumentation. The ability to use open sen-
sors and open electronic platforms could be the first step
towards the implementation of IoT-O.
The development of a system whose modular archi-
tecture allows the deployment of different low-cost plat-
forms – including the development of the data transmis-
sion system, calibration protocols, and stability test – is in
line with the strategies for global ocean observation (e.g.,
GOOS, POGO, GEOSS) and can dramatically improve
the observation of coastal waters worldwide.
Acknowledgement
The article was partially supported by the OpenMODs
project financed by the Partnership for the Observation of
the Global Ocean (POGO).
References
Abdillah, A.F., Berlian, M.H., Panduman, Y.Y.F., Akbar,
M.A.W., Afifah, M.A. et al., 2017. Design and develop-
ment of low-cost coral monitoring system for shallow water
based on internet of underwater things. Journal of Telecom-
munication, Electronic and Computer Engineering (JTEC),
9 (2-5), 97-101.
Albaladejo, C., Sánchez, P., Iborra, A., Soto, F., López, J.A.
et al., 2010. Wireless sensor networks for oceanographic
monitoring: A systematic review. Sensors, 10, 6948-6968.
Albaladejo, C., Soto, F., Torres, R., Sanchez, P., Lopez, J.,
2012. A low-Cost sensor buoy system for monitoring shal-
low marine environments. Sensors, 12, 9612-9634.
Beddows, P.A., Mallon, E.K., 2018. Cave pearl data logger:
A flexible Arduino-based logging platform for long-Term
monitoring in harsh environments. Sensors, 18 (2), 530.
Bermudez, L., Delory, E., O’Reilly, T., del Rio Fernandez, J.,
2009. Ocean observing systems demystified. In: Proceed-
ings of the OCEANS 2009, MTS/IEEE Biloxi-Marine Tech-
nology for Our Future: Global and Local Challenges, (Pis-
cataway, NJ: IEEE).
Boehme, L., Lovell, P., Biuw, M., Roquet, F., Nicholson, J. et
al., (2009). Animal-borne CTD-Satellite Relay Data Log-
gers for real-time oceanographic data collection. Ocean
Science 5, 685-695.
Boero, F., Putti, M., Trainito, E., Prontera, E., Piraino, S. et al.,
2009. First records of Mnemiopsis leidyi (Ctenophora) from
the Ligurian, Thyrrhenian and Ionian Seas (Western Medi-
terranean) and first record of Phyllorhiza punctata (Cnidar-
ia) from the Western Mediterranean. Aquatic Invasions, 4
(4), 675-680.
Bresnahan, P.J., Cyronak, T., Martz, T., Andersson, A., Waters,
S. et al., 2017. Engineering a Smartfin for surf-zone ocean-
ography. In: OCEANS 2017-Anchorage (pp. 1-4). IEEE.
Brewin, R.J., Hyder, K., Andersson, A.J., Billson, O., Bres-
nahan, P.J. et al., 2017. Expanding aquatic observations
through recreation. Frontiers in Marine Science, 4, 351.
Carminati, M., Stefanelli, V., Luzzatto-Fegiz, P., 2016. Mi-
cro-USB Connector Pins as Low-Cost, Robust Electrodes
for Microscale Water Conductivity Sensing in Oceano-
graphic Research. Procedia Engineering, 168, 407-410.
CMFRI, K., 2018. Nano News NF-POGO Alumni Network for
Oceans-NF-POGO Alumni E-Newsletter.
Crise, A.M., Ribera D’Alcalà, M., Mariani, P., Petihakis, G.,
Robidart, J. et al., 2018. A conceptual framework for devel-
oping the next generation of Marine OBservatories (MOBs)
for science and society. Frontiers in Marine Science, 5, 318.
Daniel, A., Laes Huon, A., Barus, C., Beaton, A.D., Blandfort,
D. et al., 2019. Towards a harmonization for using in situ
nutrient sensors in the marine environment. Frontiers in
Marine Science, 6, 773.
Danovaro, R., Carugati, L., Berzano, M., Cahill, A.E., Car-
valho, S., et al. 2016. Implementing and innovating marine
monitoring approaches for assessing marine environmental
status. Frontiers in Marine Science, 3, 213.
Davis, J., 2016. A Novel Aquatic Sensor and Network. Doctor-
al dissertation, Rensselaer Polytechnic Institute, New York,
United States of America.
Delory, E., Castro, A., Waldmann, C., Rolin, J.F., Woerther, P.
et al., 2014. Objectives of the NeXOS project in developing
next generation ocean sensor systems for a more cost-ef-
ficient assessment of ocean waters and ecosystems, and
fisheries management. In: OCEANS 2014-Taipei (pp. 1-6).
IEEE.
Demetillo, A.T., Japitana, M.V., Taboada, E.B., 2019. A system
for monitoring water quality in a large aquatic area using
wireless sensor network technology. Sustainable Environ-
ment Research, 29(1), 1-9.
DFRobot, Dissolved oxygen sensor manual [online] available:
https://www.dfrobot.com.
Digiacomo, P., Muelbert, J., Malone, T., Parslow, J., Sweijd, N.
et al., 2012. Requirements for Global Implementation of the
Strategic Plan for Coastal GOOS.
Falco, P., Belardinelli, A., Santojanni, A., Cingolani, N., Russo,
A. et al., 2007. An observing system for the collection of fish-
ery and oceanographic data. Ocean Science, 3 (2), 189-203.
Faustine, A., Mvuma, A.N., Mongi, H.J., Gabriel, M.C., Tenge,
et al., 2014. Wireless Sensor Networks for Water Quality
Monitoring and Control within Lake Victoria Basin.
Friedrichs, A., Busch, J.A., Van der Woerd, H.J., Zielinski, O.,
2017. SmartFluo: A method and affordable adapter to mea-
sure chlorophyll a fluorescence with smartphones. Sensors,
17 (4), 678.
Garcia-Soto, C., van der Meeren, G.I., 2017. Advancing citizen
science for coastal and ocean research. European Marine
Board IVZW.
Gerin, R., Zennaro, M., Rainone, M., Pietrosemoli, E., Poulain,
P.M. et al., 2018. On the design of a sustainable ocean drift-
er for developing countries. EAI Endorsed Transactions on
Internet of Things, 4 (13), 155483.
Glud, R.N., Gundersen, J.K., Ramsing, N.B., 2000. Electro-
chemical and optical oxygen microsensors for in situ mea-
266 Medit. Mar. Sci., 22/2 2021, 255-269
surements. In Situ Monitoring of Aquatic Systems: Chemi-
cal Analysis and Speciation, Wiley, Chichester, pp. 19-73.
Hodgson, G., 2001. Reef Check: The first step in communi-
ty-based management. Bulletin of Marine Science, 69 (2),
861-868.
Hodson, D., 1988. Experiments in science and science teach-
ing. Educational philosophy and theory, 20 (2), 53-66.
Hussain, I., Das, M., Ahamad, K.U., Nath, P., 2017. Water sa-
linity detection using a smartphone. Sensors and Actuators
B: Chemical, 239, 1042-1050.
Jiang, P., Xia, H., He, Z., Wang, Z., 2009. Design of a water
environment monitoring system based on wireless sensor
networks. Sensors, 9, 6411-6434.
Johnson, K.S., Jannasch, H.W., Coletti, L.J., Elrod, V.A., Martz,
T.R. et al., 2016. Deep-sea DuraFET: a pressure tolerant
pH sensor designed for global sensor networks. Analytical
chemistry, 88 (6), 3249-3256.
Kao, C.C., Lin, Y.S., Wu, G.D., Huang, C.J., 2017. A comprehen-
sive study on the internet of underwater things: applications,
challenges, and channel models. Sensors, 17 (7), 1477.
Kirton, J., 2018). A G7 Summit of Significant Success at
Charlevoix 2018, G7 Research Group, July 13. Available
http://www.g7.utoronto.ca/evaluations/2018charlevoix/
kirton-performance-full.html.
Kumari, C. U., Samiappan, D., Rao, T.R., Sudhakar, T., 2016.
Mach-Zehnder Interferometer based high sensitive water
salinity sensor for oceanographic applications. In: 2016
IEEE Annual India Conference (INDICON) (pp. 1-4).
IEEE.
Kumari, C.U., Samiappan, D., Kumar, R., Sudhakar, T., 2019.
Fiber optic sensors in ocean observation: A comprehensive
review. Optik, 179, 351-360.
Lauro, F.M., Senstius, S.J., Cullen, J., Neches, R., Jensen, R.M.
et al., 2014. The common oceanographer: crowdsourcing
the collection of oceanographic data. PLoS biology, 12(9),
e1001947.
Lockridge, G., Dzwonkowski, B., Nelson, R., Powers, S., 2016.
Development of a low-cost arduino-based sonde for coastal
applications. Sensors, 16 (4), 528.
Lumpkin, R., Özgökmen, T., Centurioni, L., 2017. Advances in
the application of surface drifters. Annual Review of Marine
Science, 9, 59-81.
Marcelli, M., Piermattei, V., Madonia, A., Mainardi, U., 2014.
Design and Application of New Low-Cost Instruments
for Marine Environmental Research. Sensors, 14, 23348-
23364.
Martín, F.J.F., Rodriguez, J.C.C., Anton, J.A., Perez, J.V., Sánc-
hez-Barragán, I. et al., 2006. Design of a low-cost optical
instrument for pH fluorescence measurements. IEEE Trans-
actions on Instrumentation and Measurement, 55 (4), 1215-
1221.
Méndez-Barroso, L.A., Rivas-Márquez, J.A., Sosa-Tinoco, I.,
Robles-Morúa, A., 2020. Design and implementation of
a low-cost multiparameter probe to evaluate the temporal
variations of water quality conditions on an estuarine la-
goon system. Environmental Monitoring and Assessment,
192 (11), 1-18.
Mills, G., Fones, G., 2012. A review of in situ methods and
sensors for monitoring the marine environment. Sensor Re-
view, 32 (1), 17-28.
Koss, R.S., Miller, K., Wescott, G., Bellgrove, A., Boxshall, A.
et al., 2009. An evaluation of Sea Search as a citizen science
programme in Marine Protected Areas. Pacific Conserva-
tion Biology, 15 (2), 116-127.
Miloslavich, P., Bax, N.J., Simmons, S.E., Klein, E., Appeltans,
W., et al., 2018. Essential ocean variables for global sus-
tained observations of biodiversity and ecosystem changes.
Global Change Biology, 24, 2416-2433.
Morton, S., Gano, S., 2015. Phytoplankton Monitoring Net-
work: Using. Available: https://coastalscience.noaa.gov/
research/stressor-impacts-mitigation/pmn/.
Muller-Karger, F.E., Miloslavich, P., Bax, N.J., Simmons, S.,
Costello, M.J. et al., 2018. Advancing marine biological
observations and data requirements of the complementary
essential ocean variables (EOVs) and essential biodiversity
variables (EBVs) frameworks. Frontiers in Marine Science,
5, 211.
Murphy, K., Heery, B., Sullivan, T., Zhang, D., Paludetti, L. et
al., 2015. A low-cost autonomous optical sensor for water
quality monitoring. Talanta, 132, 520-527.
Organisation for Economic Co-Operation and Development
(OECD). 2017. The Ocean Economy in 2030. IWA Publishing.
Parra, L., Rocher, J., Escrivá, J., Lloret, J., 2018. Design and
development of low cost smart turbidity sensor for water
quality monitoring in fish farms. Aquacultural Engineer-
ing, 81, 10-18.
Pham, T.T., Burnett, D., Handugan, L., Frashure, D., Chen, C.J.
et al., 2007. A Low-Cost, Data-Logging Salinity Sensor. In:
OCEANS 2007 (pp. 1-5). IEEE.
Piermattei, V., Madonia, A., Bonamano, S., Martellucci, R., et
al., 2018. Cost-Effective Technologies to Study the Arctic
Ocean Environment. Sensors, 18 (7), 2257.
Pinto, V.C., Araújo, C.F., Sousa, P.J., Gonçalves, L.M., Minas,
G., 2019. A low-cost lab-on-a-chip device for marine pH
quantification by colorimetry. Sensors and Actuators B:
Chemical, 290, 285-292.
Reyers, B., Stafford-Smith, M., Erb, K.H., Scholes, R.J., Sel-
omane, O., 2017. Essential variables help to focus sustain-
able development goals monitoring. Current Opinion in
Environmental Sustainability, 26, 97-105.
Roemmich, D., Johnson, G.C., Riser, S., Davis, R., Gilson, J.
et. al. 2009. The Argo Program: Observing the global ocean
with profiling floats. Oceanography, 22 (2), 34-43.
Rudnick, D., Costa, D., Johnson, K., Lee, C., Timmermans,
M.L., 2018 (January). ALPS II-Autonomous Platforms and
Sensors. A Report of the ALPS II Workshop. In: ALPS II
Workshop.
Ryabinin, V., Barbière, J., Haugan, P., Kullenberg, G., Smith,
N. et al. 2019. The UN Decade of Ocean Science for Sus-
tainable Development. Frontiers in Marine Science, 6, 470.
Sanchez-Iborra, R., Cano, M.D., 2016. State of the art in LP-WAN
solutions for industrial IoT services. Sensors, 16 (5), 708.
Solomon, J., 1980. Teaching children in the laboratory. Taylor
& Francis.
Tanhua, T., McCurdy, A., Fischer, A., Appeltans, W., Bax, N.
et al., 2019. What we have learned from the framework for
ocean observing: Evolution of the global ocean observing
system. Frontiers in Marine Science, 6, 471.
Thaler, A., Sturdivant, K., 2013. Oceanography for Everyone -
The OpenCTD. Retrieved from http://www.rockethub.com/
267
Medit. Mar. Sci., 22/2 2021, 255-269
projects/26388-oceanography-for-everyone-the-openctd#-
description-tab.
Trevathan, J., Johnstone, R., Chiffings, T., Atkinson, I., Berg-
mann, N. et al., 2012. SEMAT -the next generation of in-
expensive marine environmental monitoring and measure-
ment systems. Sensors, 12(7), 9711-9748.
UNEP, F., 2016. Investment portfolios in a carbon constrained
world: the second annual progress report of the portfolio
decarbonization coalition.
Wang, Z.A., Moustahfid, H.A., Mueller, A., Mowlem, M.C.,
Michel, M. et al., 2019. Advancing observation of ocean
biogeochemistry, biology, and ecosystems with cost-effec-
tive in situ sensing technologies. Frontiers in Marine Sci-
ence, 6, 519.
Wright, N.G., Chan, H.K., 2016. Low-cost Internet of Things
ocean observation. In: OCEANS 2016 MTS/IEEE Monterey
(pp. 1-5). IEEE.
Yang, Q., Yoo, S.J., 2018. Optimal UAV path planning: Sensing
data acquisition over IoT sensor networks using multi-ob-
jective bio-inspired algorithms. IEEE Access, 6, 13671-
13684.
http://www.nexosproject.eu/ (Accessed 11 December 2020)
http://www.schema-ocean.eu/ (Accessed 11 December 2020)
https://www.commonsenseproject.eu/ (Accessed 12 December
2020)
http://www.senseocean.eu/ (Accessed 12 December 2020)
https://pogo-ocean.org/innovation-in-ocean-observing/activ-
ities/openmods-open-access-marine-observation-devices/
(Accessed October 2020)
https://create.arduino.cc/projecthub (Accessed September
2020)
268 Medit. Mar. Sci., 22/2 2021, 255-269
APPENDIX
Summary table for the most popular marine sensors/probes
Many commercial probes for oceanographic measurements can be reported and discriminated on the basis of
price, but sensor accuracy is the most important. SBE 37 MicroCAT is an expensive probe whose cost is justified by
very high-resolution sensors used also for open ocean applications; YSI 600OMS is a moderately expensive probe
equipped with lower quality sensors; Idronaut Ocean Seven has very high sensitivity sensors with a cost compara-
ble to YSI. It is rather difficult to find cheap commercial optical sensors. These usually range between €2,000 and
€10,000: Cyclops-7F Turner Design, ECO FL Wetlabs, SeaPoint, UniLux Chelsea, MicroFlu TriOS (Hodson, 1988;
Piermattei et al., 2019).
Appendix Table 1. Main state-of-art specifications of commercial and cost-effective marine sensors/probes.
Sensor - Probe Measure Accuracy Resolution/
MDL*
PRICE
Category
CTD SENSORS/
PROBES
SBE 37
MicroCAT
Temperature
Conductivity
± 0.002 °C
±0.003 mS/cm
0.0001 °C
0.0001 mS/
cm
€€€€
YSI 600OMS Temperature
Conductivity
± 0.15 °C
± 0.5 mS/cm
0.01 °C
0.001-0.1
mS/cm
€€€€
Idronaut Ocean
Seven CTD
Temperature
Conductivity
± 0.002 °C
± 0.003 mS/cm
0.0002 °C
0.0003 mS/
cm
€€€€
Underway CTD
Teledyne
Oceanscience
Temperature
Conductivity
± 0.004 °C
±0.005 S/m
0.0002 °C
0.0005 S/m
MIDAS CTS
Valeport
Temperature
Conductivity
± 0.01 °C
±0.01 mS/cm
0.002 °C
0.005 mS/cm
MINOS
CTD AML
Oceanographic
Temperature
Conductivity
± 0.005 °C
±0.01 mS/cm
0.001 °C
0.001 mS/cm
CTD350
Greenspan
Analytical
Temperature
Conductivity
± 0.2 °C
±1%
-
CTD Aanderaa
Data Instruments
Temperature
Conductivity
± 0.1 °C
±0.15 mS/cm
0.05 °C
0.075 mS/cm
Castaway-CTD
SonTek-Xylem
Brand
Temperature
Conductivity
± 0.05 °C
±0.25% ±5 S/m
0.01 °C
1 S/m
€€€
Van Essen CTD Temperature
Conductivity
± 0.1 °C
±1%
0.01 °C
±0.1%
€€
OTT CTD Hach
Environmental
Temperature
Conductivity
± 0.1 °C
±1.5%
0.01 °C
0.01 mS/cm
€€€
OPTICAL
SENSORS/
PROBES
Cyclops-7F
Turner Design
Chla Fls
Turbidity
0.03 µg/l
0.05 NTU
€€
ECO FL Wetlabs Chla Fls 0.02 µg/l €€-€€€€
SeaPoint Chla Fls 0.02 µg/l €€€
UniLux Chelsea Chla Fls 0.01 µg/l €€
MicroFlu-chl
Trios
Chla Fls 0.02 µg/l €€
6025 Chlorophyll
Sensor YSI
Chla Fls 0.1 µg/l
C-STAR WET
Labs
Transmittance €€€-€€€€
EXO Turbidity
YSI
Turbidity 0.3 FNU €€€
269
Medit. Mar. Sci., 22/2 2021, 255-269
Sensor - Probe Measure Accuracy Resolution/
MDL*
PRICE
Category
Hydroscat-4
HOBI Labs
4-Wavelength Backsc.
and fluorometer
OCR-500 Series
Irradiance
Satlantic
Wavelenght 400 – 865
nm
€€€€
PRR-2600
Radiometer
Biospherical
Instruments
Irradiance,
Reflectance, PAR
CHEMICAL
SENSORS/
PROBES
2710 OxyGuard
probe
Rickly
Hydrological
Dissolved Oxygen ± 0.5 ppm €
SBE 43 Sea-Bird
Electronics
Dissolved Oxygen 2% of saturation €€€€
Oxygen Optodes
Aanderaa
Instruments
Dissolved Oxygen < 5% 0.4%
WQ401 Global
Water
Dissolved Oxygen ± 0.5% of full scale €€
EXO Optical DO
YSI
Dissolved Oxygen 0.1 mg/l €€
CS511 Campbell
Scientific
Dissolved Oxygen ± 2% €
DO AMT Dissolved Oxygen ± 2% of saturation ± 0.1% of
saturation
N510 Nexsense Dissolved Oxygen ± 0.2 mg/l
WQ-FDO Optical
DO
Global Water
Dissolved Oxygen ± 0.2 ppm ± 0.01% of
saturation
€€
SBE 18 Sea-Bird
Electronics
pH ± 0.1 pH €€€€
WQ201 Global
Water
pH ± 2% €
6589 Fast
Response YSI
pH ± 0.2 pH ± 0.01 pH
Hydrolab HACH
Environment
pH ± 0.2 pH ± 0.01 pH
Deep-Water pH
AMT
pH ± 0.05 pH ± 0.01 pH
COST-EFFECTIVE
SENSORS/PROBES
Cost-Effective
Atlas Scientific Temperature ± 0.15 °C - €
Conductivity ± 2% - €
pH ± 0.002 - €
Smartfin Temperature ± 0.1 °C - €
‘Leeuw’ sensor Chla Fls 0.3 µg/l - €
CTD-SRDL Temperature ± 0.005 °C - €
ArLoc Temperature ± 0.01°C €
Chla Fls 0.01 µg/l €
* Minimum Detectable Limit
