Progetto – Pro.G.Brix 14-18

3.2. Project purpose and problem to be solved through innovative solutions

The high specialization of agricultural companies has a significant environmental impact with negative effects such as loss of soil fertility and pollution, as well as having a significant impact on the climate. Climate issues are combined with those of agro-environmental sustainability and innovation. To date, costs in the area are generally high to obtain quality and safe products, while sales prices vary significantly, and there is a widespread lack of propensity for process innovation, with difficulties in planning cultivation phases. In the province of Latina, production is mainly intended for the fresh market and for processing into fourth-range vegetables. The outlet markets are mainly foreign and mainly aimed at large-scale retail trade and the imbalance of market power to the detriment of agricultural producers is evident, who are subjected to prices and conditions of supply and payments by large-scale retail trade. For growth in the sector, aspects such as the evolution of technologies applied to horticultural production must necessarily be considered, in order to allow for the improvement of production efficiency and market-oriented quality while respecting the environment. Specialized companies manage to obtain better economic results in the medium-long term, but are exposed to frequent market crises, linked both to external variables (international context, economic cycles of consumption, etc.) and to local problems (climate risks, phytosanitary problems, excess production). The development and adoption in the Agro Pontino of a cultivation protocol that allows the production of watermelon with a guaranteed Brix level of 14-18 is essential for the protection of the climate and soil and its sustainable use, a priority of the agro-climatic-environmental policy of rural development. It would also allow for greater efficiency in the distribution of water, fertilizers and pesticides, since the cultivation needs during the entire vegetative cycle of the crop would be better respected, and environmental risks, especially those related to excess inputs, would be reduced to a minimum, and it would allow the reduction of production costs to a minimum. Related to what has just been considered, among the various characteristics, the Brix level not only defines the harvesting times, but above all decides the unit price of the product and the level of profitability of the plants. The adoption of the protocol in the Agro Pontino would allow companies to have greater respect for all the environmental issues mentioned above, offering a higher quality and healthier product, with an added value, much more competitive on the market, since fully satisfying the quality parameters required by the market means increasing the value of production.

3.3. Proposed innovation

The innovation that we intend to introduce in the Lazio Region consists in the creation of a cultivation protocol that allows, through management based on the use of innovative instruments, to obtain a specific product: watermelon with a guaranteed Brix level of 14-18.

This protocol will be managed starting from the field survey, using digital spectroscopy techniques (operating in the wavelength range between 400 and 2500 nm), and/or smartphones and will allow, whatever the initial condition, to obtain watermelons with a guaranteed brix degree of 14-18. This will allow to improve the competitiveness of primary producers by integrating them better into the agri-food chain, creating an added value for the product..

This protocol will also cover all post-harvest activities up to marketing. In fact, the brix level may undergo variations in the phases following harvest.

This is fully part of what is called precision agriculture, an agricultural management strategy that uses modern instruments and is aimed at carrying out agronomic interventions taking into account the actual crop needs and the biochemical and physical characteristics of the soil, a practice well studied and applied also in the horticultural sector. The process will have to start from an analysis of the initial conditions of the soil, and will have to develop a management system through which to reach the pre-established sugar level, furthermore the fully automated system will allow to control and possibly modify the process in the event that at a certain point in the process it becomes clear that the results will deviate from those expected. All the data will be available in real time through a smartphone, through which all the cultivation practices carried out will be automatically recorded (i.e. the field notebook) in addition to updating the loading and unloading registers of pesticides and fertilizers.

Innovative monitoring techniques will be used with reference to the study and evaluation of changes recorded during cultivation and cultivation practices. This is thanks to the development, fine-tuning, implementation and verification of techniques based on spectrophotometry operating in the VIS (400-1000 nm), NIR (1000-1700 nm) and SWIR (1000-2500 nm), capable of carrying out rapid, simple and efficient monitoring in relation to the different cultivation practices. These techniques, robust and reliable like conventional ones, will allow for a rapid and low-cost systematic analysis and survey of the characteristics and growth status of the fruit, together with the monitoring of plants and soils. Significant advantages are obtained thanks to the possibility of conducting a high number of analyses directly in the field in a short time, obtaining useful information about the characteristics and state of the plants, of primary importance for correct cultivation and to guarantee the desired sugar level, with a view to improving and benefiting in socio-economic and energy-environmental terms.

We will use methodologies specific to chemometric analyses and recognition and classification algorithms derived from this type of investigation. To achieve this objective, from the software point of view, we will use the libraries contained in the PLS_Toolbox (Eigenvector Research, Inc.) operating in a MATLAB environment (The Mathworks, Inc.). Specific procedures developed autonomously and originally by the research group will therefore be developed, implemented, fine-tuned and validated. The investigations will be developed through the acquisition, archiving, pre-processing and analysis of hyperspectral data (i.e. digital spectra relating to both the plant's leaf apparatus and the product (i.e. watermelon), acquired in the 400-2500 nm range at different spectral resolutions), aimed both at fine-tuning classification/recognition models of the characteristics of the different products, and at building correlation models between this information and information obtained for the same products through traditional analysis.

In the context of highly specialized horticulture, organic farming is attracting more and more interest. The use of the protocol developed can also be applied in the field of organic production. The production obtained can be certified and guaranteed.

3.3a Coordination and interaction methods between partners

Coordination and interaction between partners will be managed through periodic meetings and computerized project management on a common IT platform.

3.4. Specific objectives

The main objective is to increase farm income through the production of watermelon with a guaranteed Brix level of 14-18, producing a low environmental impact and through simplified management of cultivation practices thanks to the combined use of hyperspectral digital spectrophotometry systems operating in the VIS (400-1000nm), NIR (1000-1700 nm) and/or SWIR (1000-2500 nm), capable of carrying out rapid, simple and efficient monitoring of cultural progression in relation to different cultivation practices and smartphones. This will improve the competitiveness of primary producers by better integrating them into the agri-food chain through quality schemes, the creation of added value for agricultural products, the promotion of products in local markets, short supply chains, producer associations and organizations and inter-professional organizations.

The use of innovative non-destructive and non-invasive monitoring techniques aimed at preparing a watermelon cultivation protocol with guaranteed sugar content will allow participating companies to significantly improve their production processes by optimizing them with a more efficient organization, thanks to the transition from a traditional system to precision agriculture, through digitalization of production control processes, consequently obtaining a superior product in quality, with better results on the market (increase in price, increase in sales) and on the health of the consumer and the environment, also in a system consistent with organic agriculture.

The proposal, based on the previously defined operational scenarios, would pave the way for the application of effective and efficient non-invasive and non-destructive monitoring, control and traceability techniques, based on easy monitoring and early detection, in relation to watermelon cultivation. Such an approach will also produce a lower economic impact of the investigations, also allowing for an increase in their number to the advantage of a more extensive and systematic control of the product.

The approach based on hyperspectral spectroscopy techniques will allow, from a technical-scientific point of view, the achievement of the following objectives: i) the construction of hyperspectral libraries characteristic of the different state of crops (i.e. plants, watermelons and soils) in relation to their characteristics, ii) the development of chemometric correlation models between information obtained through hyperspectral techniques and conventional ones, iii) the availability of an integrated sensor/algorithmic approach of "easy use" such that it can be used not only at laboratory scale, but also directly in the field for "fast point analyses" and/or at production scale, maintaining the level of reliability unchanged, in terms of recognition, characterization and traceability of the different products, in combination and/or in substitution with that obtainable at laboratory scale through conventional procedures.

3.5. Description of the project in actions/activities and products

The project will be able to take place thanks to the formation of a partnership. This will be composed of agricultural and marketing companies belonging to the fruit and vegetable sector of the Lazio region and in particular of the Agro Pontino area whose objective will be to obtain a specific product: watermelon with a guaranteed Brix level of 14-18. A public research body, the Research and Services Center for Sustainable Technological Innovation (Ce.R.S.I.Te.S) of the University of Rome - La Sapienza and two private research bodies will also be involved: Eurolab S.r.l. and the European Center for Managerial Studies. Other professional figures will be involved for this project.

The research activities will be carried out through:

the use of innovative non-destructive and non-invasive monitoring techniques aimed at preparing cultivation protocols for a watermelon fruit with guaranteed sugar content. These techniques will be used with reference to the study and evaluation of the various changes recorded during cultivation and the various cultivation practices, leading to a possible increase in production through a system consistent with organic farming;
the development, fine-tuning, implementation and verification of techniques, based on spectrophotometry techniques operating in the VIS (400-1000nm), NIR (1000-1700 nm) and SWIR (1000-2500 nm), capable of carrying out rapid, simple and efficient monitoring of cultural progression in relation to different cultivation practices.

In order to introduce the concepts and operating methods that are the basis of the project, the companies involved in the partnership will hold operational meetings aimed at the theoretical-practical illustration of the detection and diagnostic methods, based on digital/hyperspectral punctual and/or image spectroscopy techniques, for the evaluation of the characteristics of the different products in relation to all points of the supply chain and the corresponding processing/production phases.

Demonstrative applications will be developed during the different cultivation phases, in order to demonstrate the effectiveness of the proposed hyperspectral spectroscopy techniques.

Almost continuous support and assistance from specialized personnel is provided for all companies and for all phases of project development.

Partners will be in contact through the Feedentity data exchange platform. Coordination meetings will be organized on a monthly basis.

The project will be implemented in the following phases:

1st phase – Choice of plots where to set up the experiment and analyse the characteristics of the soil and the water used;

2nd phase – Preparation of a cultivation management system;

3rd phase – Application of what has been prepared with relative monitoring through spectrophotometry systems operating in the VIS (400-1000nm), NIR (1000-1700 nm) and SWIR (1000-2500 nm), to be used directly in the field and in connection with smartphones and to the Feedentity platform, which examines the sugar content during the ripening phases);

4th phase – Study of changes in sugar concentration;

5th phase – Results and discussion;

6th phase – Conclusion and preparation of the protocol.

The instrumental approach and analytical methodologies will be used and developed in relation to the definition of different operational PHASES. These PHASES in terms of content and actions are reported below.

PHASE 1 - Choice of plots where to set up the experiment and analyse the characteristics of the terrain and of the water used

During this first phase, different areas (i.e. plots of land) will be studied in which to carry out the experimentation. The characteristics of the soil and water that will be used for irrigation purposes will also be analyzed.

PHASE 2 – Setting up a system for managing cultivation

Preparation of a cultivation management system consisting of:

one or more control units for the management of specific monitoring systems installed in the field for the purpose of recording the values of environmental parameters to be related to the crop cycle,
preparation of a fertilization plan based on the crop,
preparation of cultivation practices.

PHASE 3 – Monitoring through spectrophotometry systems

During PHASE 3, spectrophotometry systems operating in the VIS (400-1000nm), NIR (1000-1700 nm) and/or SWIR (1000-2500 nm) range will be developed, to be used directly in the field and in connection with smartphones and the Feedentity platform, aimed at preparing a methodological approach for monitoring the sugar content during the watermelon ripening phases.

In this phase, two different spectrometers will be tested: the portable VIS-SWIR spectrophotoradiometer

FieldsSpec® 4 (350-2500 nm) and the portable spectrophotoradiometer USB NIR MicroNIR™ (950-1650 nm).

Reflectance data will be acquired by spectrometer with two sample presentation modes (i.e. watermelon):

acquisition of whole watermelon – in contact with the peel;
acquisition of cut watermelon – in contact with the pulp.

For each sample analyzed, the sugar content values (Brix degree) will be obtained using traditional techniques (i.e. refractometric method) in support of the spectroscopic analyses. These analyses will be carried out following the ripening phases of the watermelon. The sample size, in this phase, will be of fundamental importance for the construction of a robust regression model, suitable for the correlation between the Brix data obtained using traditional methods and the reflectance data.

PHASE 4 – Study of changes in sugar concentration

In this phase, the results of the investigations conducted in PHASE 3 will be evaluated. On the basis of the values detected, strictly linked to the fertilization plans and cultural practices previously defined, changes will be made, where and if necessary, in order to maximize the quality of the product.

During this phase, for each sample analyzed, the sugar content values (Brix degree) will be obtained using traditional techniques (i.e. refractometric method). These analyses will be carried out following the ripening phases of the watermelon.

In addition to the spectroscopic analyses, the relative sugar content data will be made available. The partial least squares (PLS) regression technique will be used to evaluate the correlation between the acquired spectra and the ripening state of the analyzed watermelons. PLS models will be developed for each data set (Geladi et al., 1986; Wold et al., 2001; Abdi, 2003) using the values of TSSC (°Bx) as Y.

If the results of the regression analyses are acceptable, it will be possible to proceed with the construction of a classification model capable of discriminating, on the basis of the reflectance spectra associated with the Brix degree of the analysed watermelon, the following classes:

watermelon with Brix degree < 14
watermelon with Brix degree = 14-18
watermelon with Brix degree > 18

or, alternatively, a two-class classifier:

watermelon with Brix degree = 14-18
watermelon with Brix degree 18

Measurement of the content of dissolved solid substances

Among the soluble solids, sugars are those found in the highest concentrations within the pulp of a fruit. Along with sugars, in the pulp of a fruit, there are also other soluble substances (i.e. pectin and amino acids). The content of dissolved solids, or TSSC, is traditionally evaluated by refractometer and reported in degrees Brix (Garner et al., 2015). A classic portable refractometer allows you to measure the refractive index of a liquid. In the case of fruit pulp, the refractometer allows you to determine the critical angle of refraction (ChemBuddy, 2011). The light that passes through the liquid placed on the surface of the refractometer prism is "bent" and that is, refracted. The refracted light is then focused on a graduated scale, magnified by a lens in such a way as to be clearly visible for reading purposes (MISCO, 2013; A.KRÜSS Optronic, 2013). In this way, the sugar content of the kiwi being analyzed can be measured; i.e. assuming that its pulp has the same refractive index as 1 g of sucrose in 100 g of aqueous solution, then we will have a TSSC equal to 1° Bx (Bionutrient Food Association, 2017).

PHASE 5 – Results and discussion

The results will be analyzed in order to evaluate the results of the investigations conducted through spectrometry both with reference to the evaluation of the Brix degree, but also and above all in relation to the methods of use of the method and the procedures for processing the information acquired through spectrophotometry.

The two main parameters that will be used to evaluate the goodness of the VIS-SWIR spectra and Brix degree regression models are: Root Mean Square Error (RMSE) and 𝑅 (Martens et al., 1989; Naes, 2002). The RMSE parameter provides the possibility to evaluate the differences between the estimated response 𝒀 ̑ and the observed response 𝒀. While, R2 is a parameter used to evaluate the model fitting. These parameters will be referred to the calibration, cross-validation and prediction phases. While, to evaluate the importance of the contribution of the descriptive variables to the final model, both the variables significant to the description of Y must be taken into account (and the variables relevant to the description of X, summarized by parameters such as: regression vector, selectivity ratio and VIP scores.

The classification performances will instead be evaluated on the basis of the following parameters: Sensitivity, Specificity, Precision, Misclassification Error and Accuracy, referring to the calibration, cross-validation and test set prediction phase.

The classification performances will also be finally evaluated as a function of the instrumental devices, VIS-SWIR FieldsSpec® 4 and USB NIR MicroNIR™, used to perform the spectral acquisitions. In fact, these instruments perform acquisitions in different wavelength ranges, respectively 350-2500 nm and 950-1650 nm, and are characterized by very different costs and hardware configurations, the comparison of which is of great interest in terms of application strategies.

PHASE 6 – Conclusion and preparation of the protocol

In this phase, the conclusions reached by the study will be reported and an operational protocol will be prepared for monitoring the crops in order to be able to develop watermelon production with a guaranteed Brix level of 14-18.