Fertilizers, which are used world-wide since the early 20th century, offer higher yields in agricultural production. Fertilizers are source of nitrogen that plants need to take up. However excess nitrogen poses significant environment risks such as global warming, acidification and eutrophication. Particularly nitrate, which is highly soluble in water, if not taken up by plants, would run off into rivers, lakes, reservoirs and coasts or percolate into ground water. High concentration of nitrogen in surface waters would result in eutrophication which is indicated by over-proliferation of blue-green algae, decrease of light penetration, oxygen deprivation and eventual loss of biological variety. Agriculture is estimated to be responsible from 56% of non-CO2 anthropogenic greenhouse gas (GHG) emission. N2O is a concerned GHG that is sourced by the lost nitrogen as NOx and NH3 due to leaching and runoff from fields. It deteriorates air quality and cause depletion of ozone layer. Excess NH3 can attenuate the buffering capacity of soils and cause acidification. Increased acidity in soil is associated with decreased fertility. One may define the cultivation as the chain of processes with memory property when it comes to the motivation of sustainability. Good nitrogen management is one of the crucial steps to sustain economically feasible agriculture without damaging the environment. The gap between the total nitrogen input and output should be as small as possible to prevent the nitrogen leakage and to keep the fertilizer costs at minimum values. Achieving the success in this challenging task requires knowledge, experience, planning, technology and precise monitoring. Some of the parameters that determine the efficiency of fertilizers are application timing, content of fertilizer, soil pH, soil texture and the phenology of the crop. All those parameters should be assessed together at local domain in real-time. We propose development of an online platform and integrated sensors that will help farmers and wine producers in making decisions for fertilizer application timing, quantity and location by modeling the existing conditions in the relevant farm. The platform will provide recommendations to the farmers as an output by digesting hyper-local weather forecast, real-time soil humidity, soil temperature, soil pH, soil nutrition, air temperature, air humidity data and biological features of the crop. We have been working on a decision making tool for the farmers in Turkey. The beta version of the platform is online under the domain name growity.co (benimhavam.com), which has been used by the Turkish farmers for the last 1 year. Hyper-local weather forecast data for the Turkish territory is provided to the end users. The users benefit from the interpretation data for air temperature, air humidity, precipitation, soil temperature, soil humidity, sunshine duration, precipitation accumulation and the archive data. We are aware of the fact that the human activity on individual farms can change the conditions on the relevant field. Irrigation for example, changes the humidity of the soil, as well as the air. Therefore apart from the precipitation accumulation information, which is obtained by summation of the very near term precipitation forecast data, the humidity levels following the irrigation should be taken into account to make the assessments precisely. To meet this requirement, we propose development of sensor that will continuously measure the soil humidity, temperature, nutrition and acidity together with air temperature and air humidity. The data will be collected and transferred to the cloud system via wireless router units. The variations in the climatic conditions and the on-site measurements will be evaluated together to provide accurate recommendations to the end-users. The project will be conducted in collaboration with our partner wine producer firm, LusoVini, in Portugal. LusoVini is a group of business that is involved in the production line of wine starting from vine grape cultivation till bottling. They sell their products in national as well as in international markets. Their experience in the field and expertise in wine grape cultivation will guide the research team in optimization of fertilizer application. Therefore we initiate developing our models on wine/grape production. We will acquire/process hyper-local weather forecast data for the territories of Portugal. The grape yield records of the company will be studied based on the fertilizer usage and meteorological archive data for the relevant coordinates. Models will be generated by evaluating the correlations. As the significant step of the project, the sensors will be developed that will make on-site, real time measurements. Monitoring and forecasting will run in parallel to achieve precise farming practices, which will eventually reduce operation costs while minimizing the environmental impact.