SDSS for agricultural applications is quite complex; it requires knowledge from various multidisciplinary areas, such as crop agronomy, computer hardware and software, mathematics, and statistics. For example, to understand crop growth, it is necessary to know how many variables affect crop growth, how each variable affects crop growth? Each crop requires a different optimum value for growth . DSS model was developed to support decision-making in West Africa to help local managers assess the Water Energy Food Environment, furthermore, proposing ideal administration answers for add to expanding the creation of food crops at the territorial level in the transboundary Mékrou stream bowl in the waterway bowl between Benin, Burkina Faso, and Niger. They supported crop productivity at the regional level for intensifying agricultural practices, enhancing inputs such as nutrient fertilizers and water irrigation, etc. . The authors showed that this DSS could coordinate a few models, for example, a biophysical agricultural model, a simplified regression metamodel, linking crop production with external inputs, a linear programming and a multi-objective genetic algorithm optimization routines for finding efficient agricultural strategies; and a user-friendly interface for input/output analysis and visualization Fig. 13. NARSS team succeed in developing software to enhance farm management under Egyptian conditions based on IoT technologies; this software is semi-automated in the first version as it can receive the data from two sources from exile sheet or based on IoT. The software can achieve numerous functions such as farm management zone , spatial mapping analysis of soil characteristics, and agricultural-dedicated solutions. The smart management program should be flexible in modifying modeling equations, parameter values, adding and removing elements, and it should contain alternatives solutions that address a wide range of different phenomena in the field. IoT technology and web services are used to improve the automation of data collection with higher Spatio-temporal accuracy to facilitate and improve the decision-making process. The software provides some services to the users,raspberry cultivation pot including an absolute mapping service, real time visualization of data, spatial analysis using virtual coordinates.

The spatial interpolation and mapping could be achieved using Inverse distance weighting based on the data collected from nodes locating in the field. The software can also illustrate a contours map that showed the distribution range of each phenomenon and also can be generated grid values dividing the field into zones of predefined ranges of the selected parameter; each range indicates a specific meaning or requires a different management criterion. When the quality index parameter is selected to be analyzed, also a legend indicates the quality classes of different land zones appears. The real-time data is used together with the stored laboratory data to analyze and evaluate the properties of soil, yield, and other factors regarding farm management. Thus the ranges that have the same symmetric range are thus identified and require spatial management . The gauges visualization is used to visualize the phenomena used for soil parameters analysis for different nodes and show the analysis result in real-time.Quite alarmingly, carbon storage of agricultural soil has been reported to decrease by 31% during the first decade after conversion into farmland due to land-use change and to further diminish to less than 50% in half a century, accompanied by losses in crop yield. A similar trend was detected from Finnish cultivated fields where the average C loss was 17% over a 35-year cultivation period, and the authors suggested that the change in management practices in recent decades towards increasing cultivation of annual crops, as well as climate change, has contributed to soil C losses in boreal cultivated soils. Intensive soil cultivation causes leaching of essential nutrients and physical damage to soil structures, eventually negatively changing ecosystem services provided by the soil microbiome. Microbiome soil ecosystem services include decomposition, formation of soil aggregates, cycling of nitrogen , aid in nutrient and water uptake by plants, pathogen control, mitigation of greenhouse gases and C sequestration into soil as microbial bio- and necromass. For instance, mould board ploughing, which is used as a common tillage practice in conventional intensive cultivation has been reported to cause many physical, chemical and biological changes in soil including reduced abundance and diversity of soil organisms.According to a large meta-analysis study, microbial biomass C and N were on average 41% and 51% higher in organic systems, respectively, as well as having 59% increased total phospholipid fatty acids and 32–84% increased enzyme activities, compared to conventional systems. According to another meta-analysis based on an extensive literature review, organic farming generally also had positive impacts on the biodiversity of soil microbiota.

Several studies have reported on the increase of microbial diversity or changes in community composition due to organic farming. In addition, organic farming has been reported to lower soil-derived plant diseases and increase disease suppressiveness. Organic fertilizers are reported to induce changes in soil microbial communities that would promote soil health. According to EC-regulation, organic crop rotations are not allowed to have cereals more than three consecutive years, and more than 30% of crop rotation plants must be legumes. Furthermore, synthetic fertilizers and pesticides are prohibited in organic farming. However, management practices used in organic farming e.g., organic fertilizers and more diverse crop rotations are utilized more and more in conventional farming approaching the idea of an integrated farming system. Furthermore, land use intensification differently affects bacteria, archaea and fungi. Microbial communities respond differently to management practices such as tillage intensity. Symbiotic arbuscular mycorrhizal fungi colonizing many crop plant roots particularly benefit from management practices used commonly in organic farming, such as diverse crop rotation and reduced tillage. The microbial community composition is important since the soil fungal diversity is suggested to be an intrinsic factor in the health of managed soils. A broad and comprehensive scientific understanding on the long term impacts of organic farming on soil processes and microbial communities is currently incomplete. Indeed, we are not aware of any previous studies that have been conducted in arable soils of the boreal region and in parallel for forage and cereal crop rotation. In this study, the micro-biome was investigated both in spring and autumn sampling campaigns, since temporal factors are known to strongly affect microbial communities, especially in agricultural soils. We used basal respiration as a measure of microbial activity, quantitative PCR for estimating microbial abundance, fumigation extraction to estimate microbial biomass derived C and N, and target gene region amplicon sequencing to reveal microbial community composition. Our aim was to detect whether long-term organic and conventional farming have induced changes in microbial activity, biomass, richness, as well as community composition according to their respective farming practices, and we expected that the direction and magnitude of putative changes are dependent on crop type , season, and the specific microbial group investigated. The sampled experimental field site is located on sandy soil in Toholampi; Ostrobothnia, western Finland . The site was constructed for erosion and nutrient leaching studies as described by Turtola and Kemppainen in 1998. Four different crop rotations in total were established at the field in 2001 to compare conventional and organic farming systems for hypothetical cereal and milk production farms with four-year crop rotations . The main focus of this study was the effect of different nitrogen sources and fertilization intensities on nitrogen leaching and crop yields.

The experiment was designed as a randomized block design consisting of 16 plots including four replicates for each crop rotation. Organic crop rotations and farming practices were designed to meet the EU requirements for organic farming in terms of crop rotation, fertilization and plant protection. Fertilization was based on the biological nitrogen fixation of legumes and use of cattle manure applied as a slurry. Plant protection was executed proactively through crop rotation and tillage. Organic crop rotation producing cereals was planned to cooperate with a dairy farm which provided manure in return for silage. During the entire 18-year period the OCer rotation received on average 50 kg ha-1 a-1 total nitrogen in manure, applied in the first and last year of the four-year crop rotation . Organic crop rotation of the milk production farm cultivating forage crops was planned to be self-sufficient in fodder and manure. The crop rotation produced fodder to feed the dairy cows. The manure produced by the cows was used as organic fertilizer. Average manure application rate during 2001–2018 was 85 kg Tot-N ha-1 a-1 . Since 2005 manure has been applied annually. Conventional cereal crop rotation was fertilized with synthetic fertilizers according to the limits of the Agri-Environmental Program in Finland and was approximately 86 kg Tot-N ha-1 a-1 . Conventional forage crop rotation of a dairy farm was fertilized with a manure application rate of 110 kg Tot-N ha-1 a-1 . Annual fertilization with applied manure was complemented with synthetic fertilizers according to the limits of AEP and was about 61 Tot-N kg ha-1 a-1. Fertilization practices during the experimental years 2001–2018 are presented in detail in Supplementary Table S1. In addition,low round pots field plots under the conventional systems received plant protection agents for weed control presented in Supplementary Table S2. Thus, the sampling year 2018 was the second year of the 4-year crop rotation and cultivated plants in rotations were timothy and clover ley for OCer and OMilk, barley for CCer and timothy and meadow fescue ley for CMilk. However, in 2017, the year preceding spring sampling in 2018, barley was cultivated as a main crop in all four rotations. Soil DNA was extracted with a NucleoSpin Soil kit according to the protocol of the manufacturer from homogenised soil samples from which all roots were removed. DNA concentration and purity were determined with a NanoDrop Lite spectrophotometer.

DNA samples were sequenced at the Institute of Genomics of Tartu University, Estonia. AMF fungi were sequenced from four composite DNA samples obtained from autumn samples with AMF targeting the 18S rRNA gene using primers AML2 and universal eukaryotic primer WANDA. For bacteria the targeted V4 region of the 16S SSU rRNA and for fungi the ITS2 region were amplified in a two-step polymerase chain reaction . Bacterial and fungal PCR were performed using the 16S rRNA primers 515F and 806R and the ITS primers ITS4 and gITS7, respectively, with 8bp dual index for 24 cycles. The final PCR fragments were run as paired-end 2 × 300 bp with the MiSeq platform using MiSeq v3 kit producing ca. 20–25 M reads per flow cell. Quantitative PCR for the partial fungal ITS region, and bacterial and archaeal 16S rRNA genes, was conducted as described by Peltoniemi et al., in 2015, for all samples separately. After the first quality filtering steps, raw bacterial 16S rRNA sequence data consisted of 915 740 reads clustering into 13 824 OTUs, fungal ITS2 data consisted of 663 744 reads clustering into 2800 OTUs, and raw AMF 18S rRNA sequence data consisted of 157 011 reads clustering into 1328 OTUs. Second quality filtering was done as described by Soinne et al., in 2020. OTUs that had affiliations other than bacteria or fungi, as well as singleton OTUs and reads with relative proportion below 0.01% were removed from the data. Furthermore, bacterial OTUs were consolidated according to accession numbers in the Silva database, fungal OTUs to the exact same species hypothesis in UNITE, and AMF OTUs according to genbank accession numbers in MaarjAM database. Only 25% of the 18S rRNA sequences had >90% identity to their reference. The final bacterial data consisted of 472 119 and 347 709 reads for spring and autumn samples, respectively, clustering into 2110 OTUs. For fungi, the spring and autumn data consisted of 268 238 and 272 290 reads, respectively, and clustered into 1190 OTUs. The final AMF data consisted of 38 940 reads for four bulked autumn samples, clustering into 39 OTUs. Raw sequence data is deposited to the sequence read archive of NCBI/EMBL database under the BioProject id PRJNA637213 with the accession numbers SAMN15098534-15098565 for bacterial 16S rRNA and fungal ITS2 data, and SAMN15098881- SAMN15098884 for AMF 18S rRNA data. All statistical analyses were conducted in R studio version 1.2.5001 and R version 3.6.0 or 3.6.1. Differences in means of basal respiration rates, soil pH, total soil C and N, microbial biomass C and N, extractable C and N, 16S rRNA gene and ITS-region copy numbers, and bacterial and fungal OTU numbers , between the farming systems , and for the spring and autumn data separately, were investigated with function lmer producing a linear mixed model that takes into account the impact of the block design.