Field scale assessment of subsurface drainage systems (SSDS) for efficient crop water management with improved crop productivity for agriculture in the northern Finland.

Precision agriculture is essential to optimize the crop water use even in the environmental conditions with adequate water availability. It ensures the optimal water retention and reduced nutrient leaching for agricultural field. By doing so, precision agriculture also confirms tackling the situations like changing climate with varied seasonal weather patterns, availability of adequate water resources for crop growth and altering surface water quality due to runoff from agriculture. 
Although, the agriculture in the Nordics is not under acute pressure of water scarcity however the situations of unprecedented weather conditions during the crop growing season requires necessary attention. Moreover, the subsequent effects of long-term changing climate predictions on the surface as well as groundwater availability in the region are concerning. Under such circumstances the conventional use of SSDS implementing controlled drainage (CD) approach with its impact on field-scale hydrology with a shallow groundwater state were assessed in recent studies. However, neither its subsequent effects on the field-scale crop productivity nor its integration with possible strategies to optimize crop water use under long-term predicted state of subsurface hydrology in the region were investigated.
This study presents the long-term climate assessment for agriculture in the Northern Finland with its impact on the seasonal crop yield. Also, with the use of process-based model approach the study attempts to present a possible eco-friendly strategy with necessary updates in the existing SSDS to optimize the crop water use under the adverse conditions of long-term forecasted state of subsurface hydrology in the region. In addition, the study presents the field scale crop productivity by ensuring the enhanced water retention and reduced nutrients from agriculture with precision in farm management practices to sustain or improve the crop productivity. The simulation results with crop water productivity tool over historical dataset showed over 40 to 60 percent rise in the seasonal crop yield under adverse climate conditions. Whereas the results for overall amount of soil water required to replenish the crop water need showed a difference of almost 0.016994 MCM per ha in case of effective integration of SSDS with more efficient water application systems for agriculture. 
Moreover, this work introduces Data learning approach which talks about Integration of multi-source data (DI) and Machine Learning (ML) approach to real-world data. We present a broader perspective followed while developing applications based on Data Learning approach. We present a data learning method and results for a case study field which involves process based as well as machine learning approach.