Increasing salinity of agricultural waterways is a global problem with critical economic and ecological impacts. When cropland is irrigated with moderately saline waters, salts accumulate in the soil leading to reduced crop yields or permanent land fallowing. While this problem is neither new (ancient irrigation systems led to the degradation of the once fertile soils across Mesopotamia), nor declining in scale (soil salinization is now estimated to impact nearly one-third of all irrigated land worldwide), there are few decision tools for evaluating the techno-economic feasibility of intervention which is the focus of this proposal. Developing decision tools is critical to informing farmers, managers and others who are interested in slowing salinization trends and ensuring the sustainability and resilience of global food systems.
Agricultural waterway salinization imposes considerable economic and ecological damages. The proposed research will develop methods for valuing those damages and comparing the cost-effectiveness of technology and policy interventions. The proposed research will also develop novel methods for efficiently generating a broad, multi-resolution dataset of agricultural waterway salinization critical to implementing these decision models. Finally, the proposed research will evaluate the techno-economic feasibility of salinity reduction interventions, including subsidizing land fallowing, limiting tile drain discharge, and desalinating saline agricultural waters, in a multi-objective decision framework. This research will test and implement the proposed methodology in the Panoche Water and Drainage District and the Westlands Water District, salinity-impacted districts within the San Joaquin River Basin of California and very nearby UC Merced. Collecting, processing, and leveraging data high-resolution data to inform environmental decisions remains a methodological challenge in agricultural systems. The PIs propose to develop a comprehensive decision analysis framework that leverages remote and autonomous sensing to rapidly and cost-effectively evaluate salinity management practices for agricultural waterways. Design and integration of multi-modal sensor networks will inform fundamental relationships between water quality, soil salinity, ecological health, and land use in agricultural environments. Incorporating this information into a decision analysis framework will aid agricultural producers, regulators, and state infrastructure managers in evaluating technologies and policies for salinity management across multiple, often competing, objectives. The first objective of the proposed research is to implement a valuation model to assess the benefits of salinity reduction to agricultural and ecological systems. The second objective is to develop a hierarchical, multi-modal data collection methodology using remote sensing, autonomous robotic watercraft sensors, and sparse, distributed, static sensor networks for efficiently and cost-effectively developing models of agricultural water salinization. The third objective is to evaluate the cost-effectiveness of distributed salinity reduction technologies and policies available to agricultural producers and regulators to minimize the economic and ecological impacts associated with salinized agricultural waterways. This will result in novel, hierarchical, low cost methods for generating high-resolution data sets on agricultural waterway salinity that can be extended to the detection and monitoring of other non-point source emissions in the agricultural industry. The proposed research will also develop novel valuation methods for quantifying the benefits of reducing agricultural waterway salinity for private actors (growers) and the public (ecosystems) at a sufficient resolution to inform policy and technology implementation, a major barrier to policy and technology development for combating salinization issues. Finally, the proposed research will develop algorithms for positioning static sensors to maximize the value of information collected. Throughout, the PIs will develop methods to manage uncertainty in the processing of large datasets. The proposed work will enhance the sustainability of irrigated agriculture by providing a quantitative decision framework in which to compare public and private costs and benefits of salinity reduction. It will also promote workforce development in the emerging field of autonomous and robotic sensing for environmental decision-making by engaging UC Merced undergraduate students in data collection, data management, and data visualization research. UC Merced is a minority serving institution in the Central Valley of CA, and students participating for course credit or full time summer employment will be trained for emerging, high tech jobs in the local agricultural industry. Finally, the PIs will continue the development of an education module for high school students that provides hands-on exposure to environmental science, robotics, and data processing via deployment of autonomous robotic watercraft sensors.
