Collaborative Geodesign and Spatial Optimization for Fragmentation-Free Land Allocation

Yiqun Xie, Bryan C. Runck, Shashi Shekhar, Len Kne, David J. Mulla, Nicolas Jordan, Peter Wiringa

TL;DR

This paper tackles the complex challenge of allocating land for agricultural production to maximize both food output and ecosystem services, like water quality, while crucially avoiding land fragmentation. They propose three frameworks—collaborative geodesign, spatial optimization, and a hybrid approach—to help stakeholders navigate these trade-offs and create practical, efficient land-use plans.

What Problem Does This Paper Address?

As the global population grows, so does the demand for agricultural food production. This puts immense pressure on our environment, particularly on water resources, due to practices like soil erosion and fertilizer runoff. Effectively managing agricultural land is key to balancing food production with environmental sustainability. A major hurdle in this is "land fragmentation"—creating small, irregularly shaped plots of land for different uses. This fragmentation makes it difficult to use modern, large-scale farming equipment, significantly reducing efficiency and increasing costs. The problem is further complicated by the need to consider diverse stakeholder interests (farmers, environmentalists, regulators) who often have conflicting priorities. Existing approaches struggle to simultaneously address the spatial partitioning of land, the choice of land management practices, and the critical constraint of avoiding fragmentation, all while incorporating stakeholder input.

Methods

The authors propose three frameworks to address the Fragmentation-Free Land Allocation (FF-LA) problem:

  • Collaborative Geodesign: This is a stakeholder-driven approach using a technology-supported system. It facilitates collaboration among diverse stakeholders (farmers, conservationists, etc.) to collectively design land allocation plans. The system allows users to create, visualize, and evaluate different land-use scenarios, fostering a social learning process to resolve conflicting objectives.
  • Spatial Optimization: This is a computer-driven approach that uses integer programming to find optimal land allocation solutions. It explicitly models and enforces hard spatial constraints, such as minimum patch area and regular shapes (e.g., rectangles), to guarantee fragmentation-free results. They introduce a dynamic-growth tiling framework (DGTF) to heuristically search for solutions in large-scale watersheds.
  • Hybrid Framework: This approach integrates the strengths of both collaborative geodesign and spatial optimization. It allows stakeholders to intervene in the optimization process, guiding the algorithm by fixing certain land uses or focusing optimization on specific sub-regions. This aims to leverage the quantitative power of optimization with the qualitative insights and practical constraints that humans bring to decision-making.

Datasets

The study uses a case study of the Seven Mile Creek watershed in southwestern Minnesota. The watershed was discretized into a 30m x 30m grid. Quantitative coefficients for water quality and economic costs for different land management practices (LMPs) were computed using the Soil and Water Assessment Tool (SWAT) and the InVEST model, which were calibrated by domain experts.

Key Insights and Ideas

  • Fragmentation is a critical practical constraint: The paper strongly emphasizes that avoiding land fragmentation isn't just an aesthetic concern; it directly impacts the economic viability of farming by hindering the use of modern equipment.
  • Stakeholder input is essential but challenging: While collaborative geodesign can engage diverse groups, achieving consensus on conflicting objectives (e.g., water quality vs. market return) is difficult and can lead to less-than-optimal solutions from an efficiency standpoint.
  • Spatial optimization offers powerful solutions but needs context: Purely algorithmic optimization can generate highly efficient, fragmentation-free plans that significantly outperform stakeholder-driven designs in terms of environmental benefits (like sediment reduction). However, these solutions may not fully capture nuanced human preferences or local context.
  • A hybrid approach is the most promising: The paper argues that the most effective path forward lies in integrating human-driven design with algorithmic optimization, allowing stakeholders to guide and refine computationally generated solutions.

Key Results

In the Seven Mile Creek watershed case study:

  • The stakeholder-driven collaborative geodesign resulted in a land allocation plan that reduced sediment by 182 tons/year compared to the baseline, but the plan showed significant land fragmentation.
  • Spatial optimization without spatial constraints (treating each grid cell independently) led to severe land fragmentation, making it impractical for farming.
  • Spatial optimization with hard spatial constraints (minimum area and width) successfully generated fragmentation-free land allocation maps. These optimized solutions achieved a significantly lower amount of sediment (ranging from 1508 to 1664 tons/year) compared to the stakeholder design, while maintaining at least the same economic return. The degree of sediment reduction was inversely related to the strictness of the spatial constraints (tighter constraints led to higher sediment).

Limitations and Future Work

Limitations:

  • The paper acknowledges the difficulty of social learning in collaborative geodesign when stakeholders face conflicting objectives.
  • While spatial optimization provides efficient solutions, it struggles to incorporate qualitative information about local cultural contexts and practical constraints that are hard to quantify.
  • The current spatial constraints are primarily rectangular, and the paper notes that real-world landscapes often have irregular boundaries.

Future Work:

  • Further study and implementation of the hybrid framework, integrating spatial optimization into collaborative geodesign for a more complete system.
  • Development of new interactive algorithms for the hybrid framework.
  • Extending the shape constraints beyond rectangles to accommodate a variety of regular shapes.

Summary updated

Abstract

Demand for agricultural food production is projected to increase dramatically in the coming decades, putting at risk our clean water supply and prospects for sustainable development. Fragmentation-free land allocation (FF-LA) aims to improve returns on ecosystem services by determining both space partitioning of a study area and choice of land-use/land-cover management practice (LMP) for each partition under a budget constraint. In the context of large-scale industrialized food production, fragmentation (e.g., tiny LMP patches) discourages the use of modern farm equipment (e.g., 10- to 20-m-wide combine harvesters) and must be avoided in the allocation. FF-LA is a computationally challenging NP-hard problem. We introduce three frameworks for land allocation planning, namely collaborative geodesign, spatial optimization and a hybrid model of the two, to help stakeholders resolve the dilemma between increasing food production capacity and improving water quality. A detailed case study is carried out at the Seven Mile Creek watershed in the midwestern US. The results show the challenges of generating near-optimal solutions through collaborative geodesign, and the potential benefits of spatial optimization in assisting the decision-making process.
BibTeX
@article{xie2017collaborative,
  author = {Yiqun Xie and Bryan C. Runck and Shashi Shekhar and Len Kne and David J. Mulla and Nicolas Jordan and Peter Wiringa},
  title = {{Collaborative Geodesign and Spatial Optimization for Fragmentation-Free Land Allocation}},
  journal = {ISPRS Int. J. Geo Inf.},
  year = {2017},
  doi = {10.3390/ijgi6070226},
}
Figure