# Modelling Spatial Compositional Data: Reconstructions of past land cover and uncertainties

Source: [Modelling Spatial Compositional Data: Reconstructions of past land cover and uncertainties](<https://www.sota2.com/research/paper/68fa7d851e209131cea9370c>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T06:23:26.944Z

Authors: Behnaz Pirzamanbein, Johan Lindstr\\"om, Anneli Poska, Marie-Jos\\'e Gaillard (5) \_\_INSTITUTION\_4\_\_ Centre for Mathematical Sciences, Lund University, Sweden, (2) Centre for Environmental, Climate Research, Lund University, Sweden, (3) Department of Physical Geography, Ecosystems Analysis, Lund University, Sweden, (4) Institute of Geology, Tallinn University of Technology, Estonia, (5) Department of Biology, Environmental Sciences, Linnaeus University, Sweden)

Published: 2015-11-19T22:09:15.000Z

## Abstract

In this paper, we construct a hierarchical model for spatial compositional data, which is used to reconstruct past land-cover compositions (in terms of coniferous forest, broadleaved forest, and unforested/open land) for five time periods during the past $6\\,000$ years over Europe. The model consists of a Gaussian Markov Random Field (GMRF) with Dirichlet observations. A block updated Markov chain Monte Carlo (MCMC), including an adaptive Metropolis adjusted Langevin step, is used to estimate model parameters. The sparse precision matrix in the GMRF provides computational advantages leading to a fast MCMC algorithm. Reconstructions are obtained by combining pollen-based estimates of vegetation cover at a limited number of locations with scenarios of past deforestation and output from a dynamic vegetation model. To evaluate uncertainties in the predictions a novel way of constructing joint confidence regions for the entire composition at each prediction location is proposed. The hierarchical model's ability to reconstruct past land cover is evaluated through cross validation for all time periods, and by comparing reconstructions for the recent past to a present day European forest map. The evaluation results are promising and the model is able to capture known structures in past land-cover compositions.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Task Planning and Allocation on Human-Robot Task Planning and Allocation (TPA) (test)](<https://www.sota2.com/research/sota/task-planning-and-allocation-on-human-robot-task-planning-and-allocation-tpa-test.md>): Task Planning and Allocation; Human-Robot Task Planning and Allocation (TPA) (test); H1,R1 Time/Cost: 1013.93.

- [Human-robot task planning and allocation on Production Task unseen product quantities Num6](<https://www.sota2.com/research/sota/human-robot-task-planning-and-allocation-on-production-task-unseen-product-quantities-num6.md>): human-robot task planning and allocation; Production Task unseen product quantities Num6; Makespan: 1032.91.

- [Human-robot task planning and allocation on Production Task unseen product quantities Num9](<https://www.sota2.com/research/sota/human-robot-task-planning-and-allocation-on-production-task-unseen-product-quantities-num9.md>): human-robot task planning and allocation; Production Task unseen product quantities Num9; Makespan: 1555.34.

- [Human-robot task planning and allocation on Production Task Num1 (unseen product quantities)](<https://www.sota2.com/research/sota/human-robot-task-planning-and-allocation-on-production-task-num1-unseen-product-quantities.md>): human-robot task planning and allocation; Production Task Num1 (unseen product quantities); Makespan: 1336.32.

- [Human-robot task planning and allocation on Production Task Num5 (unseen product quantities)](<https://www.sota2.com/research/sota/human-robot-task-planning-and-allocation-on-production-task-num5-unseen-product-quantities.md>): human-robot task planning and allocation; Production Task Num5 (unseen product quantities); Makespan: 879.73.

- [Human-robot task planning and allocation on Production Task Num8 (unseen product quantities)](<https://www.sota2.com/research/sota/human-robot-task-planning-and-allocation-on-production-task-num8-unseen-product-quantities.md>): human-robot task planning and allocation; Production Task Num8 (unseen product quantities); Makespan: 1403.76.

- [Human-robot task planning and allocation on Production Task unseen product quantities Num2](<https://www.sota2.com/research/sota/human-robot-task-planning-and-allocation-on-production-task-unseen-product-quantities-num2.md>): human-robot task planning and allocation; Production Task unseen product quantities Num2; Makespan: 817.98.

- [Human-robot task planning and allocation on Production Task Num3 (unseen product quantities)](<https://www.sota2.com/research/sota/human-robot-task-planning-and-allocation-on-production-task-num3-unseen-product-quantities.md>): human-robot task planning and allocation; Production Task Num3 (unseen product quantities); Makespan: 1051.68.

- [Human-robot task planning and allocation on Production Task Num4 (unseen product quantities)](<https://www.sota2.com/research/sota/human-robot-task-planning-and-allocation-on-production-task-num4-unseen-product-quantities.md>): human-robot task planning and allocation; Production Task Num4 (unseen product quantities); Makespan: 1138.52.

- [Human-robot task planning and allocation on Production Task Num7 (unseen product quantities)](<https://www.sota2.com/research/sota/human-robot-task-planning-and-allocation-on-production-task-num7-unseen-product-quantities.md>): human-robot task planning and allocation; Production Task Num7 (unseen product quantities); Makespan: 1322.52.

- [Human-robot task planning and allocation on Production Task unseen product quantities Num10](<https://www.sota2.com/research/sota/human-robot-task-planning-and-allocation-on-production-task-unseen-product-quantities-num10.md>): human-robot task planning and allocation; Production Task unseen product quantities Num10; Makespan: 1720.59.

- [Human-robot task planning and allocation on Production Task Mean (unseen product quantities)](<https://www.sota2.com/research/sota/human-robot-task-planning-and-allocation-on-production-task-mean-unseen-product-quantities.md>): human-robot task planning and allocation; Production Task Mean (unseen product quantities); Makespan: 1225.94.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1511.06417>)

[PDF](<https://arxiv.org/pdf/1511.06417.pdf>)
