Understanding landscape-primary productivity and biodiversity relationship through graph metrics

Abstract

Besides landscape ecological studies, the traditional approach to trying to understand ecological processes that occur in a landscape is the use of spatial statistics. However, this does not take into account that many of these processes cannot be observed without considering the multiple interactions that occur between patches of different land use in the landscape. The objective of this research was to explore the use of graph metrics in understanding the processes at the landscape scale, specifically its productivity and plant biodiversity, using three different landscapes. A bibliographic review of the graph metrics was performed, which was separated into landscape and local scales, and into groups within each scale. The usefulness and ecological significance of the metrics was evaluated, the relationship between them and with productivity and biodiversity was analysed. In total, 13 landscape scale metrics and 11 local scale metrics with ecological significance were selected. It was found that a large part of the metrics were able to identify differences between the three landscapes and between locations at local scale, but with different variabilities over time. The metrics had a higher relationship with productivity at both scales, achieving correlations over 70% between the predicted and actual values of productivity, while the biodiversity models achieved a correlation over 45%. Several metrics at both scales were important for predicting both productivity and biodiversity. This study highlights the utility and flexibility of graph theory to understand processes in landscapes, in the context of biodiversity conservation in agricultural landscapes and in landscape ecology.Besides landscape ecological studies, the traditional approach to trying to understand ecological processes that occur in a landscape is the use of spatial statistics. However, this does not take into account that many of these processes cannot be observed without considering the multiple interactions that occur between patches of different land use in the landscape. The objective of this research was to explore the use of graph metrics in understanding the processes at the landscape scale, specifically its productivity and plant biodiversity, using three different landscapes. A bibliographic review of the graph metrics was performed, which was separated into landscape and local scales, and into groups within each scale. The usefulness and ecological significance of the metrics was evaluated, the relationship between them and with productivity and biodiversity was analysed. In total, 13 landscape scale metrics and 11 local scale metrics with ecological significance were selected. It was found that a large part of the metrics were able to identify differences between the three landscapes and between locations at local scale, but with different variabilities over time. The metrics had a higher relationship with productivity at both scales, achieving correlations over 70% between the predicted and actual values of productivity, while the biodiversity models achieved a correlation over 45%. Several metrics at both scales were important for predicting both productivity and biodiversity. This study highlights the utility and flexibility of graph theory to understand processes in landscapes, in the context of biodiversity conservation in agricultural landscapes and in landscape ecology.Besides landscape ecological studies, the traditional approach to trying to understand ecological processes that occur in a landscape is the use of spatial statistics. However, this does not take into account that many of these processes cannot be observed without considering the multiple interactions that occur between patches of different land use in the landscape. The objective of this research was to explore the use of graph metrics in understanding the processes at the landscape scale, specifically its productivity and plant biodiversity, using three different landscapes. A bibliographic review of the graph metrics was performed, which was separated into landscape and local scales, and into groups within each scale. The usefulness and ecological significance of the metrics was evaluated, the relationship between them and with productivity and biodiversity was analysed. In total, 13 landscape scale metrics and 11 local scale metrics with ecological significance were selected. It was found that a large part of the metrics were able to identify differences between the three landscapes and between locations at local scale, but with different variabilities over time. The metrics had a higher relationship with productivity at both scales, achieving correlations over 70% between the predicted and actual values of productivity, while the biodiversity models achieved a correlation over 45%. Several metrics at both scales were important for predicting both productivity and biodiversity. This study highlights the utility and flexibility of graph theory to understand processes in landscapes, in the context of biodiversity conservation in agricultural landscapes and in landscape ecology.Besides landscape ecological studies, the traditional approach to trying to understand ecological processes that occur in a landscape is the use of spatial statistics. However, this does not take into account that many of these processes cannot be observed without considering the multiple interactions that occur between patches of different land use in the landscape. The objective of this research was to explore the use of graph metrics in understanding the processes at the landscape scale, specifically its productivity and plant biodiversity, using three different landscapes. A bibliographic review of the graph metrics was performed, which was separated into landscape and local scales, and into groups within each scale. The usefulness and ecological significance of the metrics was evaluated, the relationship between them and with productivity and biodiversity was analysed. In total, 13 landscape scale metrics and 11 local scale metrics with ecological significance were selected. It was found that a large part of the metrics were able to identify differences between the three landscapes and between locations at local scale, but with different variabilities over time. The metrics had a higher relationship with productivity at both scales, achieving correlations over 70% between the predicted and actual values of productivity, while the biodiversity models achieved a correlation over 45%. Several metrics at both scales were important for predicting both productivity and biodiversity. This study highlights the utility and flexibility of graph theory to understand processes in landscapes, in the context of biodiversity conservation in agricultural landscapes and in landscape ecology.Besides landscape ecological studies, the traditional approach to trying to understand ecological processes that occur in a landscape is the use of spatial statistics. However, this does not take into account that many of these processes cannot be observed without considering the multiple interactions that occur between patches of different land use in the landscape. The objective of this research was to explore the use of graph metrics in understanding the processes at the landscape scale, specifically its productivity and plant biodiversity, using three different landscapes. A bibliographic review of the graph metrics was performed, which was separated into landscape and local scales, and into groups within each scale. The usefulness and ecological significance of the metrics was evaluated, the relationship between them and with productivity and biodiversity was analysed. In total, 13 landscape scale metrics and 11 local scale metrics with ecological significance were selected. It was found that a large part of the metrics were able to identify differences between the three landscapes and between locations at local scale, but with different variabilities over time. The metrics had a higher relationship with productivity at both scales, achieving correlations over 70% between the predicted and actual values of productivity, while the biodiversity models achieved a correlation over 45%. Several metrics at both scales were important for predicting both productivity and biodiversity. This study highlights the utility and flexibility of graph theory to understand processes in landscapes, in the context of biodiversity conservation in agricultural landscapes and in landscape ecology.