Agricultural land suitability analysis: State-of-the-art and outlooks for integration of climate change analysis

Machine learning methods, specifically, convolutional neural networks (CNNs), have emerged as an integral part of scientific research in many disciplines. However, these powerful methods often fail to perform pattern analysis and knowledge mining with spatial vector data because in most cases, such data are not underlying grid-like or array structures but can only be modeled as graph structures. The present study introduces a novel graph convolution by converting it from the vertex domain into a point-wise product in the Fourier domain using the graph Fourier transform and convolution theorem. In addition, the graph convolutional neural network (GCNN) architecture is proposed to analyze graph-structured spatial vector data. The focus of this study is the classical task of building pattern classification, which remains limited by the use of design rules and manually extracted features for specific patterns. The spatial vector data representing grouped buildings are modeled as graphs, and indices for the characteristics of individual buildings are investigated to collect the input variables. The pattern features of these graphs are directly extracted by training labeled data. Experiments confirmed that the GCNN produces satisfactory results in terms of identifying regular and irregular patterns, and thus achieves a significant improvement over existing methods. In summary, the GCNN has considerable potential for the analysis of graph-structured spatial vector data as well as scope for further improvement.

A graph convolutional neural network for classification of building patterns using spatial vector data

How land transfer affects agricultural land use efficiency: Evidence from China’s agricultural sector

As a main type of urban construction land, urban-industrial land is used to provide the judging criteria for construction land scale in the planning period when urban population, industrial development, investment scale, and other conditions are uncertain in China; however, research on expected indicator such as urban-industrial land in overall land use plan mainly focuses on qualitative analysis; quantitative analysis research has not yet been carried out. Using MATLAB R2016a software modeling tools to establish GM (1, 1) model and RBF neural network model, respectively, this paper predicted the demand of urban-industrial land in Beijing-Tianjin-Hebei Urban Agglomeration. Comparing the predicated results with the actual value of urban-industrial land in Beijing, Tianjin, and 11 prefecture-level cities in Hebei Province, we determined the reasonable prediction model for urban-industrial land after testing the accuracy of the two prediction models. The results showed that the RBF neural network model was the more reasonable prediction model for urban-industrial land. Using the predicted results of the RBF neural network model, combining expected indicators of overall land use plan (2006–2020) in Beijing and Tianjin, as well as 11 prefecture-level cities in Hebei Province in the planning target year, determined remaining usable time of urban-industrial land. Finally, combined with the actual scale of urban-industrial land in 2015 and the predicated scale of urban-industrial land in 2020, the remaining usable time of each city’s urban-industrial land was calculated in terms of the average annual growth rate of urban-industrial land from 2009 to 2015. According to the comparative relationship between the remaining usable time and the remaining time of the overall land use plan (5 years), urban-industrial lands were divided into three kinds of regulation zones: reasonable reduction zone, optimized adjustment zone, and core development zone. The policy implications for urban-industrial land in each regulation zone were also provided. This paper can provide reference for regulation zoning of urban-industrial land in developing countries and regions.

Demand prediction and regulation zoning of urban-industrial land: Evidence from Beijing-Tianjin-Hebei Urban Agglomeration, China.

This article presents a discussion about the issue of soil resource management in the context of sustainable development. These resources are one of the basic resources on Earth, conditioning the sustainable development of present and future generations by providing the capacity for food production. The study contains an indicator-based assessment of changes in available soil resources as a result of progressive urbanization processes in the suburban area of a big city. The case study was the Wrocław Larger Urban Zone in Poland. The applied methodology allowed for the valorization of urban sprawl processes on agricultural land. As a result, municipalities were divided into groups that characterize different suburbanization processes and their impact on the available soil resources. By using the proposed approach, it is possible to support the spatial development management process in order to protect the valuable components of the environment. The results of the research show that 29 rural precincts located around Wrocław have a high soil transition index (W≥50), which proves their inefficient agricultural management system on a regional scale. The study is an attempt to answer a question on the practical implementation of sustainable development goals that were included in Agenda 2030.

/pdf/indicator-based-assessment-for-soil-resource-management-in-3lsci1mz4o.pdf

Indicator-based assessment for soil resource management in the Wrocław larger urban zone of Poland

A GIS-based agricultural land-use allocation model coupling increase and decrease in land demand

The complex, dynamic and non-linear characteristics of land use change (LUC) modelling necessitate the development of advanced models and methods. In this research, a novel and stepwise approach is developed for modelling integrated LUC at the regional level based on land suitability, land demand and land use conversion rules. The aim of this model is to use different models (i.e. geographic information systems, remote sensing, fuzzy logic, cellular automata and multi-criteria decision analysis) in a logical way to improve model development, model evaluation and prediction of LUC. The proposed model was implemented in ‘Borkhar and Meymeh’ township, Iran, for two periods (1986–1998 and 1998–2005). Results for producer accuracy, user accuracy, overall accuracy and figure of merit were 0.345, 0.357, 0.351, 0.234 respectively in the first period (1986–1998) and 0.302, 0.287, 0.0.294, 0.202 respectively in the second period (1998–2005). Our results suggest that the developed model is accurate, when com...

Developing a methodology for modelling land use change in space and time

Building pattern extraction is an essential step in building generalization. Although many studies have already been conducted, there is a lack of a framework for extracting building patterns. To o...

An integrated framework for linear pattern extraction in the building group generalization process

Building clustering is an important task that should be performed prior to building generalization operations. One of the most common approaches for building clustering is the use of density-based ...

Parastoo Pilehforooshha

Papers

A GIS-based agricultural land-use allocation model coupling increase and decrease in land demand

Developing a methodology for modelling land use change in space and time

An integrated framework for linear pattern extraction in the building group generalization process

A local adaptive density-based algorithm for clustering polygonal buildings in urban block polygons

A new model combining building block displacement and building block area reduction for resolving spatial conflicts