A Population Spatialization Model at the Building Scale Using Random Forest

Mapping livestock density distribution in the Selenge River Basin of Mongolia using random forest

Mapping dynamically distributed livestock in the vast steppe area based on statistical data collected by administrative units is very difficult as it is limited by the quality of statistical data and local geographical environment factors. While, spatial mapping of livestock gridded data is critical and necessary for animal husbandry management, which can be easily integrated and analyzed with other natural environment data. Facing this challenge, this study introduces a spatialization method using random forest (RF) in the Selenge River Basin, which is the main animal husbandry region in Mongolia. A spatialized model was constructed based on the RF to obtain high-resolution gridded distribution data of total livestock, sheep & goats, cattle, and horses. The contribution of factors influencing the spatial distribution of livestock was quantitatively analyzed. The predicted results showed that (1) it has high livestock densities in the southwestern regions and low in the northern regions of the Selenge River Basin; (2) the sheep & goats density was mainly concentrated in 0-125 sheep/km2, and the high-density area was mainly distributed in Khuvsgul, Arkhangai, Bulgan and part soums of Orkhon; (3) horses and cattle density were concentrated in 0-25 head/km2, mainly distributed in the southwest and central parts of the basin, with few high-density areas. This indicates that the RF simulation results effectively depict the characteristics of Selenge River Basin. Further study supported by Geodetector showed human activity was the main driver of livestock distribution in the basin. This study is expected to provide fundamental support for the precise regulation of animal husbandry in the Mongolian Plateau or other large steppe regions worldwide.

Accurate Recognition of Building Rooftops and Assessment of Long-Term Carbon Emission Reduction from Rooftop Solar Photovoltaic Systems Fusing GF-2 and Multi-Source Data

Rooftop solar photovoltaic (PV) retrofitting can greatly reduce the emissions of greenhouse gases, thus contributing to carbon neutrality. Effective assessment of carbon emission reduction has become an urgent challenge for the government and for business enterprises. In this study, we propose a method to assess accurately the potential reduction of long-term carbon emission by installing solar PV on rooftops. This is achieved using the joint action of GF-2 satellite images, Point of Interest (POI) data, and meteorological data. Firstly, we introduce a building extraction method that extends the DeepLabv3+ by fusing the contextual information of building rooftops in GF-2 images through multi-sensory fields. Secondly, a ridgeline detection algorithm for rooftop classification is proposed, based on the Hough transform and Canny edge detection. POI semantic information is used to calculate the usable area under different subsidy policies. Finally, a multilayer perceptron (MLP) is constructed for long-term PV electricity generation series with regional meteorological data, and carbon emission reduction is estimated for three scenarios: the best, the general, and the worst. Experiments were conducted with GF-2 satellite images collected in Daxing District, Beijing, China in 2021. Final results showed that: (1) The building rooftop recognition method achieved overall accuracy of 95.56%; (2) The best, the general and the worst amount of annual carbon emission reductions in the study area were 7,705,100 tons, 6,031,400 tons, and 632,300 tons, respectively; (3) Multi-source data, such as POIs and climate factors play an indispensable role for long-term estimation of carbon emission reduction. The method and conclusions provide a feasible approach for quantitative assessment of carbon reduction and policy evaluation.