Penerapan Model Geographically Weighted Regression(GWR) Pada Produksi Ubi Jalar

Sweet potatoes are a major source of carbohydrate, after rice, corn, and cassava. Sweet potato is consumed as an additional or side meal, except in Irian Jaya and Maluku, sweet potato is used as staple food. The main problem faced in increasing sweet potato production is still relies on certain areas, namely Java Island, as the main producer of sweet potato. Differences in production is what often causes the needs of sweet potato in various regions can not be fulfilled and there is a difference price of sweet potato. To fulfill the needs of sweet potato in Java, mapping areas of sweet potato production need to be made so that areas with potential for producing sweet potato can be developed while areas with insufficient quantities of sweet potato production may be given special attention. Due to differences in production in some areas of Java which depend on soil conditions, altitude, rainfall and temperatures, a model of sweet potato production will be developed using the GWR model. Based on the Geographically weighted regression model for each regencies / cities in Java Island, it can be concluded that the largest sweet potato production coming from Kuningan with R2 equal 99.86%.<br />Keywords : Geographically weighted regression, model, sweet potato

Download Full-text

Spatiotemporal Changes in Vegetation Cover and Its Influencing Factors in the Loess Plateau of China Based on the Geographically Weighted Regression Model

Forests ◽

10.3390/f12060673 ◽

2021 ◽

Vol 12 (6) ◽

pp. 673

Author(s):

Chen Yang ◽

Meichen Fu ◽

Dingrao Feng ◽

Yiyu Sun ◽

Guohui Zhai

Keyword(s):

Regression Model ◽

Influencing Factors ◽

Loess Plateau ◽

Geographically Weighted Regression ◽

Vegetation Cover ◽

Weighted Regression ◽

Vegetation Coverage ◽

The Loess Plateau ◽

Geographically Weighted Regression Model ◽

Spatiotemporal Changes

Vegetation plays a key role in ecosystem regulation and influences our capacity for sustainable development. Global vegetation cover has changed dramatically over the past decades in response to both natural and anthropogenic factors; therefore, it is necessary to analyze the spatiotemporal changes in vegetation cover and its influencing factors. Moreover, ecological engineering projects, such as the “Grain for Green” project implemented in 1999, have been introduced to improve the ecological environment by enhancing forest coverage. In our study, we analyzed the changes in vegetation cover across the Loess Plateau of China and the impacts of influencing factors. First, we analyzed the latitudinal and longitudinal changes in vegetation coverage. Second, we displayed the spatiotemporal changes in vegetation cover based on Theil-Sen slope analysis and the Mann-Kendall test. Third, the Hurst exponent was used to predict future changes in vegetation coverage. Fourth, we assessed the relationship between vegetation cover and the influence of individual factors. Finally, ordinary least squares regression and the geographically weighted regression model were used to investigate the influence of various factors on vegetation cover. We found that the Loess Plateau showed large-scale greening from 2000 to 2015, though some regions showed decreasing vegetation cover. Latitudinal and longitudinal changes in vegetation coverage presented a net increase. Moreover, some areas of the Loess Plateau are at risk of degradation in the future, but most areas showed a sustainable increase in vegetation cover. Temperature, precipitation, gross domestic product (GDP), slope, cropland percentage, forest percentage, and built-up land percentage displayed different relationships with vegetation cover. Geographically weighted regression model revealed that GDP, temperature, precipitation, forest percentage, cropland percentage, built-up land percentage, and slope significantly influenced (p < 0.05) vegetation cover in 2000. In comparison, precipitation, forest percentage, cropland percentage, and built-up land percentage significantly affected (p < 0.05) vegetation cover in 2015. Our results enhance our understanding of the ecological and environmental changes in the Loess Plateau.

Download Full-text