No abstract

Abstract:The spatial structure of Beijing has changed dramatically since the reforms of the late 1970s. It is not clear, however, whether these changes have been sufficient to transform the city's monocentric spatial structure into a polycentric one. This paper uses 2010 enterprise registered data to investigate the spatial distribution of employment in Beijing. Using a customized grid to increase the spatial resolution of our results, we explore the city's employment density distribution and investigate potential employment subcenters. This leads to several findings. First, Beijing still has strong monocentric characteristics; second, the city has a very large employment center rather than a small central business district; third, five subcenters are identified, including four in the suburbs; and fourth, a polycentric model that includes these subcenters possesses more explanatory power than a simple monocentric model, but by only four percentage. We conclude that the spatial structure of Beijing is still quite monocentric, but may be in transition to a polycentric pattern.

Rapid urban land expansion and the resulting arable land loss have put food security in China at risk. This paper investigates the characteristics and mechanism of arable land conversion in Beijing using a logistic model based on land-use data for 2001 and 2010. The results suggest that (1) arable land conversion tends to occur near built-up areas, city centers and major roads; (2) arable land that lies closer to irrigation canals and country roads is less likely to be converted to urban use; (3) arable land that is bigger in size and has a more regular shape has a lower probability of conversion to urban use; and (4) the Prime Farmland Protection policy and related land-use plan have played a positive role in preserving arable land, demonstrated by the probability for arable land conversion inside a prime farmland boundary is 63.9 percent less than for land outside the boundary. Based on these findings and on sustainable-development principles, we suggest that, rather than an exclusive focus on controlling the quantity of arable land, the location and characteristics of the arable land should be a primary consideration when designing urban policies and plans.

No abstract

The soil fertility index (SFI) is an indicator that is commonly used to evaluate the soil fertility in the rice paddy regions of China. However, calculating an SFI requires the laboratory measurement of multiple soil properties, which adds to the costs and complexity of conventional methods. Visible and near‐infrared (vis–NIR, 400–2,500 nm) spectra might offer opportunities to cost‐effectively and rapidly assess soil properties and the SFI. In this study, we evaluated soil fertility properties and their derived SFIs for paddy fields in southern China using vis–NIR spectra with the partial least‐squares regression model. The results showed very good prediction accuracy of vis–NIR spectroscopy for soil organic matter, clay and sand contents, with the former two having an apparent spectral response in the near‐infrared range. A good predictive ability was obtained for total nitrogen, silt, available nitrogen and total potassium. In contrast, the prediction accuracy was only moderate for cation exchange capacity and available phosphorus, and it was poor for the pH, total phosphorus and available potassium without apparent spectral responses. A bootstrapped partial least squares regression model predicted the SFI directly from the vis–NIR spectra accurately (R2 of 0.80 and ratio of performance to interquartile range of 3.12), whereas SFI computed from vis–NIR estimates of the individual indicators was less accurate. This shows that vis–NIR spectroscopy can improve the efficiency of soil fertility assessments within the cultivated paddy rice regions of southern China.Highlights Soil fertility index (SFI) can be predicted directly with spectroscopy. Prediction accuracy of vis–NIR spectroscopy was good for SOM, TN and clay. SFI estimated directly from vis–NIR spectra was superior to SFI calculated from predicted individual indicator. Successful prediction of SFI from vis–NIR spectra can be attributed to the high significance of SOM and TN.

China has a unique land use system in which there are two types of land ownership, namely, state-owned urban land and farmer collective-owned rural land. Despite strict restrictions on the use rights of farmer collective-owned land, rural land is, in fact, developed along two pathways: it is formally acquired by the state and transferred into state ownership, or it is informally developed while remaining in collective ownership. Taking Beijing, the capital, as an example, and using data from land use surveys in the Changping district of the city, this paper examines the spatial patterns and characteristics of land development along these two pathways, and sets up a multinomial logit model to test whether land development with and without ownership change was affected by the same factors. The findings suggest that: (1) development on collective-owned land is more spatially scattered, and its links to public infrastructure are weaker; (2) transportation infrastructure, farmer income, spatial location, and previous land usage are the major factors that affect land development; nevertheless, the influences of the factors are different for the two pathways.

No abstract

Deep learning is characterized by its strong ability of data feature extraction. This method can provide unique advantages when applying it to visible and near-infrared spectroscopy for predicting soil organic matter (SOM) content in those cases where the SOM content is negatively correlated with the spectral reflectance of soil. This study relied on the SOM content data of 248 red soil samples and their spectral reflectance data of 400–2450 nm in Fengxin County, Jiangxi Province (China) to meet three objectives. First, a multilayer perceptron and two convolutional neural networks (LeNet5 and DenseNet10) were used to predict the SOM content based on spectral variation and variable selection, and the outcomes were compared with that from the traditional back-propagation neural network (BPN). Second, the four methods were applied to full-spectrum modeling to test the difference to selected feature variables. Finally, the potential of direct modeling was evaluated using spectral reflectance data without any spectral variation. The results of prediction accuracy showed that deep learning performed better at predicting the SOM content than did the traditional BPN. Based on full-spectrum data, deep learning was able to obtain more feature information, thus achieving better and more stable results (i.e., similar average accuracy and far lower standard deviation) than those obtained through variable selection. DenseNet achieved the best prediction result, with a coefficient of determination (R2) = 0.892 ± 0.004 and a ratio of performance to deviation (RPD) = 3.053 ± 0.056 in validation. Based on DenseNet, the application of spectral reflectance data (without spectral variation) produced robust results for application-level purposes (validation R2 = 0.853 ± 0.007 and validation RPD = 2.639 ± 0.056). In conclusion, deep learning provides an effective approach to predict the SOM content by visible and near-infrared spectroscopy and DenseNet is a promising method for reducing the amount of data preprocessing.