New method for matching spatial and temporal data inhuman-Earth system model coupling

The bidirectional coupled operation of the human-Earth system model is hindered by the spatial and temporal mismatch of the Earth system data and comprehensive economic system data. The Earth system model uses geologically gridded data whereas the economic evaluation model uses national or regional administrative statistical data. In addition, the Earth system model and economic evaluation system model are generally not synchronized because they use different time steps. Thus, the modeling methodologies and operational trajectories are totally different, thereby making it difficult to operate them in a single model. To match the two types of data on spatial scales, we proposed transforming the statistical data collected in administrative units into gridded data by using economic models and the inverse operation to convert the gridded data into administrative data. On temporal scales, we proposed to streamline the selected economic data for various time scales to the same time step used by the Earth system model (e.g., both on a “per year” basis). For the first time, we designed an “area-weighted conversion method”, which can be used in a forward direction to convert administrative data into gridded data or in a backward direction to transform gridded data into administrative data, thereby matching the economic data in administrative units with the gridded data in the Earth system model. The practical steps in the area-weighted conversion method are as follows. First, using the Lambert projection, net grids are built with 1°×1° latitude and longitude resolution using the base map for China and the global base map, respectively, thereby yielding 3795 cells for the Chinese mainland and 62640 cells for the globe. Next, in ArcGIS, the net grid is built and the factors are converted into areal units, and then connected to the base map with the economic indices per unit area. Connecting the names of the administrative entities is a crucial step. Using the union operation in the analysis tools, the established net grid cuts the base map into irregular shapes. The exported table of properties contains the areas of the irregular shapes and economic indices per unit area. Finally, after applying the weights to the data mentioned above, the map is reconnected to the ArcGIS net grid. Using the forward and backward methods, we connected the data in the Earth system model and economic evaluation model, and operated them in a coupled manner. Our study was based on data for 2010–2018 obtained from the China National Bureau of Statistics and the United Nations’ databases. We applied the proposed method to data for the Chinese mainland and other countries, such as populations, GDP, and CO2 emissions, with a 1°×1° grid spatial scale and statistical data at various temporal scales were converted into a per year basis. The simulation and test results were satisfactory. The distributions of various indices and characteristics were consistent with reality, thereby indicating that the proposed method is capable of providing a reliable basis for the bidirectional coupled variable operation of the human-Earth system model. Our future research will involve feeding the spatial-temporal matched data converted using our method into the human-Earth system model to conduct simulations, analyzing the result and impacts, and further improving the spatial-temporal resolution and real-time capability.