Replication Data for: "Improving litterfall production prediction in China under variable environmental conditions using machine learning algorithms"

Data: "data_updated_0618-empty rows removed-3-2.xlsx" contains 968 records of total annual litterfall production (Mg/ha/yr) collected at 314 forest sites covering the full geographical range of Chinese forests. The sites were distributed across various climatic zones, spanning latitudes from 18.26° to 51.50° N, longitudes from 82.25° to 129.53° E, altitudes from 0 to 4115 m above sea level, mean annual temperatures (MAT) from −5.4 to 25.4 °C, and mean annual precipitation (MAP) levels from 370 to 2800 mm. In addition to the geographical location and climate conditions, associated stand information is also included, such as forest type, stand origin, stand age, mean diameter at breast height (DBH), mean tree height, and stand density. Trap size (i.e., the surface area of the litter traps) is also included as it is a potentially important factor affecting litter collection. Code: "Litterfall v1.2_[total]_public release.ipynb" contains a complete pipeline of data parsing, cleaning, preprocessing, model training, and prediction.

数据文件"data_updated_0618-empty rows removed-3-2.xlsx"包含968条记录，记录了覆盖中国森林全地理范围的314个森林站点的年凋落物总产量（单位：Mg/ha/yr）。这些站点分布于不同气候带，纬度范围为18.26°N至51.50°N，经度范围为82.25°E至129.53°E，海拔范围为海平面以上0至4115米，年平均温度（mean annual temperature, MAT）范围为-5.4至25.4°C，年平均降水量（mean annual precipitation, MAP）范围为370至2800mm。此外，数据集还包含相关林分信息，如森林类型、林分起源、林分年龄、平均胸径（mean diameter at breast height, DBH）、平均树高及林分密度。陷阱大小（即凋落物收集陷阱的表面积）也被纳入其中，因其是影响凋落物收集的潜在重要因素。代码文件"Litterfall v1.2_[total]_public release.ipynb"包含数据解析、清洗、预处理、模型训练及预测的完整流程。