Pedotransfer functions for extended hydraulic models using high resolution soil hydraulic measurements

对土壤-植物大气系统中水分运移的适当了解对许多有关环境和农业问题的主题起着重要作用。用Richards方程模拟水分运移时，必须知道土壤的水力函数，即持水函数和导水函数。这些函数通常是通过精心测量并随后对数据进行适当的模型拟合而得出的。如果这样的数据不可用或无法获取，则通常使用PedoTransfer函数(PTF)。这些PTF通过回归方法从土壤的基本性质(质地、有机碳含量或容重)推导出水力函数的参数，这些基本性质通常是可用的。这对于大规模环境过程的分布式建模尤其重要，例如集水区尺度的水文建模或气象模型中的土壤表面湿度建模。在过去的几十年里，许多PTF都是为了这样的目的而产生的。然而，对这类PTF的比较表明，一般的PTF对于本地应用来说并不精确，因此需要针对本地和区域应用进行特殊校准。在我们的建议(I)中，我们假设到目前为止发展的PTF忽略了水力性质的基本特征，如膜的导水率，从而导致估计参数的系统误差。此外，(Ii)这些PTF是基于分辨率和湿度范围有限的数据。由于现代测量技术，在扩展的水分范围内存在大量高分辨率的水力数据。基于这些数据开发了新的扩展和一致的水力模型。该项目的目的是将新的模型与新的数据结合起来，得出新的改进的PTF。与经典的PTF进行比较，可以看出新开发的PTF是否以及在多大程度上导致了对土壤水力性质的更好估计。

An appropriate knowledge of water transport in the soil-plant atmosphere system plays a major role for many topics regarding environmental and agricultural questions. For modeling water transport with the Richards equation, the soil hydraulic functions, i.e. the water retention and hydraulic conductivity function must be known. These functions are usually derived by elaborate measurements with subsequent fitting of appropriate models to the data. Pedotransfer functions (PTF) are often used if such data are not available and cannot be taken. These PTF derive by means of regression approaches parameters of the hydraulic functions from basic soil properties (texture, organic carbon content or bulk density), which are often available. This is especially important for distributed modeling of environmental processes on large scales, like hydrological modeling at the catchment scale or modeling of the soil surface moisture in meteorological models. In the past decades, many PTF have been derived for such purposes. Yet, comparison of such PTF showed that general PTF are not precise for local applications and therefore a special calibration for local and regional applications was required. In our proposal (i) we hypothesize that the hitherto developed PTF neglect essential features of hydraulic properties, like water conductivity in films, and thus lead to systematic errors in the estimated parameters. Furthermore, (ii) these PTF are based on data with limited resolution and moisture range. Due to modern measurement techniques, there exists a multitude of highly resolved hydraulic data in an extended moisture range. New extended and consistent hydraulic models have been developed based on these data. The aim of the project is to combine the new models with the new data to derive new improved PTF. A comparison with classic PTF will show if, and how much, the newly developed PTF lead to better estimates of soil hydraulic properties.