SPAMS10 Delfland: Soil motion parameters to model relative surface elevation changes

SPAMS (Simple Parameterization for the Motion of Soils) is a model that provides a first approximation of vertical soil motion for a particular location described by four parameters, given the availability of meteorological data. SPAMS estimates surface motion parameters based on physical processes by assuming that soil displacement is mainly driven by meteorological factors, and distinguishes between reversible and irreversible subsidence. The model uses precipitation and evapotranspiration data from nearby meteorological stations. The SPAMS parameters can be estimated using a certain period of observations from Interferometric Synthetic Aperture Radar (InSAR), extensometers, or other instruments that can measure vertical displacement. The estimated SPAMS parameters are then used to model relative surface elevation over time. These parameters rely on the assumption that the conditions of the area of interest, i.e., soil stratigraphy, ground water management, and land use, remain the same over time.<br>This repository provides the so-called SPAMS10 parameters over the Delfland region in the Netherlands. The SPAMS10 dataset includes:the scaling factor of precipitation: xP [-], see [1]the scaling factor of evapotranspiration: xE [-], see[1]the irreversible subsidence rate: xI [mm/day], see [1]the integration time: tau [days], see [1]the variance of xP: var_xP,the variance of xE: var_xE,the variance of xI: var_xI,the covariance between xP and xE: cov_xPxE,the covariance between xP and xI: cov_xPxI, andand the covariance between xE and xI: cov_xExI.<br>Each set of 10 SPAMS parameters in the dataset refers to a single parcel, including the geolocation of its center-of-mass and relevant contextual information for each parcel. The contextual information includes the ID of the closest meteorological station and the soil code describing the shallow soil type. The meteorological data is provided by the Royal Netherlands Meteorological Institute (KNMI), and details about soil code can be accessed through the basisregistratie ondergrond (BRO—PDOK). Additionally, the residuals sum of squares between InSAR and SPAMS and the degrees of freedom are provided to evaluate model suitability and identify whether an error is present in the mathematical model.<br>The dataset is stored in a parquet data format. In addition, the JSON file provides the metadata related to this dataset. The GitHub repository provides an example script on how to read and model relative surface elevation changes using SPAMS parameters provided in this SPAMS10 dataset.<br>[1] Conroy, P., S. A. N. van Diepen and R. F. Hanssen, "SPAMS: A new empirical model for soft soil surface displacement based on meteorological input data", Geoderma, vol. 440, pp. 1-4, 2023, https://doi.org/10.1016/j.geoderma.2023.116699.

SPAMS（土壤运动简单参数化模型，Simple Parameterization for the Motion of Soils）是一种可在具备气象数据的前提下，基于四个参数对特定地点的垂直土壤运动进行初步近似估算的模型。该模型基于物理过程估算地表运动参数，假设土壤位移主要由气象因素驱动，并区分可逆沉降与不可逆沉降。该模型采用邻近气象站的降水与蒸散发数据。
SPAMS参数可通过干涉合成孔径雷达（Interferometric Synthetic Aperture Radar，InSAR）、伸长计或其他可测量垂直位移的仪器获取的一段时期观测数据进行估算。估算得到的SPAMS参数可随后用于模拟随时间变化的相对地表高程。这些参数的推导基于如下假设：目标区域的条件，即土壤地层、地下水管理与土地利用状况，随时间保持不变。
本仓库提供荷兰代尔夫兰（Delfland）区域内的所谓SPAMS10参数。SPAMS10数据集包含以下内容：降水缩放因子：xP [-]，详见文献[1]；蒸散发缩放因子：xE [-]，详见文献[1]；不可逆沉降速率：xI [mm/day]，详见文献[1]；积分时间：τ [days]，详见文献[1]；xP的方差：var_xP；xE的方差：var_xE；xI的方差：var_xI；xP与xE的协方差：cov_xPxE；xP与xI的协方差：cov_xPxI；以及xE与xI的协方差：cov_xExI。
数据集中每一组共10个SPAMS参数对应一个单一地块，包含该地块质心的地理位置以及相关上下文信息。上下文信息包括最近气象站的ID与描述浅层土壤类型的土壤代码。气象数据由荷兰皇家气象研究所（Royal Netherlands Meteorological Institute，KNMI）提供，土壤代码的详细信息可通过地下基础登记系统（basisregistratie ondergrond，BRO—PDOK）获取。此外，数据集还提供了InSAR与SPAMS之间的残差平方和以及自由度，用于评估模型适配性，并识别数学模型是否存在误差。
该数据集以Parquet数据格式存储。此外，配套的JSON文件提供了本数据集的元数据。本GitHub仓库还提供了示例脚本，用于演示如何利用本SPAMS10数据集中的SPAMS参数读取数据并模拟相对地表高程变化。
[1] Conroy P, van Diepen S A N, Hanssen R F. "SPAMS: A new empirical model for soft soil surface displacement based on meteorological input data"[J]. Geoderma, 2023, 440: 1-4. https://doi.org/10.1016/j.geoderma.2023.116699.