Fresh groundwater far off the coast: 3D numerical simulations of groundwater flow at the New Jersey shelf

本研究的目的是了解地下水循环在新泽西大陆架（NJS）的数值模拟的基础上。在1970年后期，在大西洋边缘取心项目（Atlantic Margin Coring Project）的过程中，沿美国东部大陆边缘沿着首次发现了盐度明显低于海水的地下水。据证明，这种地下淡水储量存在于当地距离海岸线100多公里的范围内。在NJS的IODP 313航次过程中恢复了特别详细的孔隙水盐度记录，揭示了三个钻井位置中每个位置的垂直盐度突变。以往的研究要么解释了存在的淡水在NJS由现今向海定向大气地下水流或淡水侵位在末次冰期。然而，这些过程中哪一个最终在NJS产生了新鲜地下水，目前仍存在争议。此外，所观察到的垂直盐度变化的原因也知之甚少。为了回答这些问题，我们将建立一个详细的三维水文地质模型来模拟NJS的地下水流。在数值模拟的基础上，确定了以下待检验的工作假设：1. NJS的新鲜地下水是在末次冰期沉积的。2.目前，海底淡水地下水的排放范围不超过离岸100公里。3.海底淡水的分布受低渗透性沉积物的控制。为建立三维水文地质模型提供了良好的数据库。它包括ODP第150、174 A航次和IODP第313航次期间采集的大量二维地震剖面、三维地震数据（截至2015年12月）以及岩石物理和测井数据。地震资料的深度偏移将用于建立深度上的三维水文地质模型。将根据IODP第313航次后建立的已公布岩石地层格架，对各个模型层序进行定义。每个层序将被细分为地质相，其中每个相类型的特征在于从ODP/IODP测井和实验室测量获得的岩石物理性质。通过使用多点地质统计学方法来模拟相分布，从三维地震数据的AVO分析得出的岩石物理特性也可以被纳入模型中。通过数值地下水流模拟来检验上述工作假设，需要仔细确定模型的初始和边界条件，包括更新世海平面和温度变化、现今海岸淡水排放、水头（最终考虑冰盖地形）和热通量。这项研究的结果可能适用于整个美国东部大西洋边缘，淡水是广泛遇到的，并会产生更好的了解海底地下水排放一般。

The aim of this study is to understand groundwater circulation at the New Jersey shelf (NJS) based on numerical simulations. In the late 1970, groundwater with salinities significantly lower than seawater was initially discovered in the course of the Atlantic Margin Coring Project along the eastern U.S. continental margin. Such fresh groundwater reserves were shown to be present within a distance of locally more than 100 km from the coastline. A particularly detailed record of pore water salinity was recovered in the course of IODP Leg 313 at the NJS, revealing abrupt vertical salinity variations at each of the three drilling locations. Previous studies either explained the existence of fresh groundwater at the NJS by present-day seaward directed meteoric groundwater flow or by fresh water emplacement during the last glacial period. However, which of these processes finally resulted in fresh groundwater at the NJS is currently debated. Also, the cause for the observed vertical salinity changes is little understood. In order to answer these questions, we will build a detailed 3D hydrogeological model for simulating groundwater flow at the NJS. On the basis of numerical simulations, the following working hypotheses to be tested are defined: 1. Fresh groundwater at the NJS was emplaced during the last glacial period. 2. Present-day submarine fresh groundwater discharge does not extend up to 100 km offshore. 3. The distribution of fresh submarine groundwater is controlled by low-permeability deposits. An excellent data base is available for building a 3D hydrogeological model. It consists of numerous 2D seismic profiles, 3D seismic data (by December 2015) as well as petrophysical and logging data, acquired during ODP Legs 150, 174A and IODP Leg 313. Depth-migration of seismic data will be performed for building a 3D hydrogeological model in depth. Definition of individual model sequences will be based on the published lithostratigraphic framework established after IODP Leg 313. Each sequence will be subdivided into geological facies, where each facies type is characterized by petrophysical properties obtained from ODP/IODP logs and laboratory measurements. By using a multiple-point geostatistical approach for modelling facies distribution, petrophysical properties derived from AVO analyses of 3D seismic data can also be incorporated into the model. Testing the working hypotheses defined above by numerical groundwater flow simulations requires careful definition of the initial and boundary conditions of the model, incorporating Pleistocene sea level and temperature variations, present-day fresh-water discharge at the coast, the hydraulic head (eventually accounting for ice sheet topography) and heat flux. Results of this study may be applicable to the entire eastern U.S. Atlantic margin where fresh water is widely encountered and will yield a better understanding of submarine groundwater discharge in general.