河口盐沼是一类独特的地貌，是由河流入海泥沙在潮汐波浪的地质塑造作用下形成，位于陆地与海洋生态系统的交汇处，受多种环境因素影响，特别是咸淡水的交互作用，导致河口盐沼土壤水盐分布变化呈显著梯度规律，从而创造多样化的生境，为许多珍稀物种提供了栖息繁殖场所。河口盐沼生态系统在具备高度活力同时也较为脆弱，其结构功能易受外部因素影响。近年来，全球气候变暖导致的海平面上升、入海水量减少以及降雨量季节分布不均等，使得来自降雨和河流的淡水补给不足以抵抗海水入侵对河口盐沼的影响。此外，农业活动、地下水开采等人类活动也是导致河口盐沼土壤盐渍化程度增大的重要因素。

土壤水盐是限制河口盐沼植被生长的重要因素，尽管耐盐植物可在高盐度土壤中存活，但土壤水盐条件的恶化仍会严重阻碍植株的生长繁殖，并改变不同物种之间的竞争关系，加快了入侵物种的扩散。因此，维持相对适宜的土壤水盐条件，对维护河口湿地植被群落乃至生态系统的稳定具有重要意义。综上，本文围绕多要素影响下河口盐沼湿地土壤水盐运移机制，以及土壤水盐变化对植被生长与竞争的影响这两个关键问题开展研究，取得的主要结论如下：

（1）多要素影响下黄河三角洲土壤水盐时空响应规律

为厘清在气候变化与人类活动等多重要素影响下黄河口盐沼湿地水盐的时空演变规律，本研究构建了基于物联网通讯的土壤水盐野外数据监测平台，进而得到黄河三角洲典型河—海梯度和陆—海梯度的地下水和土壤水盐要素，综合考虑气象、水文等多个环境因子，确定了驱动水盐运移过程的关键环境因素。结果表明，由于黄河的补给作用一方面使得沿岸带土壤脱盐，另一方面抬升了地下水埋深，使得潮滩部分区域土壤盐度上升。在无河流影响的海陆梯度上，盐度自海向陆呈下降趋势，分布区间为8.57-33.13 ppt，河流的补给使得盐度的分布区间扩展至6.78-34.64 ppt，且波动范围也有所扩大。海洋潮汐和黄河径流是造成地下水埋深波动的主要影响要素，且其对地下水的影响与样点和岸线间的距离密切相关。在海—河梯度上，气温和土壤温度对土壤水盐的影响均较小，且河流补给可能进一步降低这一影响；黄河对距其较远的N样带几乎无显著影响，而对于海—陆梯度而言，温度和降雨对含水率均有中等正向影响，而对盐度则有较强的负向影响，整体而言，地下水埋深和潮汐高度是影响黄河三角洲土壤含水率和盐度的关键因素。

物理过程与机器学习的河口盐沼湿地土壤水盐混合模型

为开展区域尺度的黄河三角洲土壤水盐时空分布的演变模拟，首先基于地质资料构建了黄河三角洲地下水动力模型，然后结合野外实测长序列数据构建了基于机器学习的土壤水盐运移预测一维模型，并将机器学习模型与地下水动力模型相结合，形成基于物理过程与机器学习的土壤水盐运移混合模型，实现区域尺度的土壤水盐分布预测。选取1998-2021年黄河干流水文序列作为研究情景进行模拟，结果表明，降雨入渗是含水层主要水量流入来源，其次是河流补给，而蒸散发则是主要流出来源。土壤含水率与地下水水位间存在较高的相关性，但地下水的年际间波动对含水率变化影响较小而与年降雨量相关，高盐度区域和高含水率区域有较为明显的重合。整体而言，高盐带与低盐带在中潮滩形成了相互交错的条带状格局。浅层土壤含水率对地下水埋深的响应曲线呈S型，且敏感区间为-0.82 m~-0.08 m，土壤盐度对地下水埋深的响应曲线呈单峰状，其敏感区间为-1.7~-0.5 m。

（3）考虑土壤水盐运移的河口盐沼植被生境分布格局模拟模型

以黄河三角洲盐沼湿地典型植被作为目标物种，基于已有研究成果确定其水盐耐受阈值，结合土壤水盐运移混合模型预测结果，对典型植被在黄河三角洲的生境分布格局进行预测。结果表明，芦苇的适宜生境主要分布在现行河道两侧、北侧潮滩和故道，柽柳的适宜生境仅分布在清8断面两岸地势较高处，翅碱蓬的适宜生境主要以带状分布在距海岸线约3-5 km处，互花米草适宜生境为入海口沙嘴附近。流量的增大导致河口北岸芦苇的适宜生境减小而南岸芦苇和翅碱蓬的适宜生境扩大，柽柳的适宜生境分布相较于前一时期基本没有改变，而互花米草的适宜生境有所减小，但其分布范围呈现向内陆扩展的趋势。

（4）考虑土壤水盐运移过程的河口盐沼植被生物量预测模型

为模拟多要素影响下水盐运移对于盐沼植被生长竞争的影响，通过将有限差分方程与元胞自动机模型相耦合，形成由土壤水盐格局驱动的植被生长与竞争预测模型，在给定的适宜生境结果和初始条件下模拟目标物种在河口盐沼湿地的扩散、生长与竞争过程。结果表明，芦苇群落主要分布在黄河河道近岸带及清水沟流路故道，柽柳主要分布在清8汊附近分布有密集的群落，翅碱蓬在潮滩上大多以小斑块形式存在，沿海岸线内侧分布有呈条带状聚集的翅碱蓬群落斑块，互花米草的生物量主要在入海口沙嘴前段，因繁殖能力较强而形成大片稳定斑块，但其群落无法向内陆扩散。当植被间存在竞争的情况下，芦苇群落的部分小型斑块被翅碱蓬群落取代，但大型斑块较为稳定；柽柳群落在中潮滩和河道近岸带的群落基本都被翅碱蓬或互花米草取代，而南侧凹地大部分柽柳群落保持稳定；翅碱蓬群落在较低潮滩被互花米草入侵，同时在中潮滩占据部分芦苇群落；互花米草群落主要分布在较低低潮滩，其面积基本未受影响；裸斑区大部分位于北岸中潮滩及口门沙嘴附近，少部分位于中潮滩与河岸带间的高盐区域内。

Estuarine salt marshes are unique landforms formed by the deposition of river sediments into the sea, shaped by the combined action of tidal waves. Situated at the confluence of terrestrial and marine ecosystems, these marshes are influenced by various environmental factors, such as brackish water interactions, resulting in a pronounced gradient distribution of soil moisture and salinity. This creates diverse habitat conditions within the estuarine salt marsh area, offering a sanctuary for numerous rare species. Although these wetland ecosystems boast high productivity and abundant biodiversity, their structure and function remain vulnerable to external factors. In recent years, global warming has led to rising sea levels, diminished water flow into the sea, and uneven seasonal distribution of rainfall, causing an insufficient freshwater supply from precipitation and rivers to counteract seawater intrusion. Furthermore, human activities such as agriculture and groundwater extraction contribute to the increased salinization of estuarine salt marsh wetlands.

Soil water and salt content play a crucial role in limiting vegetation growth within estuarine salt marshes. While salt-tolerant plants can adapt to high salinity soils, the deterioration of soil water and salt conditions can still significantly impede plant growth and reproduction. That, in turn, alters the competitive dynamics among various species and accelerates the proliferation of invasive species. Hence, preserving optimal soil water and salt conditions is of paramount importance for maintaining the stability of estuarine wetland vegetation communities and even the overall ecosystem. In conclusion, this paper emphasizes two primary aspects: the soil water-salt transport mechanism and the impact of soil water-salt migration on vegetation growth and competition under the influence of various factors. The main findings are as follows: 

(1) Spatio-temporal response patterns of soil water and salt in the Yellow River Delta under the influence of multiple factors

In order to clarify the spatio-temporal evolution of water and salt in the salt marsh wetland of the Yellow River estuary under the influence of multiple factors such as climate change and human activities, this study constructed a soil water and salt field data monitoring platform based on the Internet of Things communication, and then obtained the groundwater and soil water and salt elements of the typical river-sea gradient and land-sea gradient in the Yellow River Delta, comprehensively considering multiple environmental factors such as meteorology and hydrology. The key environmental factors driving the water and salt migration process were determined. The results show that the replenishment of the Yellow River on the one hand desalts the soil along the coastal zone, and on the other hand raises the groundwater depth, which increases the soil salinity in some areas of the tidal flat. On the sea-land gradient without river influence, the salinity showed a decreasing trend from sea to land, with a distribution range of 8.57-33.13 ppt. The distribution range of salinity extended to 6.78-34.64 ppt due to river recharge, and the fluctuation range also expanded. Ocean tide and Yellow River runoff are the main factors that cause groundwater depth fluctuation, and their influence on groundwater is closely related to the distance between the sample point and the coastline. On the sea-river gradient, the influence of air temperature and soil temperature on soil water and salt is small, and the influence may be further reduced by river recharge. The Yellow River has almost no significant influence on the N transect far away from it, but for the sea-land gradient, temperature and rainfall have moderate positive effects on water content, while strong negative effects on salinity. Overall, groundwater depth and tidal height are the key factors affecting soil water content and salinity in the Yellow River Delta.

Physical process and machine learning soil water-salt mixing model of estuarine salt marsh wetlands

In order to simulate the evolution of spatial-temporal distribution of soil water and salt in the Yellow River Delta at a regional scale, a groundwater dynamic model for the Yellow River Delta was first constructed based on geological data, and then a one-dimensional model for predicting soil water and salt migration based on machine learning was constructed by combining the machine learning model with the groundwater dynamic model. A hybrid model of soil water and salt transport based on physical process and machine learning was developed to predict the distribution of soil water and salt at regional scale. The hydrological sequence of the main stream of the Yellow River from 1998 to 2021 is selected as the study scenario. The results show that rainfall infiltration is the main inflow source of aquifer water, followed by river recharge, and evapotranspiration is the main outflow source. There is a high correlation between soil water content and groundwater level, but the inter-annual fluctuation of groundwater has little influence on the change of water content, but is related to the annual rainfall. The high-salinity region and high-moisture region have obvious overlap. On the whole, the high salt zone and low salt zone form a banded pattern in the middle tidal flat. The response curve of shallow soil water content to groundwater depth is S-shaped, and the sensitive range is -0.82 m~-0.08 m. The response curve of soil salinity to groundwater depth is unimodal, and the sensitive range is -1.7 m~ -0.5 m.

(3) Simulation model of the habitat distribution pattern of estuarine salt marsh vegetation considering the migration of soil water and salt

Taking the typical vegetation of the Yellow River Delta as the target species, the water and salt tolerance threshold was determined based on the existing research results, and combined with the prediction results of the soil water and salt transport mixed model, the habitat distribution pattern of the typical vegetation in the Yellow River Delta was predicted. The results show that the suitable habitats of reed are mainly distributed in the two sides of the current river, the suitable habitat of tamarix is only distributed in the upper terrain, the suitable habitat of wing alkali canopy is mainly distributed about 3-5 km from the coastline, and the suitable habitat of miflora is near the estuary sand mouth. The increase of flow leads to the decrease of the suitable habitat of reed on the north bank of the estuary and the expansion of the suitable habitat of reed and wing base on the south bank. The suitable habitat distribution of tamarix has not changed compared with the previous period, while the suitable habitat of Spartina has decreased, but its distribution range shows a trend to the inland expansion.

(4) The diction model of estuarine salt marsh vegetation biomass considering the soil water and salt migration process

To simulate the influence of water salt migration on salt marsh vegetation growth competition, by coupling the finite difference equation with cellular automata model, forming the soil water salt pattern driven vegetation growth and competition prediction model, to simulate the diffusion, growth and competition process. The results show that the reed community is mainly distributed in the Yellow River river coastal zone and clear ditch flow, tamarx is mainly distributed in qing 8 branch distribution of dense community, wing alkali peng mostly in the beach in the form of small patches, inner distribution along the coastline of wing alkali peng community patches, each grass biomass mainly in the mouth sand front, due to strong reproductive ability and form large stable patches, but the community cannot spread to the inland. When there is competition between vegetation, some small patches of the reed community are replaced by the wing alkali community, but the large patches are relatively stable; the community in the midtidal beach and the community is invaded by the reciprocal herb, and the community is mainly distributed in the low tide beach in the north bank and the sand mouth, which is located in the high salt area between the middle tidal beach and the riparian zone.