Desiccation cracking is one of the most serious surface failures in farmland soils. Soil structure can be reshaped to modify evaporation pathways in the redistribution of near-surface moisture. However, existing numerical research still treat soil-water movement and crack development as weakly coupled or even separate processes, despite their significance in hydrology and engineering. It is often required to reproduce crack evolution in the fields. In this study, a modeling framework was coupled with farmland soil moisture evaporation and desiccation cracking, with particular emphasis on the dynamic feedback between moisture loss and crack propagation. A soil-water movement was linked with a three-dimensional cracking model from linear elastic fracture mechanics. A systematic simulation was also integrated with moisture variation, volumetric shrinkage, and crack development, rather than as separate stages. Farmland soil was then taken from Yinchuan, Ningxia, western China. Field observations and laboratory response tests were conducted to validate the model after optimization. In the field experiment, crack images and crack depth measurements were collected from replicated test plots during post-saturation drying. Spatial distribution of crack depth was quantified under natural evaporative conditions. A second experiment was designed to establish the response relationship between soil shrinkage rate and volumetric water content. Water content was attributed to the key transfer mechanism between the moisture and cracking sub-model, leading to shrinkage deformation and subsequent fractures. The coupled framework also included feedback from the crack surfaces to the evaporation boundary and seepage behavior, indicating the influence of the cracks on moisture transport. Model parameters were calibrated using Monte Carlo search and field observations. The performance was assessed using geometric and statistical criteria. Crack images were compared from experiments and simulations, according to Minkowski density descriptors. While crack-depth relative frequency was used to evaluate the vertical distribution of cracking. The coefficient of determination (R²), index of agreement (IA), bias (BIAS), and root mean square error (RMSE) were employed to quantify the agreement between simulated and observed values. The field data showed that crack depth was concentrated within the 3–9 cm interval, indicating that the middle layer of the soil profile dominated for crack development under the drying conditions. The coupled model accurately reproduced the key features of crack-network morphology and vertical stratification. The results demonstrate that the strong predictive performance was achieved in the planar crack geometry, with R² values no less than 0.876, IA values no less than 0.953, BIAS values no greater than 0.101, and RMSE values no greater than 0.116. The framework performed better to represent crack length, area, network topology, and especially the propagation of deeper cracks, compared with the baseline desiccation-cracking model that relied primarily on surface evaporation as the driving mechanism. The high accuracy of field-scale simulations was attributed to consider the interaction between soil-water movement and surface cracking. Overall, a practical and physically grounded framework can provide for the co-evolution of moisture evaporation and desiccation cracking in farmland soils. Soil structural degradation under drying can offer the potential value for irrigation, soil-water regulation, and the crack risks in sustainable agriculture.

Field scale crop growth simulation based on the assimilation of observation data and crop growth models is an important method for optimizing field management, agricultural auxiliary decision-making, and crop growth evaluation, which is of great significance for precise management of farmland. In order to construct a numerical model that can accurately simulate the growth and yield of spring wheat in arid areas, this study combined the SWAP (soil water atmosphere plant) model with the IES (iterative ensemble smoother) algorithm to construct a SWAP-IES assimilation simulation system suitable for simulating spring wheat growth in arid areas. Using field test data from 2019 to 2020, the roles of soil water content (SW), leaf area index (LAI), and soil water combination in simulating the growth and yield of spring wheat in arid areas were evaluated, and the impact of assimilation data observation on the yield estimation accuracy of the SWAP-IES system was analyzed and evaluated. The results indicate that: 1) When only assimilating SW, the SWAP-IES system had the highest accuracy in simulating soil moisture (R²=0.87), indicating that assimilating soil moisture can lay the foundation for accurately simulating water stress conditions, evapotranspiration, etc. 2) When there was no assimilation, the R² of the SWAP-IES system for simulating LAI, plant height, and biomass of spring wheat ranged from 0.31 to 0.67, and all treatments were below medium accuracy. The accuracy of LAI simulation for spring wheat was significantly improved when assimilating LAI (R² between 0.76 and 0.96 for each treatment), while LAI+SW had the highest simulation accuracy for spring wheat biomass (R² between 0.73 and 0.92 for each treatment). There was no obvious pattern in plant height, and the simulation of plant height by assimilating LAI and LAI+SW achieved high accuracy (R² ranging from 0.71 to 0.96). Thus, it was necessary to select appropriate observation variables for assimilation simulation based on the research purpose. 3) The accuracy of the SWAP-IES system in predicting spring wheat yield without assimilation was relatively low, with R² of 0.45 and S_RE ranging from 10.89% to 40.34%. The yield estimation had improved when assimilating SW, but the results were still of medium accuracy. The accuracy of yield estimation significantly was improved when assimilating LAI (R² was 0.79), while the overall accuracy of spring wheat yield simulation is the highest when assimilating LAI+SW (R² was 0.87). During the two years, T1, T2, and T3 treatments have the lowest estimated S_RE when assimilating LAI+SW (S_RE ranging from 3.87% to 8.38%), while T4 and T5 treatments had the lowest estimated S_RE when assimilating LAI (all within 10%). 4) The effect of assimilating LAI+SW at flowering stage in single growth period observation data on improving the yield estimation accuracy of SWAP-IES system was the greatest (R² increased from 0.45 without assimilation to 0.74), followed by assimilating observation data at jointing stage and booting stage. Assimilating observation data from multiple growth stages can significantly improve the accuracy of the model's yield estimation. When assimilating LAI+SW observation data from jointing and flowering stages, the estimated yield R² was 0.79, while when assimilating LAI+SW observation data from jointing, booting, and flowering stages, the estimated yield R² reached 0.83. The SWAP-IES assimilation simulation system constructed in this study can effectively simulate the growth and yield formation process of spring wheat under different water conditions by integrating LAI+SW observation data of key growth stages such as flowering and jointing stages, especially under water stress conditions. The results can provide valuable information for using observation equipment such as drones and ground cameras to carry out spring wheat growth monitoring, yield estimation, and precise management under different water management conditions in arid areas.

Soil desiccation and cracking can be one of the most common physical phenomena during natural and man-made damage processes. There is the a significant effect on the mechanical and hydraulic properties of soil in modern agriculture. This study aimed to reveal the development and formation mechanism of desiccation cracking in farmland soil along the depth direction. The Hookean spring model proposed by Vogel was improved to construct the triangular prism network structure with the nodes under the gravity. A three-dimensional soil desiccation cracking model was established to simulate the crack depth. The crack along the depth direction was assumed as the vertical development during simulation. The farm soil desiccation cracking experiment was carried out in the Tongxin County (105°54′E, 36°58′N), Ningxia, China. The ponded water was exerted at the soil surface, until the soil was saturated at the upper 10 cm layer. Camera photography was used to capture the images of crack formation and development at 2-hour intervals for three days. The steps of digital image processing included: the non-experimental part of the image was cropped, the image size was processed to 2048×2048 pixels, while the image was applied geometrically corrected, and a color image of the crack patterns was converted into a gray image, which was then converted into a binary image using the threshold division, and finally the image was processed through the noise removal. The white areas and black pixels in the binary image were represented by the aggregates and crack skeleton, respectively. After that, the Minkowski densities were adopted to quantify the surface crack patterns. Three basic characteristics of cracks were divided into the area A, the length L, and the Euler number E of the crack (representing connectivity in the two-dimensional crack structure). The difference between the Minkowski density and crack depth of the experimental and simulated image was quantified by four evaluation indicators: the coefficient of determination, R², the consistency index, IA, and the root mean square error, RMSE. Finally, a systematic analysis was implemented on the influence of the longitudinal elastic coefficients, K_z (representing the elasticity of soil along the depth direction on the model output. Parameters The parameters of the model were calibrated and verified by the field experimental measurement and photograph. The precision of the simulation was also evaluated. The results show that: The coefficients of determination, R² for the area, length, and Euler number density were between 0.849-0.959. RMSE was between 0.005-0.083, while the BIAS was between 0.103-0.190, and the consistency index IA was 0.965-0.988. As such, the improved 3D model calibration was fully met the requirement of calibration. The model was effective to simulate the formation and development of cracks in farmland soils along the depth direction. The simulated surface crack morphology was consistent with the development of natural cracks. The smaller the longitudinal elastic coefficient was, the more outstanding the trend of crack development along the depth direction was. The peak of crack depth frequency was at soil depth of 6 cm, and the proportion of deep cracks (soil depth of 5-10 cm) was larger. On the contrary, the crack generation and development along the depth direction were inhibited, with the peak of crack depth frequency at (soil depth of 1cm) and a greater proportion of shallow cracks (soil depth of 1-4 cm). The finding can provide an ideal way to predict the development and formation of soil cracks along the depth direction