The efficacy of phosphorus (P) based fertilizers is frequently compromised by soil dynamics that render much of the applied P unavailable for crops. This study aimed to: (i) validate a new P model's prediction of plant-available P; (ii) analyze the effects of organic versus mineral fertilization on P availability and crop yield; and (iii) examine temporal changes in P pools under various fertilization regimes. Data were collected from two long-term field trials, Dikopshof and Bad Lauchstädt, in Germany, using organic (FYM), mineral (MIN), a combination of organic and mineral (MIX) fertilizers, and unfertilized treatments. The AgroC model, incorporating a new P module, accurately predicted P dynamics in cropped plots. At both sites, MIX presented the highest yield, P removal, total P and available soil P. After 120 years of repeated P fertilization, simulations at Dikopshof revealed a positive P balance in MIN (11.1 % with observed 13 %) and in MIX (15 % with observed 15 %), but negative in FYM (-4.9 % with observed -5 %). However, at Bad Lauchstädt, the P balance was negative in all treatments except in MIN (+1.04 %), indicating P depletion. Among crops, cereals showed the most variated yields, with P-use efficiency ranging from 50 % to 99 %, while sugar beet presented the highest P-use efficiency (up to 122 %). The lowest P application rates exhibited, FYM treatment, the highest P-use efficiency for all crops. Model pools were successfully linked to field-measured soil P fractions using CAL and DGT methods, providing initial predictions of various soil P fractions across different fertilization strategies.
In arid regions, groundwater resources are prone to depletion due to excessive water use and little recharge potential. Especially in sand dune areas, groundwater recharge is highly dependent on vadose zone properties and corresponding water fluxes. Nevertheless, vadose zone water fluxes under arid conditions are hard to determine owing to, among other reasons, deep vadose zones with generally low fluxes and only sporadic high infiltration events. In this study, we present an inverse model of infiltration experiments accounting for variable saturated nonisothermal water fluxes to estimate effective hydraulic and thermal parameters of dune sands. A subsequent scenario modeling links the results of the inverse model with projections of a global climate model until 2100. The scenario modeling clearly showed the high dependency of groundwater recharge on precipitation amounts and intensities, whereas temperature increases are only of minor importance for deep infiltration. However, simulated precipitation rates are still affected by high uncertainties in the response to the hydrological input data of the climate model. Thus, higher certainty in the prediction of precipitation pattern is a major future goal for climate modeling to constrain future groundwater management strategies in arid regions.
Abstract Modeling of the land surface water‐, energy‐, and carbon balance provides insight into the behavior of the Earth System, under current and future conditions. Currently, there exists a substantial variability between model outputs, for a range of model types, whereby differences between model input parameters could be an important reason. For large‐scale land surface, hydrological, and crop models, soil hydraulic properties (SHP) are required as inputs, which are estimated from pedotransfer functions (PTFs). To analyze the functional sensitivity of widely used PTFs, the water fluxes for different scenarios using HYDRUS‐1D were simulated and predictions compared. The results showed that using different PTFs causes substantial variability in predicted fluxes. In addition, an in‐depth analysis of the soil SHPs and derived soil characteristics was performed to analyze why the SHPs estimated from the different PTFs cause the model to behave differently. The results obtained provide guidelines for the selection of PTFs in large scale models. The model performance in terms of numerical stability, time‐integrated behavior of cumulative fluxes, as well as instantaneous fluxes was evaluated, in order to compare the suitability of the PTFs. Based on this, the Rosetta, Wösten, and Tóth PTF seem to be the most robust PTFs for the Mualem van Genuchten SHPs and the PTF of Cosby for the Brooks Corey functions. Based on our findings, we strongly recommend to harmonize the PTFs used in model inter‐comparison studies to avoid artifacts originating from the choice of PTF rather from different model structures.
Abstract. The reduction of information contained in model time series through the use of aggregating statistical performance measures is very high compared to the amount of information that one would like to draw from it for model identification and calibration purposes. It has been readily shown that this loss imposes important limitations on model identification and -diagnostics and thus constitutes an element of the overall model uncertainty. In this contribution we present an approach using a Self-Organizing Map (SOM) to circumvent the identifiability problem induced by the low discriminatory power of aggregating performance measures. Instead, a Self-Organizing Map is used to differentiate the spectrum of model realizations, obtained from Monte-Carlo simulations with a distributed conceptual watershed model, based on the recognition of different patterns in time series. Further, the SOM is used instead of a classical optimization algorithm to identify those model realizations among the Monte-Carlo simulation results that most closely approximate the pattern of the measured discharge time series. The results are analyzed and compared with the manually calibrated model as well as with the results of the Shuffled Complex Evolution algorithm (SCE-UA). In our study the latter slightly outperformed the SOM results. The SOM method, however, yields a set of equivalent model parameterizations and therefore also allows for confining the parameter space to a region that closely represents a measured data set. This particular feature renders the SOM potentially useful for future model identification applications.
Soil water content (SWC) plays a crucial role in the production and transport of CO 2 in soils. Classical approaches estimating the effects of SWC on soil respiration are incubation experiments, where soil structure is disturbed and transport processes are neglected. Nevertheless, such data govern the water reduction function of C turnover models. In our approach, the water reduction control parameters (WRCP) of a water reduction function were estimated from column experiments using inverse modeling. Therefore, we used the SOILCO2–RothC model in combination with multistep outflow (MSO) experiments. First, the effective hydraulic properties were estimated and then used in a second experiment to estimate the WRCP and rate constants of the resistant plant material (RPM) C pool. The results showed that the estimated hydraulic parameters can be used for the prediction of CO 2 production and transport of a second MSO experiment only if the WRCP and the C turnover rate of the RPM pool of RothC will also be optimized. Optimizing only the WRCP matched the CO 2 efflux fairly well but the WRCP at the highest matric potential, which determines the start of reduction, was too low at −1.61 cm and (water‐filled pore space [WFPS] = 99.9%). Calibrating both WRCP and the RPM rate constant matched the efflux again fairly well and the results indicate a reduction of optimal CO 2 production at water contents of 0.224 m 3 m −3 or 53.3% WFPS. Also, the estimated RPM rate constant seems to be in a reasonable range at k RPM = 2.5791 × 10 −7 cm −1 .
Abstract Terrestrial evapotranspiration (ET) is the second largest water flux in the global water cycle. It can be measured with different techniques; weighable lysimeters can provide very accurate measurements, and some very long‐term time series exist. However, these lysimeter time series are affected by data gaps that must be filled to estimate actual ET totals and long‐term trends. In this paper, we explore four different gap‐filling methods: the potential ET‐method, the ratio method, the FAO‐based water balance method, and HYDRUS modeling. These gap‐filling methods were evaluated for three time series of actual ET measured by lysimeters and meteorological data of three European sites. Separate evaluations were made for the five driest and five wettest April–October periods to investigate whether the performance of the gap‐filling methods was affected by hydrological conditions. Series of random gaps were artificially created for the three time series, including gaps of four different lengths. Actual ET was estimated for these gaps with the gap‐filling methods, which were evaluated based on RMSE and mean bias error. The results show that the ratio method outperformed other methods for gap filling of lysimeter data for Basel (Switzerland), whereas the HYDRUS method outperformed other methods for Rheindahlen (Germany). For Rietholzbach (Switzerland), the different methods performed very similarly, except that the FAO method gives slightly larger RMSEs. The gap‐filling methods do not perform very differently for dry and wet conditions. The ratio method is recommended for filling smaller gaps, and the HYDRUS method is recommended for longer gaps of 30 d.
