| 02 Sep 2025
Advancing Ecohydrological Modelling: Coupling LPJ-GUESS with ParFlow for Integrated Vegetation and Surface-Subsurface Hydrology Simulations
Abstract. Climate change accelerates the global hydrological cycle, which has escalating impacts on human health and the socioeconomic development. However, many existing Earth system models neglect the more complex processes of topography-driven vegetation-surface-groundwater interactions, thereby failing to accurately capture climate-hydrological responses. To address this gap, we integrate the three-dimensional surface-subsurface hydrological model ParFlow with the dynamic global vegetation model LPJ-GUESS to investigate how lateral groundwater flow and vegetation dynamics jointly regulate hydrological fluxes. The fully coupled ParFlow-LPJ-GUESS (PF-LPJG) model and stand-alone LPJ-GUESS model were used to run hydrological simulations at a resolution of 10 km across the Danube River Basin. A comprehensive evaluation of multiple hydrologic variables – including streamflow, surface soil moisture (SM), evapotranspiration (ET), and water table depth (WTD) was conducted using in situ and remote sensing (RS) observations based on a 38-year (1980–2018) model simulation. The results demonstrate that the PF-LPJG model substantially improves streamflow and surface soil moisture simulations without requiring parameter calibration compared to stand-alone LPJ-GUESS, mitigates the underestimation of summer low flows during dry years, increases the accuracy of peak flow timing in wet years, and achieves a Kling-Gupta Efficiency (KGE) > 0.5 and Spearman’s ρ > 0.80 at over 80 % of gauging stations. Seasonal soil moisture anomalies are better captured (R = 0.51) compared to satellite-based products. Additionally, the modelled WTD agrees well with in-situ monitoring-well data, as indicated by a low RSR value (~1.31, Root Mean Square Error-observations Standard deviation Ratio). Notably, the coupled model improves the representation of bare-soil evaporation and reduces transpiration-to-evaporation (T/E) ratio fluctuations, aligning more closely with the GLEAM v4.2 product. The coupled model PF-LPJG entails a mechanistic framework for capturing bidirectional interactions among surface-subsurface water, vegetation dynamics and ecosystem biogeochemical processes, which can be applied to other catchments or climatic conditions to deeply analyze climate-induced modification on vegetation-water-carbon interactions.
In this manuscript the authors propose a robust coupled model for integrated vegetation and surface-subsurface water flow simulations. This work is very valuable for the egusphere community. It is mostly clear but would needs some additional clarification, illustrations and discussions to improve its readability and repeatability.
Main comments:
Parflow equations and annotations need some clarifications. Describing each equation briefly by one sentence summarizing what it does / means and how they are linked/solved with respect to each other would bring some clarity to the reader (there is already some attempt, but it is still a bit confusing). Time discretization shall be introduced and justified for ParFlow and LPJ-GUESS.
The value of input parameters should be included in the manuscript. Providing references from where it was sourced is great but not sufficient, given that there is no calibration of those parameters. Consider to illustrate them with some figures: landcover map, topography and river map, annual rainfall map and timeseries at one location, hydrogeological model vertical cross section (hydrostratigraphic units, heterogenous property fields), soil data property maps per layer or vertical cross-section. For homogeneous properties inside one layer or a hydro-stratigraphic unit, a table summarizing the parameter values should be provided.
The discussion should be separated from the results section. A more thorough discussion should be written to acknowledge the limitations of the current coupled model and suggest possible modelling improvements on both aspects of the coupled model (Vegetation-Land Surface aspect and Hydro-geo-logical aspect). Given the not so good results of Water Table Depth, calibration of hydrogeological parameters should be discussed with respect to the studied Danube basin or other locations. There is also a river network pattern for ET and SM produced by the coupled model, that are not present in the ‘reference’ data; that should also be discussed.
Detailed comments:
Abstract is clear.
Introduction is clear.
2.2 Parflow
Equation (1) and lines 123-124: is it phi_p (not defined after) or psi_p in equation 1? Should it be q_s(x) in equation 1 instead of q_e(x)?
Line 126, do you mean the boundary conditions q_bc ?
Lines 129-130: are (1) and (2) not the same thing? psi_p = psi_s seems related to the sentence after.
Line 130: define psi_s here.
Equations 4 and 5: what is q_r?
2.3 Coupling model approach
Timestep discretization needs to be introduced in 2.2 to clarify the articulation of the coupled ParfFlow and LPJ-GUESS models; it seems clear that there is a daily time scale interaction, but time discretization could potentially be different between the solvers (finer different discretization).
Figure 1: for consistency, keep the same left-right ordering of soil moisture / precipitation
2.4 Data sets
Line 172: “in a lot of research” seems superfluous, remove it.
Lines 177-179: data at different resolution? Please check and clarify the resolution used for each data-type. What is “u-component of wind”? What is the CDS daily aggregation method?
How many river flow observation points from the Danube River Basin are used? Where are they located?
Can you explain why the GLEAM data can be used as a reference as it is the result of a model?
Same justification needed for the ESA CCI-SM product.
How many in-situ water table depth observation points from the Danube Basin are used? Where are they located?
2.5
Line 241:by “stabilizes less than 1 %”do you mean “stabilizes, with fluctuations less than 1 %” ?
3.1 Streamflow
Global and local results: the boxplots could also be presented at the gauging station level to quantify the performance as a function of basin size. A sketch showing the relationship of the 7 considered basins (sub-basin of the main one), would help the reader understand their relationship, rather than guessing it.
3.2 ET
Figure 4: subtitles of 1st and 2nd row too long, make it hard to read, maybe add Annual ET Difference to the colour-bar legend/label. The colormap to show the difference is not great (subplots d to f): use a seismic or bwr (blue white red) colormap centred around 0 such that white colour denotes no change, blue colours negative difference and red colours positive difference. Subplots a to c: use a colourblind linear colormap.
Subplots a-f: missing scale and North. Subplot (g): missing unit on the x axis
3.3 SM
Figure 5: what does SWC stands for? Subplot (c): use a seismic or bwr (blue white red) colormap centred around 0 such that white colour denotes no change, blue colours negative difference and red colours positive difference. Subplots a,b and d: use a colourblind linear colormap.
Figure 6: not sure how the RMSE and Sperman rho CDFs were calculated, for how many subsamples? How were the subsamples selected?
3.4 Water Table Depth
The CDF error from the PF-LPJG is smaller than from the work of Fan et al. (2013) but It does not seem close to the real observations and strongly biased (shifted cdf). That should be acknowledge and some plausible explanation given. Figure 8c should be compared to real observations at least once; it is not possible to observe to identify spring or other seasons on the graph as year graduations are too small, maybe zoom over a smaller time range to support this statement.
3.5 E T partitioning
Figure 8a is hard to read; create another subplot to separate the evaporation series from the transpiration series. Maybe an additional plot of the residuals as time series would facilitate the interpretation of these results.