A flexible open-source modular framework for ecohydrological modeling: Application and validation of GEOSPACE-1D

D'Amato, Concetta; Benettin, Paolo; Rinaldo, Andrea; Rigon, Riccardo

doi:10.5194/egusphere-2026-2385

Preprints

https://doi.org/10.5194/egusphere-2026-2385

Preprints

28 Jul 2026

| 28 Jul 2026

Status: this preprint is open for discussion and under review for Geoscientific Model Development (GMD).

A flexible open-source modular framework for ecohydrological modeling: Application and validation of GEOSPACE-1D

Concetta D'Amato, Paolo Benettin, Andrea Rinaldo, and Riccardo Rigon

Abstract. Ecohydrology is a rapidly evolving interdisciplinary field. It explores the interactions among the variety of hydrological processes and ecosystems, specifically water controls on the biota. Advances in measurement techniques and modeling frameworks have enhanced our understanding of key processes, such as vegetation-rainfall interactions, soil moisture dynamics, water-borne disease infections, and the impacts of landscape evolution on streamflow. However, most existing ecohydrological models rely on rigid, predefined components and parameterisations, which limit their flexibility to diverse case studies and datasets. Addressing these limitations requires the implementation of open-source modular frameworks that facilitate customization, comparison, and integration of several modeling components.

The present study introduces and validates GEOSPACE (Soil-Plant-Atmosphere Continuum Estimator), a flexible, open-source ecohydrological framework within the GEOframe system. GEOSPACE employs object-oriented programming principles to model mass and energy fluxes within the critical zone, focusing on processes such as evapotranspiration, soil water dynamics, and vegetation water uptake. Its modular architecture offers multiple alternative formulations for key processes, allowing researchers to tailor simulations to specific scenarios, compare different models, and extend the capabilities of the model. The study uses data from a previous experiment to demonstrate the effectiveness of GEOSPACE in simulating ecohydrological processes under diverse conditions. GEOSPACE is therefore identified as a valuable tool for addressing critical ecohydrological challenges, paving the way for innovative and reliable solutions in a changing environmental context.

Received: 26 Apr 2026 – Discussion started: 28 Jul 2026

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Preprint (PDF, 5693 KB)

Supplement (2252 KB)

Download & links

Concetta D'Amato, Paolo Benettin, Andrea Rinaldo, and Riccardo Rigon

Status: open (until 06 Oct 2026)

Post a comment Subscribe to comment alert

RC1:
'Comment on egusphere-2026-2385', Anonymous Referee #1, 02 Aug 2026 reply
The manuscript describes the application and evaluation of the GEOSPACE ecohydrological model with data from the «Spike II» lysimeter experiment. The study is well-organized and clearly written, addressing an important topic in ecohydrological modeling. The authors provide a thorough description of the experimental setup and demonstrate the model's capacity to replicate various components of the soil-plant water balance. The study is generally suitable for publication. However, in my opinion, the manuscript has several major and minor shortcomings that should be addressed before publication.
Major comments
The limitations of lysimeter measurements should be discussed more thoroughly. Although the manuscript uses lysimeter observations as the reference dataset for model validation, it provides little discussion of the lysimeter method's inherent limitations. While lysimeters are one of the reliable techniques for measuring evapotranspiration, their accuracy depends on specific experimental conditions. In particular, accurate evaporation estimates require minimal advection effects, meaning the temperature, humidity, and aerodynamic conditions of the incoming air must closely resemble those of the surrounding surface. Otherwise, additional sensible and latent heat transport may lead to over- or underestimation of evaporation. Additionally, the representativeness of lysimeter measurements may be affected by edge effects, differences in vegetation development, and local microclimatic conditions. The authors should discuss whether these requirements were met during the Spike experiment and evaluate how the lysimeter measurements' potential methodological limitations may have influenced the reported model validation.

The methodology used to partition evapotranspiration into soil evaporation and plant transpiration remains unclear. Since lysimeters only measure total evapotranspiration, transpiration cannot be considered an independently observed variable. If the partitioning is performed solely by the model, the simulated transpiration cannot be considered directly validated against experimental data. Alternatively, if transpiration is inferred from comparisons among different lysimeters, then the methodology and its underlying assumptions must be described in greater detail. This approach could be influenced by variations in vegetation structure, aerodynamic properties, soil moisture conditions, and microclimate among lysimeters, which could introduce uncertainty into the partitioning process. The manuscript would benefit from a clearer distinction between directly measured variables and model-derived quantities, as well as a discussion of the limitations of this validation approach.

The model was evaluated using a relatively small number of lysimeters, including those representing a single tree, several grass-covered lysimeters, and bare soil. While this dataset is suitable for evaluating the model's performance under these specific experimental conditions, it does not provide sufficient evidence of the model's broader applicability to different vegetation types or ecosystems. Therefore, conclusions regarding the model's general applicability should be presented more cautiously.

The limitations of the modeling approach should be discussed more thoroughly. While the manuscript briefly mentions the absence of hydraulic redistribution and stable isotope transport, it pays little or no attention to other important limitations. These include the one-dimensional formulation of the model, the simplified representation of vegetation, the absence of sensitivity analysis, and the lack of a quantitative assessment of uncertainty. A more thorough discussion of these limitations would clarify the model's current capabilities and prevent misinterpretation of the validation results.

The conclusions section repeats parts of the introduction and discussion, offering virtually no new scientific findings. In conclusion, it is appropriate to emphasize that successfully simulating evaporation at the scale of individual lysimeters does not demonstrate comparable performance under natural field conditions. Natural ecosystems exhibit much greater spatial heterogeneity in soil properties, vegetation structure, and surface-atmosphere interactions than controlled lysimeter experiments do. These factors may affect model performance and should be considered when discussing the model's broader applicability.

Minor comments
L41 There is some inconsistency in the wording of the definitions of evaporation and transpiration. The listed approaches do not explicitly estimate transpiration. The PT equation is for potential evaporation. The H method is a simple way to estimate reference evapotranspiration. The PM equation is used to estimate actual evapotranspiration.
L43 (Penman and Keen, 1948). It is one of the possible references to the Penman equation. Actually, if you like to refer to the Penman-Monteith equation for potential evapotranspiration, you need to choose another reference. Refer to J. Monteith.
L45 (D’Amato and Rigon, 2025) Actually, there are a number of much more sophisticated models.
L150 A more detailed description of the models is required.
L151 PM for evapotranspiration
L155 However, LAI and LAD can be recalculated using, for example, biomass on foliage, or they can be obtained using a hemispherical photo.
L161 The stomatal conductance of the leaf was derived from the Jarvis parameterization. How was the tree (bulk) stomatal conductance calculated? Did you recalculate the leaf to surface big-leaf stomatal conductance in your study?
L175 What about estimating the intercepted evaporation (i.e., the rainwater that was intercepted)?
L188 big-leaf approach? for individual tree?
I see there are a variety of different approaches to evapotranspiration simulation in your model frameworks. It is actually necessary to proceed with great caution when synthesizing models that differ in their approach to vegetation cover structure and physiology. Directly exchanging parameters can lead to significant errors in flux estimates.
L195 Once again, it’s not entirely clear what the authors mean by the terms “evaporation” and “evapotranspiration.” Generally, they refer to the same thing - they’re almost synonyms. Evapotranspiration, or total evaporation, is the amount of water that is released into the atmosphere as vapor as a result of transpiration and physical evaporation from the soil and the moist parts of vegetation.
Or perhaps this was calculated based on the difference in measurements from various lysimeters... But this is a relatively rough estimate... The direct extrapolation of moisture fluxes (i.e., soil evaporation) between lysimeters with different vegetation types can lead to serious errors in attributing the contribution of soil to total evaporation, for example. The same applies to grass cover....
L196-197 This looks like a discussion
L198 How was transpiration actually measured in your study? For example, was it calculated based on the soil surface's contribution to evaporation?
L220 It sounds like an advertisement for a software product... given the assumptions made in the model... such claims aren't very trusted. You should avoid making such statements in the article.
L231 The Priestley-Taylor equation is used to calculate potential evaporation (not potential evapotranspiration). Evaporation from a water surface or a wet grass surface.
L237 PT constant may vary depending on the LAI and the amount of water on the leaf surfaces.
Fig 8 These are very high transpiration rates. They are 3–5 times higher than the daily evaporation rate of a humid tropical forest.
L268 The question still remains about how to determine transpiration and how accurate that determination is.
L273 What is the statistical significance of R2 estimates?
L283 This was under conditions in a single lysimeter. How can this data be extrapolated to a real vegetation canopy?
L301 Once again, the Priestley-Taylor equation was considered a simplification of the Penman equation for a water-covered or wet surface... under conditions of minimal advection... The air entering your lysimeter has properties that are quite different from those formed near the vegetation in the lysimeter. Drier air can lead to increased transpiration and physical evaporation. So, using lysimeters for such tasks is a very risky business.
L320 Such a combination of modeling approaches should not be treated in a purely formal manner; instead, one must take into account the purpose of the models, the assumptions underlying each modeling approach, and the potential errors associated with those assumptions.
L338 Based on the results of measurements in four lysimeters and the modeling of flow within them, this is undoubtedly a very bold claim. It’s hard to believe.
L354 Heatwaves seem to have been left out of the article… The results make no mention of changes in evaporation due to high temperatures. Or was it just a passing remark?
L366 The results for the grassy plots and the lone tree hardly support such promising claims.

Reply
Citation: https://doi.org/10.5194/egusphere-2026-2385-RC1
RC2:
'Comment on egusphere-2026-2385', Anonymous Referee #2, 08 Sep 2026 reply
The manuscript “A flexible open-source modular framework for ecohydrological modelling: Application and validation of GEOSPACE-1D” is a following work of the model description article by D’ Amato et al. (2025a). It demonstrates the first application of the new model using real-world data and validates the model compared with observations. The scope of the manuscript is clear but the structure is not well organized. In addition, the comparison with other existing models can be improved with more analysis.
Major comments:
Structure: it is very strange to read the result section. In this section, the texts only cite the figures and tables but don’t describe the results. Some brief result descriptions are found in the discussion section and the authors refer to each objective clearly in the discussion section, not in the result section. As a result, the objectives of this study are achieved in the discussion section, not in the result section. However, in my opinion, the result section is the major part to describe results and achieve the objectives while the discussion section supplements the results with additional comparison or discussion. Hence, the result description should be extended substantially in the result section and the structure of the result section should comply with the objectives. Besides the result and discussion section, the method section is not complete. After reading section 2, I didn’t know that there would be a comparison between ET models, and the first two paragraphs in section 3.2 fit better in the method section. In addition, it would be good to have an overview of all modelling work at the beginning of the method section.

Section 4 (discussion) contains the results of this study, which should be condensed in the revised manuscript. In section 4.1, it would be more appropriate to compare the model performance with other existing models and discuss the model advantages and limitations in terms of model performance. The authors compared the modularity and specific processes with other models in section 4.2 and 4.3, which is good. However, model performance is also an important criterion to evaluate a new model compared with the benchmark of other existing models. In addition, performance comparison gives more value than modularity in this study because only the ET methods can be selected by users in the present version of framework.

Other comments:
Line 67: please give the full name of OMS3.

Somewhere in the introduction, it would be good to point out that the ecohydrological model discussed in this study is for point scale. There are other ecohydrological models for catchment scale or larger scales. The status and focus of these models are different from the models for the point scale.

Line 196-197: can you please explain why your model result (2%) is lower than the observations (5-10%)?

Line 201-205: you mentioned Fig. 5 before Fig. 4. Consider adjusting the sequence of the sentences.

Line 204-205: in the text, it says that the simulation was compared with data obtained from weight measurements, but I cannot find the observation in Fig. 4.

Figure 8: when I first saw this figure, I was wondering how the model simulates it for each soil layer. Then I found some explanation in section 4.3. It would be good to have more description and explanation on this figure in the result section.

Reply
Citation: https://doi.org/10.5194/egusphere-2026-2385-RC2

Concetta D'Amato, Paolo Benettin, Andrea Rinaldo, and Riccardo Rigon

Supplement

https://doi.org/10.5194/egusphere-2026-2385-supplement

Data sets

Simulation data for: A flexible open-source modular framework for ecohydrological modeling: Application and validation of GEOSPACE-1D Concetta D'Amato et al. https://doi.org/10.5281/zenodo.19773216

Model code and software

GEOSPACE-1D Concetta D'Amato et al. https://doi.org/10.5281/zenodo.16779108

Concetta D'Amato, Paolo Benettin, Andrea Rinaldo, and Riccardo Rigon

Viewed

Total article views: 246 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
158	65	23	246	36	23	28

HTML: 158
PDF: 65
XML: 23
Total: 246
Supplement: 36
BibTeX: 23
EndNote: 28

Views and downloads (calculated since 28 Jul 2026)

Month	HTML	PDF	XML	Total
Jul 2026	91	27	6	124
Aug 2026	58	35	17	110
Sep 2026	9	3	0	12

Cumulative views and downloads (calculated since 28 Jul 2026)

Month	HTML	PDF	XML	Total
Jul 2026	91	27	6	124
Aug 2026	58	35	17	110
Sep 2026	9	3	0	12

Viewed (geographical distribution)

Total article views: 214 (including HTML, PDF, and XML) Thereof 214 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 08 Sep 2026

Short summary

Here we validate GEOSPACE, a flexible open-source tool for simulating water exchange between soil, plants, and the atmosphere. Using data from a lysimeter experiment with four vegetation covers, we show that GEOSPACE accurately reproduces soil water dynamics, evapotranspiration, and root water uptake. Its modular design allows selecting among alternative process formulations, making it a reliable tool for ecohydrological research.


Total:	0
HTML:	0
PDF:	0
XML:	0