It is becoming increasingly common to observe researchers extracting data from public databases on the internet and applying it to open-source models without first understanding how the measurements were made, knowing their original purpose, and, worse, disregarding previous work with the data. They believe that the data is merely waiting for them to use it and that no one has ever used it.

In this context, Yanyan Cheng and her collaborators used publicly available data from three eddy covariance flux systems, two located in natural forests and one in a city, to evaluate the performance of the Noah with Multi-Parameterizations (Noah-MP) land surface model to calibrate it and improve its representativeness in tropical settings. Regarding the model evaluation for an urban setting, which is why I am writing this comment, it appears that the authors' lack of knowledge about eddy covariance flux measurements and the characteristics of the evaluated site (Telok Kurau, Singapore) led them to draw conclusions out of context that contradict field observations and focused surveys.

Before discussing such contradictions, it is worth noting that the authors did not undertake a literature review of previous work on the topic using the same database. A quick Google Scholar search would have returned seven previous studies that evaluated the performance of different land surface models to simulate energy partitioning using the same heat flux data from Telok Kurau (Demuzere et al., 2017; Liu et al., 2017; Harshan et al., 2018; Simón‐Moral et al., 2020; Meili et al., 2020, 2021; Sánchez et al., 2022).

Cheng et al. concluded that the Noah-MP land surface model for urban settings, which integrates a single-layer urban canopy model (SLUCM), substantially underestimates latent heat. According to them, this discrepancy arises from the simplified approach to urban hydrology in SLUCM and errors in the reported land fraction of impervious surfaces, such as buildings, roads, and sidewalks, by the researchers who conducted the flux measurements.

The model's performance was evaluated using a brute-force method with nine predetermined parameters. Each parameter was modified incrementally, while the remaining parameters were left fixed at default values. The selected values ​​were those that, when used by the model, yielded the smallest difference with the observed data. There may not be much to criticize here, however, all of the evaluated parameters were related to vegetation and soil, and none to the built-up area features. This is despite the fact that the aforementioned studies clearly indicate the need to improve the characterization of thermal and radiative properties of the different urban surfaces, as well as to employ correct representations of urban morphology, in addition to using better hydrological modules and vegetation parameterizations.

The finding that simplified urban hydrology treatments yield poor results in latent heat modeling is correct. However, this is not a novel result. The authors of this work should have reviewed previous efforts to improve such treatments for Singapore’s case, using Telok Kurau flux data as a reference (Meili et al., 2020, 2021; Jongen et al., 2022). It would have been valuable if Cheng et al. had compared their results with those of these authors.

The purported error found by Cheng et al. about the fraction of land covered by impermeable surfaces is objectionable to those researchers who spent days walking streets and alleys to validate information obtained from maps and aerial photographs of Telok Kurau. Refer to Velasco et al. (2013). Cheng and colleagues used 100 m resolution images, for which no reference is provided, to arrive at such a conclusion, in contrast to the house-to-house surveys of green areas and trees carried out by the aforementioned researchers.

The authors' conclusions were derived from a simplistic analysis that only considered the annual averages of latent heat and sensible heat obtained from the model. A thorough assessment cannot be achieved just by examining annual averages; it is evident that evaluating the model's performance throughout the day under different weather conditions and climatological seasons is essential to identify which factors and aspects need further investigation.

In the same context, it is unclear why the authors limited the study to the model's outputs for sensible and latent heat, ignoring the other components of the energy balance. At the very least, they should have tested the model's ability to reproduce incoming and outgoing short- and long-wave radiation, data for which is available in the database used as a reference.

The model's ability to reproduce soil moisture dynamics was appropriately mentioned on several occasions throughout the text, but it was neglected when analyzing the Telok Kurau results. The database that the authors used does not include soil moisture data. That database only covers the first year of measurements, when soil measurements were not yet available. If the authors had conducted a proper literature review, as mentioned above, they would have discovered soil temperature and water content data collected as part of soil respiration measurements conducted under the same flux measurements initiative in an urban district in Singapore (Velasco et al., 2021). They would also have discovered valuable data on soil properties that would have helped in improving the model's performance.

Unfortunately, this is not the first time that heat flux data from Telok Kuaru has been used to run numerical models without in-depth analysis that would contribute to improving knowledge about urban climatology in the tropics. The results of such studies are not reliable because they are based on simplistic assumptions, incorrect interpretations, and flawed methodologies. Modelers are invited to team up with researchers who do field measurements to first understand the phenomenon before validating and adjusting their numerical models (Velasco, 2018).

References

Demuzere, M., Harshan, S., Järvi, L., Roth, M., Grimmond, C.S.B., Masson, V., Oleson, K.W., Velasco, E. and Wouters, H., 2017. Impact of urban canopy models and external parameters on the modelled urban energy balance in a tropical city. Quarterly Journal of the Royal Meteorological Society, 143(704), 1581-1596. https://doi.org/10.1002/qj.3028.

Harshan, S., Roth, M., Velasco, E. and Demuzere, M., 2018. Evaluation of an urban land surface scheme over a tropical suburban neighborhood. Theoretical and Applied Climatology, 133(3), 867-886. https://doi.org/10.1007/s00704-017-2221-7.

Jongen, H.J., Steeneveld, G.J., Beringer, J., Christen, A., Chrysoulakis, N., Fortuniak, K., Hong, J., Hong, J.W., Jacobs, C.M., Järvi, L. and Meier, F., 2022. Urban water storage capacity inferred from observed evapotranspiration recession. Geophysical Research Letters, 49(3), e2021GL096069. https://doi.org/10.1029/2021GL096069.

Liu, X., Li, X.X., Harshan, S., Roth, M. and Velasco, E., 2017. Evaluation of an urban canopy model in a tropical city: the role of tree evapotranspiration. Environmental Research Letters, 12(9), 094008. https://doi.org/10.1088/1748-9326/aa7ee7.

Meili, N., Manoli, G., Burlando, P., Bou-Zeid, E., Chow, W.T., Coutts, A.M., Daly, E., Nice, K.A., Roth, M., Tapper, N.J. and Velasco, E., 2020. An urban ecohydrological model to quantify the effect of vegetation on urban climate and hydrology (UT&C v1. 0). Geoscientific Model Development, 13(1), 335-362. https://doi.org/10.5194/gmd-13-335-2020.

Meili, N., Manoli, G., Burlando, P., Carmeliet, J., Chow, W.T., Coutts, A.M., Roth, M., Velasco, E., Vivoni, E.R. and Fatichi, S., 2021. Tree effects on urban microclimate: Diurnal, seasonal, and climatic temperature differences explained by separating radiation, evapotranspiration, and roughness effects. Urban Forestry & Urban Greening, 58, 126970. https://doi.org/10.1016/j.ufug.2020.126970.

Sánchez, B., Roth, M., Simón-Moral, A., Martilli, A. and Velasco, E., 2021. Assessment of a meteorological mesoscale model's capability to simulate intra-urban thermal variability in a tropical city. Urban Climate, 40, 101006. https://doi.org/10.1016/j.uclim.2021.101006.

Simón‐Moral, A., Dipankar, A., Roth, M., Sánchez, C., Velasco, E. and Huang, X.Y., 2020. Application of MORUSES single‐layer urban canopy model in a tropical city: Results from Singapore. Quarterly Journal of the Royal Meteorological Society, 146(727), 576-597. https://doi.org/10.1002/qj.3694.

Velasco, E., Roth, M., Tan, S.H., Quak, M., Nabarro, S.D.A. and Norford, L., 2013. The role of vegetation in the CO 2 flux from a tropical urban neighbourhood. Atmospheric Chemistry and Physics, 13(20), 10185-10202. https://doi.org/10.5194/acp-13-10185-2013.

Velasco, E., 2018. Go to field, look around, measure and then run models. Urban climate, 24, 231-236. https://doi.org/10.1016/j.uclim.2018.04.001.

Velasco, E., Segovia, E., Choong, A.M., Lim, B.K. and Vargas, R., 2021. Carbon dioxide dynamics in a residential lawn of a tropical city. Journal of Environmental Management, 280, 111752. https://doi.org/10.1016/j.jenvman.2020.111752.