There are some minor comments on word choice and the structure of the title, but the influence of biology needs at least to be identified, if not included by deduction of some of the data. (specific details are in the associated word file.

This paper describes an update to the established P-LSH ET model, incorporating a new soil moisture constraint and model calibration to improve global performance. Model validation is assessed for global wet and dry climate zones against flux tower measurements and independent watershed level ET estimates to document relative model improvements in relation to the current model (v1) and other global ET records (GLEAM, Penman-Monteith-Leuning). Overall, the results demonstrate clear and meaningful P-LSHv2 ET performance improvement relative to ET observations and the other models. The addition of a model soil moisture constraint more strongly enhances model accuracy in dry climate zones, while providing a more realistic representation of environmental controls on ET trends. The paper is well written and comprehensive, with multiple levels of evidence to support the findings and conclusions, and well illustrated figures and tabular summaries that give the reader a clear understanding of model improvements. I therefore consider the paper to be suitable for publication once the authors address the following the following minor issues.

The authors state that tower measurements, including surface soil moisture, are used to drive and evaluate the ET algorithm (Ln 236-238). However, it’s unclear whether the tower level performance assessment (Section 4.2) is based on model ET simulations derived from local tower meteorological measurements or GLDAS inputs. Additional explanation is needed here.

The authors compare model ET performance and soil moisture sensitivity between wet and dry climate zones defined from a global climate aridity index (AI). However, the simple climate AI partitioning groups energy-limited cold land areas, including northern taiga and tundra, into the dry climate category (e.g. Fig. 2) even though these areas have generally wet soils with minimal soil moisture constraints during the short summer growing season. Thus, tundra is grouped with other GRS and OSH dominant land covers even though these other areas may represent much warmer-drier climate zones (e.g. sub-tropical Africa & western CONUS drylands).  Failure to distinguish energy limited zones may contribute to the excessive model soil moisture constraint indicated in tundra (Fig. 5) and the corresponding relative ET model underestimation in this region (e.g. Fig. 14). Additional discussion is needed along these lines.

Ln 42: Text should be modified similar to: MODIS data do not cover the pre-2000 period and are of insufficient length to represent longer-term interannual variability and trends, and attribution analysis in ET. The revised statement more correctly acknowledges the longer MOD16 ET record available from the NASA Terra satellite. Moreover, while the MODIS record is too short to capture climate “normals” that would require a minimum 30-year span, the data record does represent a comprehensive (500m, 8-day) multi-decadal global operational satellite ET record, which has been used to evaluate more recent interannual variability and trends (e.g. Hall et al. 2023, Roman et al. 2024).  

Ln 14: “curcial” should be “crucial”.

Ln 236: Please define what is meant by “surface” soil moisture here; e.g., 0-5cm depth?

Ln 497: Include supporting citation on the noted net radiation decline since 2016.

Citations mentioned above:

Hall, D.K., J.S. Kimball, R. Larson, N.E. DiGirolamo, K.A. Casey, and G. Hulley, 2023. Intensified warming and aridity accelerate terminal lake desiccation in the Great Basin of the western United States. Earth and Space Science, 10, 1, DOI:10.1029/2022EA002630.

Roman, M., C. Justice, I. Paynter, et al., 2024. Continuity between NASA MODIS Collection 6.1 and VIIRS Collection 2 products. Remote Sensing of Environment, 103, 113963, https://doi.org/10.1016/j.rse.2023.113963