Please see the uploaded file of "Response to Referee 1".

Please see the uploaded file of "Response to Referee 1".

We thank you for using our dataset and giving the above positive comments and nice suggestions. We have carefully considered your comments and have responded as follows.

First, this issue is mainly about the set of values for years larger than 31. Specifically, by classifying the pixels with values >31 into one category and then displaying forest age, the problem of spatial discontinuity you mentioned can be resolved. In the next version of the dataset, we will share it after resolving this issue.

Second, as you mentioned, there are some CCDC-family algorithms, which may be more suitable for turbulence monitoring and/or classification of land cover. We use CCDC to track the breakpoints of Landsats time series, for the three main reasons: (1) CCDC is the classical algorithm in turbulence detection. We considered using it to estimate the age of forest and the results in this paper already demonstrated that it is an acceptable choice; (2) other CCDC-family algorithms might be sensitive to detect breakpoints. In future research, we will examine whether the use of other versions will necessarily further increase the accuracy; (3) GEE cloud platform provided the basic CCDC in its official algorithm libraries, which is more suitable for large-scale mapping than other CCDC-family algorithms currently.

Third, the differences you mentioned may come from three parts: (1) Differences in statistical time. The ninth national forest inventory (NFI) of China is covering the period 2014–2018, however, our dataset is covering the period 1990–2020; (2) Differences in the methods of forest age statistics. The NFI classified the forest into five forest classes (such as young, mid-aged, near-mature, mature, and over-mature forests), and the age range of each class is vary with tree types. For example, the natural Pinus massoniana Lamb less than 20 years old belongs to the young stage, while the natural Abies fabri less than 40 years old also belongs to the young forest. However, we definite the 1-31-year-old forests as young forests; (3) Mapping error. As mentioned in Section 5.4 of the manuscript, there are still uncertainties in estimating the age of forests.

Thank you very much for your positive comments and suggestions to improve our manuscript. We have carefully considered your comments and revised the paper accordingly. We have modified the sentence in Line 518.