The authors present an analysis of two primary controls on slope stability in Northern Bavaria, Germany: geology and vegetation. The topic is important to protect life, property and infrastructure locally. The results also present possible contributions to our understanding of slope stability that would be applicable elsewhere.

I have two major comments that I believe will help to improve the paper.

1. Motivation/what’s new? It is known that slope stability is influenced by both geologic and vegetation controls, the authors could better identify the knowledge gap and clearly illustrate how their study fills this knowledge gap. Specifically, the abstract jumps straight to the actions performed without motivating/asking a clear research question. The introduction only reaches a clear motivation towards the end – focusing on the vegetative controls of landsliding in shallow and low angled hillslopes. Is this the key knowledge gap (what controls shallow and deep landsliding on low angle slopes?) This should be made clearer in the abstract/intro to justify the study and used to better explain results in the discussion section.

2. Clearly define the two types of landslides and proposed controls. As written, the two types of landslides (deep and shallow) and the specific controls the authors investigate (geologic properties, vegetative root strength, respectively) are not clearly presented. Whereas some general background is appropriate in the introduction, there should be a sharpening of focus that clearly defines landslides of different depth, and the respective controls investigated in this study. It is initially unclear why the authors investigated tree root strength when the majority of landslides were all deeper than 2 m where no roots were found. The discussion of the tree root data similarly lacks focus and a take-home point because it is not clear why these data are included in the study.

Additionally, there are useful tree data (DBH, age, stand density) that would add to the study.

Specific comments:

Abstract         opening sentence is true, but what is knowledge gap paper attempts to fill? Clearly identify two types of landslides (shallow and deep) and the knowledge gaps on what controls these types of landslide on shallow slopes.

10                    ‘rooted area’ is supposed to be root area ratio?

14                    how do high pore pressures develop due to geologic conditions? Do you mean due to hydrologic conditions? Or increased pore pressure along low permeability boundary?

20                   final 1-2 sentences of abstract would be stronger if they followed the ‘two types of landslides’ outlined above and distinguished how the mechanisms controlling slope stability are different in each (geology – forests)

Intro                why does the introduction start with a summary of sedimentary rocks? The paper is focused on geologic/vegetation controls on slope stability and as a reader I expect the principal topic to be one of those listed in the title.

43                    Also Schmidt, Roering, Ziemer, Terwilliger & Waldron.

55                    also Ziemer (https://www.fs.usda.gov/treesearch/pubs/8693)

61-62  Ziemer and Terwilliger and Waldron (https://pubs.geoscienceworld.org/gsa/gsabulletin/article-abstract/103/6/775/182576/Effects-of-root-reinforcement-on-soil-slip)

40-62              tighten language as there is some repetition

62-67              Good motivation for study – but should also clearly distinguish between shallow and deep and the controls of geology and vegetation. This  reasoning should be in abstract

117                 cite RMS from previous investigations and briefly summarize what was found

127                 dead/cut trees were excluded, but dead/cut trees continue to provide strength until they rot away. See Ziemer:

140                 only 1 species (Scots Pine) was measured in this study and roots .

195                 should g_s represent the saturated bulk density of the soil?

215                 goal to ‘test if root cohesion would be sufficient to stabilize the soil’ of shallow landslides should be mentioned in the introduction.

Fig 2 legend  ‘transekt’ should be ‘transect’, since Rhätolias-Feuerletten boundary is so important, consider changing color to make it stand out.

Figure 4         explain in legend the criteria used to identify failure plane boundary – I had to go back and search to find line 122 about Figure S3 and the identified shear plane depth

Figure 5         legend should include scarps, caption should tell reader locations of panels a, b, c, referring to the maps in figure 2. Fürstenanger is the only location with a spatial pattern in species – with Scots Pine concentrated near headscarp. Is this important?

293                 This sentence is not clear. What does 0.19 refer to?

Figure 6         why do the authors plot root diameter against tensile strength in MPa instead of against tensile force at failure? I recommend including the previously published data to show the stated similarities with other species

317                 unclear sentence, instead of stating ‘get’ unstable, I suggest ‘become’ unstable or fall below FoS of 1.

Figure 7         why are there no data for 0.5 cm depth in the European Beech? And, are the authors sure there are no roots deeper than 0.5 m that would add tensile strength to the soil?

Fig 8 caption ‘We assume an angle of internal friction of 8.4°. We vary cohesion between…

334                 ‘All these locations are underlain by…’

Figure 9         I like the figure, the different colors are hard to see.

349                 sentence structure ‘Of the 125 observed landslides, 95% occurred at the R-F boundary…’

360                 ‘In between the lower high-resistivity cells…’

362                 ‘The lower part of the landslide was characterized by flat topography, low-resistivity areas…’

392                 unclear what this sentence is trying to communicate ‘Water can move laterally…’

444                 what effect might lateral root cohesion have on such a broad landslide?

In this paper the authors analyzed the geologic and vegetation control on deep and shallow landslides in the scarplands of Southern Germany. Their objective was to understand geologic conditions and the role forest management might play in helping slope stability, especially when these slopes face changing vegetation and hydrologic conditions in a changing climate.

The interplay between vegetation and sub-surface conditions to better understand regional landslide hazards is a long-studied and important topic of interest. This paper presents many useful datasets from Northern Bavaria (e.g. ERT data for several large landslides, root strength from Scots Pine, detailed landslide mapping, and soil strength properties). However, the paper would greatly improve with a more clearly defined hypotheses and motivation, which are then justified by the data and the discussion. As written, this is almost two papers: 1) about rooting strength and controls on recent shallow landsliding, and 2) about modelling the unique scarpland geology that leds to pervasive large, deep landslides that are now seemingly stable.

A few major comments:

Abstract – This does not flow well and lacks a clear framing of the authors’ hypotheses. The abstract mentions many methods but it is not clear how all these methods intersect.

Introduction – The end of the introduction seems to describe the overall motivation (looking at role of vegetation on low slopes with landslides) that should be made clear much earlier and returned to throughout. For this section, I suggest shortening the forestry summary and expanding on the scarpland morphology and differences between shallow/deep landslides to introduce your hypotheses. With a clear motivation and hypotheses, the methods can then be justified.

Additionally, the paper would be strengthened by describing the different timescales of interest (e.g. recent land-use/forestry management and shallow landslides, longer-term climate shifts possibly leading to reactivation or initiation of deep landslides, timing of original (now “fossil”) landslides).

Discussion – As written, this section seems to raise more questions than it answers. For example, why such a focus on tree roots when the landslides analyzed are much much deeper than the rooting depth? Or, for stability modelling, why not model the geologic and hydrologic conditions needed for the original failures to test the importance of scarpland geology (e.g. Perkins et al., 2017)? This section could be strengthened by clearly describing: what is novel from this study, how low angle hillslopes compare to steep vegetated hillslopes, and intersection of geology and deep landslides with vegetation and shallow landslides.

Specific comments:

ABSTRACT

13: does low slope refer to the geologic contact or hillslope angle (likely hillslope, but confusing after talking about dipping angles)? Does high pore pressure refer to a measured, modelled, or inferred point?

14: geologic conditions should be hydrologic

18: why is European beech specifically helpful to landslide stability? If making this recommendation, include results leading to this conclusion.

INTRODUCTION:

24: “sedimentary origin” should be specified as “scarplands”

24-25: jump into very detailed geology and landslide classifications without setting up overall objectives

36: Using only a depth cutoff is a little misleading, typically shallow=landslide rooted in soil and deep=landslide rooted in rock.

50: change “therefore” to “and”

63: If this is the motivational framework, introduce early on and include in abstract. Frame your hypotheses or research questions based on this motivation. For example, in low slope scarplands do you expect vegetation to have more or less influence than steep mountains?

METHODS:

125-126: Studies show that significant root strength can persist for up to ~10 years (e.g. Ammann et al., 2009-Norway Spruce). What is age of trees vs. age of landslides?

127: insert “…selected 15 individual free-standing…”

128: remove “at 15 trees”

182: how was this material collected? Does this include bedrock material? Or just landslide material?

RESULTS:

119: what do you mean by “follow no expositional pattern”?

DISCUSSION:

352-353: why not try to model this instead? Similar to Perkins et al., 2017

376-377: Why such a focus on trees then? Why not focus on the mechanics of these deep landslides using stability modeling and ERT results?

378: Where on the landslide? Just deposit/mobile material?

390-395: There are a lot of hypotheses presented here without much to back-up conclusions. Seems like the more interesting modeling problem if the authors want to understand the role of scarpland geology

426-427: what are the depths of shallow landslides occurring on these larger landslides?

 FIGURES:

NEW FIG(s): Suggest adding schematic of geology and/or typical slope profile, and photo of typical soil pit with roots.

FIG 5: missing symbology

Reviewer References:

Ammann, Martin & Böll, Albert & Rickli, Christian & Speck, Thomas & Holdenrieder, Ottmar. (2009). Significance of tree root decomposition for shallow landslides. For. Snow Landsc. Res. 82. 79-94.

Dhakal, A. S., & Sidle, R. C. (2003). Long‐term modelling of landslides for different forest management practices. Earth Surface Processes and Landforms: The Journal of the British Geomorphological Research Group, 28(8), 853-868.

Perkins, J. P., Reid, M. E., & Schmidt, K. M. (2017). Control of landslide volume and hazard by glacial stratigraphic architecture, northwest Washington State, USA. Geology, 45(12), 1139-1142. https://doi.org/10.1130/G39691.1