the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Connecting CSR theory and LPJmL 5.3 to assess the role of environmental conditions, management and functional diversity for grassland ecosystem functions
Stephen Björn Wirth
Arne Poyda
Friedhelm Taube
Britta Tietjen
Christoph Müller
Kirsten Thonicke
Anja Linstädter
Kai Behn
Susanne Rolinski
Abstract. Forage supply and soil organic carbon storage are two important ecosystem functions of permanent grasslands, which are determined by climatic conditions, management and functional diversity. However, functional diversity is not independent of climate and management, and it is important to understand the role of functional diversity and these dependencies for ecosystem functions of permanent grasslands. Especially since functional diversity may play a key role in mediating impacts of changing conditions. Large-scale ecosystem models are used to assess ecosystem functions within a consistent framework for multiple climate and management scenarios. However, large-scale models of permanent grasslands rarely consider functional diversity. We implemented a representation of functional diversity based on the CSR theory and the global spectrum of plant form and function into the LPJmL dynamic global vegetation model forming LPJmL-CSR. Using a Bayesian calibration method, we parameterised new plant functional types and used these to assess forage supply, soil organic carbon storage and community composition of three permanent grassland sites. These are a temperate grassland, a hot and a cold steppe for which we simulated several management scenarios with different defoliation intensities and resource limitations. LPJmL-CSR captured the grassland dynamics well under observed conditions and showed improved results for forage supply and/or SOC compared to LPJmL 5.3 at three grassland sites. Furthermore, LPJmL-CSR was able to reproduce the trade-offs associated with the global spectrum of plant form and function and similar strategies emerged independent of the site specific conditions (e.g. the C- and R-PFTs were more resource exploitative than S-PFTs). Under different resource limitations, we observed a shift of the community composition. At the hot steppe for example, irrigation led to a more balanced community composition with similar C-, S- and R-PFT shares of above-ground biomass. Our results show, that LPJmL-CSR allows for explicit analysis of the adaptation of grassland vegetation to changing conditions while explicitly considering functional diversity. The implemented mechanisms and trade-offs are universally applicable paving the way for large-scale application. Applying LPJmL-CSR for different climate change and functional diversity scenarios may generate a range of future grassland productivity.
- Preprint
(761 KB) -
Supplement
(702 KB) - BibTeX
- EndNote
Stephen Björn Wirth et al.
Status: final response (author comments only)
-
RC1: 'Comment on bg-2023-55', Anonymous Referee #1, 10 May 2023
General evaluation of the research paper
The paper presented by the authors addresses a very relevant and important topic in the field of DGVM model development. For far too long, the representation of grasses and the herbaceous layer have been given far too little focus in most DGVMs with respect to structural and functional diversity. Only recently, development of more detailed grass layer representations in DGVMs are starting to emerge but compared to tree-layer representation this work is still at a comparatively early stage of development. Grassland ecosystems and savannas cover a substantial fraction of the land surface and provide important ecosystem functions and services to a multitude of people while simultaneously being threatened by the effects of climate change and resource over-exploitation. Therefore, developing vegetation models that are capable of representing within-grass layer dynamics, diversity and processes is crucial to assess the impact of different management strategies and environmental change. I therefore deem the paper a relevant and important scientific contribution.
The CSR theory is a widely known concept and therefore a valid approach to implement functional diversity and trade-offs within the herbaceous layer of the model. One may question whether the implementation in its current form using a Bayesian calibration method to parameterize the new PFTs for three specific sites can be generalized for large-scale application, but in the given context of the study, the approach seems sound and justified to me. The shown results in many cases match ecological expectations and improve results compared to the old model version, further corroborating the chosen approach.
The paper is well-written and clearly structured. I therefore recommend publication pending minor revisions and clarifications detailed below.
Detailed comments
Introduction:
line(s) 36/37: You might also add the role of atmospheric CO2-concentration. CO2-fertilization effects can shift the competitive balance in grassland communities in locations where both C3 and C4 grasses are present.
line(s) 42: “high temperatures can lead to an increase of microbial decomposition”. Only in combination with sufficient moisture. In arid regions, decomposition comes more or less to a stand-still during the dry season due to the water limitation that affects the microbial community. Rains at the beginning of the wet season then lead to peak emissions when microbial decomposition picks up again.
line(s) 44/45 "...may be beneficial for grassland productivity depending on its intensity". Maybe add: “by removing moribund plant material and triggering growth (over-)compensation.”
line(s) 49: “for the species” – “for the functional types”. I’d rather consistently keep the focus on functional types.
line(s) 52: “indirectly through alterations of the resource limitations” – add: “…that can cause shifts in the competitive balance between functional types”.
Methods
line(s) 105: “hot-steppe pasture in South Africa”: this is a somehow unusual terminology / vegetation classification. The Syferkuil site usually is referred to as savanna rangeland in other publications.
line(s) 107/108: That means no tests of fertilizer X defoliation intensity combinations? That could be another interesting experiment to add, at least for the simulations.
line(s) 115/116: Are the trait values you use to describe the strategies from within a continuous range, or discrete fixed values? For example, if you use SLA as a trait to distinguish between acquisitive and conservative strategies, then you will automatically cover the extremes as well as in-betweens if you allow SLA to be a continuous trait that can range between a minimum and maximum value (see, e.g., Scheiter et al., 2013, Langan et al., 2017).
line(s) 120 “Overview of managed grasslands in LPJmL” – “Overview of managed grassland representations in LPJmL” seems a more fitting title for this section.
line(s) 123/124: one polar, one temperate and one tropical grass: C4-type photosynthesis for the tropical grass? Knowing classic LPJ, I deem it likely that this is the case, but good to mention explicitly.
line(s) 130/131: (no water limitation, ref). – forgot to add the actual reference here.
Table 1: Forage supply [MgDM ha-1]: Terminology not entirely clear: Peak standing biomass? Annual withdrawal quantity (through mowing / grazing)? What is the temporal reference frame – annual?
line(s) 166-168: Does this new scheme also account for root biomass distribution in different soil layers, and therefore varying water availability between different soil layers? So that the total water uptake is the biomass-weighted uptake sum across soil layers? Or is it simpler than that?
line(s) 186: I suppose that means that SLA as a trait is a PFT-specific constant? I.e., it cannot vary over the lifetime of individual, or between different individuals of the same PFT?
line(s) 191/192: Does LPJmL distinguish between forbs and grasses, and if so, how is this implemented? And for grasses: does it distinguish between C3 and C4 photosynthetic pathway? Is age-mortality the only reason for mortality, or are there other causes implemented as well (e.g., due to negative annual C-balance, due to water stress, due to fire, etc.)?
line(s) 193: “a biomass increase of the average individual dependent on the available area” – rephrase? “the area-specific biomass increase of the average individual”
section 2.3.3: general question on mortality: does the model distinguish between annual and perennial herbaceous PFTs? I.e., do you have a PFT with enforced death after one growing season? Enforcing annual types should implicitly strongly select for fast resource acquisition at the expense of durable structural components, and a strong focus on reproductive performance (see, e.g., Pfeiffer et al., 2019).
line(s) 197: “we retained the approach of establishing saplings instead of seeds” – I assume that refers to the tree PFTs? A bit unusual to refer to establishing grasses or herbs as “saplings”. I assume that you must have excluded tree PFTs from the simulations of the grassland sites, allowing grasses/forbs only? Otherwise, it is likely that a forest type or savanna type would have established as potential natural vegetation at least at the German and South African sites. You should add the information of how you handled the tree component of the model in the section where you describe your simulation protocol. Also clarify how establishment is done specifically for the grasses / herbaceous layer.
line(s) 199/200: So just to make clear that I understand correctly: the average individuals are clones, i.e., all of the same PFT, but you introduced the clone-concept to be able to account for PFT-specific reproduction aspects, such as seed numbers, germination rates, and seedling survival probability? If so, you should make it clearer than it is currently. It goes in the direction of the problems faced by models that simulate actual, true individual plants and their reproduction and establishment.
line(s) 203: age-dependent mortality: hard set (at a specific age), or based on an age-dependent likelihood? And: the age-dependency differs between the different strategy types? And what is the allowed maximum number of average individuals, and the maximum number of grass-layer PFTs that can now coexist within one grid cell?
line(s) 205/206: “It can be assumed that few individuals that maintain a high cover and biomass must be larger…” – I assume all individuals that are part of one PFT have the same size and biomass, given that you are still using the average individual concept? So, adding new young individuals will lower the size and decrease the age of all clone individuals within the PFT due to the averaging. But this implies that a strongly reproduction-oriented PFT strategy would automatically have a smaller average individual size, a young average age, and a larger number of clone individuals representing the PFT. This has implications for the age-dependent mortality, as highly reproductive strategy types are then less likely to reach the age where age-dependent mortality hits. Did you consider this aspect?
Table 2: Maybe add a column that specifies the predominant gradient associated with the parameter. You mention it in the text of this section, but it would be helpful to also have it as a brief overview in the table. I find the distinction between biotic and abiotic dimension a bit arbitrary/confusing with respect of the definition. Referring directly to the respective gradient (stress gradient for biotic, disturbance gradient for abiotic)
would seem more intuitive for me.
Table 2: Hierarchy: How did you determine the hierarchy? Based on your expert assessment?
Table 2: Light extinction coefficient: Independent from SLA, or correlated? High-SLA leaves should have more transmission than low-SLA leaves.
Table 2: Maximum transpiration unit [mm] – if this is to be a rate, then the time part of the unit is missing. [mm/day]?
line(s) 237/238: The root efficiency coefficient does affect the competitiveness between plants (biotic interaction), but it also relates to the stress gradient (abiotic) with respect to water uptake capacity. This is an example illustrating why using “biotic” and “abiotic” as dimensions is maybe not the best way to make the distinction.
line(s) 240/241: The light extinction coefficient describes the fraction of light intercepted by each additional leaf layer, right? As the amount of light that can transmit a leaf layer depends on the thickness of the leaf, one would expect kbeer to be correlated with SLA, which, unlike kbeer, you define as abiotic dimension. It would be good if you sort this out more clearly.
line(s) 241/242: the leaf area index of a sapling represents the offspring size - What do you define as "offspring size"? The height of the offspring, or its starting biomass, or its projected foliar coverage? I'm not sure LAIsap is a good description of offspring size, as its meaning is rather vague without a clearer definition. Whether a seedling/sapling of given leaf biomass has a high or low LAI is a function of its SLA, so LAIsap for a given unit of leaf biomass essentially is nothing else as another way to refer to SLA.
Table 3: Flip order of columns “variable” and “site”, as site is unique and variable is tied to site and non-unique.
line(s) 287/288: “the current representation of some processes within the model” – which processes specifically?
line(s) 299: 390 years - your spin-up duration? Did you add a transient simulation period after the spin-up (how long? For what time-period?). One can only guess based on the time-axis labeling in the figures that follow in the results section. Please specify this with some more detail.
Modelling protocol: What is the temporal resolution the CSR-model version runs on? Monthly, or daily? How do you initialize community composition with respect to present PFTs and shares of PFTs at the beginning of the simulation? Based on the field-based observations? If so, how would you do it in a situation where you did not know the field situation of sites, e.g., for a large-scale or global simulation? (Question for the discussion, I guess).
Results
Figure 1: Please specify temporal reference frame for panels a, d, and g - is it the annual sum (yield), the peak season leaf biomass (leaf biomass), the grazing period duration offtake (grazing offtake)?
General question on all scenarios that included animal grazing: Is preferential grazing, i.e., selection of more palatable over less palatable PFTs, accounted for by the new CSR model version? Unlike mowing or biomass removal by fire that is indiscriminate, biomass removal by herbivores can alter community composition quite substantially, especially under high grazing pressure. If preferential grazing is not yet implemented, this should be added as a limitation in the respective section of the discussion, and could be pointed out as a future need for development.
line(s) 365-368: Ecologically, the shift towards more investment into above-ground biomass (growth (over-)compensation) and towards a more resource-exploitative strategy (construction of “cheaper” leaves with reduced life duration is plausible. However, I do not see right away why the minimum canopy conductance should decrease due to grazing?
line(s) 406/407: How does the relative contribution of the S- and R-PFT to the forage supply compare to their relative abundance or relative contribution to FPC? I.e., did they contribute more or less than could be expected according to their relative abundance within the community?
line(s) 442/443: “In the irrigated scenario, only the S-PFT contributed to forage supply.” - That is a bit surprising? One would expect that irrigation reduces stress resulting from water limitation, therefore opening the community more strongly for the C-PFT.
line(s) 473/474: “…still dominated by the S-PFT.” - Is this a legacy effect from the pre-irrigation time period's community composition? If run long enough without resource limitation (i.e., with irrigation on), would the S-PFT type be replaced by the C-PFT type, and if yes, how long do you expect this would take? Can be part of the discussion, if not already discussed there.
Discussion
General remark: how do you intend to use the CSR-model in the future, if you ideally need an a-priori determination of the ideal PFT parameterization depending on site, community, and management? And how can communities respond to changing management or environmental conditions if the parameterization of the PFTs cannot be dynamically adjusted during the simulation based on a selection mechanim that filters for the best-suited parameterization under the given circumstances?
line(s) 494/495: “IN LPJmL-CSR, growth of the vegetation was faster than in LPJmL 5.2, which led to higher yields for all cuts.” – Elaborate briefly on the causes for the faster growth in the new model version.
line(s) 504: “but selected a livestock density of 1.0 cows ha-1” – use “livestock units” rather than cows (how about steers, heifers, etc.); And: Is this to determine the amount of manure input? The temperate grassland was not grazed but mowed, so livestock density does not make much sense with respect to grazing off-take?
line(s) 506: Briefly describe the processes / mechanisms that lead to increased carbon input to the soil in the CSR-version compared to the old version.
line(s) 526/527: Here finally the information that I was missing in the methods section. You should add this information to the modeling protocol section (that you did exclude the tree PFTs from your site-scale simulations.
line(s) 528/529: You should try to give a reason for the "why" of this, instead of simply repeating the result. For example, an explanation could be that grazing was not the only / the main stress for herbaceous vegetation at this savanna site. The site has a pronounced dry-vs-wet season dynamics, and therefore water limitation as a stress factor, maybe also N-limitation, may be causes for the dominance of the S-type irrespective of the grazing management.
line(s) 540/541: You could test this by specifically allowing no other PFT than the S-type to enforce a monoculture.
line(s) 544/545: Was your simulation time period with irrigation long enough to allow establishment of a new steady state with respect to community composition? In my experience, community composition shifts are one of the slower processes and can take quite a number of years before reaching a new steady-state after a change in forcing has occurred.
line(s) 545/546: “However, periods of drought can induce and additional disturbance.” – Correct, but not in this case, because due to the irrigation you had drought eliminated.
line(s) 549: “LPJmL 5.3 underestimated the observed forage supply…” – I'm not sure about your usage of the term "forage supply" (generally throughout the manuscript) - is forage supply, according to your definition, the potentially available biomass offered by the rangeland, or do you actually rather mean "the amount of feed required by the animals" (which should then be termed as "forage demand”?
line(s) 552/553: I do not understand: how does feed demand change forage supply? Forage supply is a biomass potential offered by the plant community. Increased feed demand, as described here by your correction, should not increase the forage supply of the plant community (unless through growth overcompensation), but rather reduce the supply due to the increased demand from the animal side?
line(s) 554/555: The fact that animal demand could not be met AND above-ground biomass collapsed is a rather clear indication of over-grazing / exceeding of rangeland carrying capacity. In this context, maybe also discuss changes in the PFT community composition, i.e., changes in the prevailing strategy types. It can be expected that such a shift in strategy types occurs under such circumstances.
line(s) 562/563: You did not combine fertilization with irrigation, right? Do you expect that fertilization in combination with irrigation would increase leaf biomass beyond the level reached with irrigation alone?
line(s) 575: “Fertilization had no effect on SOC” – Not surprising, given that fertilization without irrigation did not increase leaf biomass and therefore C-input to the soil.
line(s) 580/581: “it seems that an S-strategy remained advantageous” - Again, I wonder about the turnover time required by the model to let a community transition from one steady-state to a new steady-state.
line(s) 600: And it may be interesting how grass-tree coexistence (typical for savanna sites as the one one in South Africa) will affect grass layer community composition compared to a situation where trees are excluded from the simulation.
line(s) 606/607: “Generally, a change in resource availability does only change the conditions for the establishment of a community but does not directly affect the established vegetation” – Environmental filtering can also affect the established community by increasing mortality for specific strategy types within the community, not only by changing establishment success of given strategy type. Since you seem to have no other mortality causes aside from age-dependent mortality in the model (at least not for the grass layer), you will not see this effect, but it does exist, nonetheless.
line(s) 614: Why are N-fixers not separate PFTs in the model? I'm a bit surprised that they are not.
line(s) 622/623: So the assumption is that grazing is non-preferential, correct? I.e., grazers do not favor one PFT over another, for example based on criteria that characterize palatability / nutrition value. This is a simplification in the model that should be discussed briefly, as herbivores usually do not function the same way as mowing (or fire) that removes biomass indiscriminatingly.
line(s) 624: “tolerance or avoidance” – Avoidance would for example (aside from temporal avoidance) be realized by being unpalatable. As your grazing is non-preferential, being a grazing avoider type based on palatability would not make a difference in your model as the animals would not discriminate against the avoider. This is a limitation you should mention.
line(s) 629/630: “… and the PFTs had to follow a grazing-tolerance strategy.” - The fact that grazing avoidance can only be achieved through life cycle adaptation and not through palatability likely causes a bias in your community composition. You should at least mention this possibility.
line(s) 632/633: “At the cold steppe site, grazing only happened during the growing season and both grazing tolerance an avoidance could be useful strategies.” – Well, likely not avoidance in the way you can represent it in the model (temporal avoidance). If grazing happens during the growing season, and your only way to implement avoidance is through life cycle adaptation, i.e., temporal avoidance, this will push avoiders to the non-growing season as time when no grazing happens. But I don't see how avoiders could succeed by shifting their existence focus to exactly the season when growth is not possible?
line(s) 643-645: This challenge could be circumvented by moving away from a PFT-concept with fixed pre-defined parameter values for each PFT, which implicitly limits the number of strategies that can be realized, for example by defining typical value ranges for the given parameters of a strategy type. Within these continuous ranges, a strategy type can assume many trait value combinations that define its location within the trait space occupied by the strategy type, and therefore allows more plasticity within a strategy type, e.g., a plant could be a moderate, intermediate, or extreme S-strategy type.
line(s) 645/646: The challenge will be to expand this site-scale-focused approach to a generalized large-scale / global approach, because it will not be possible to parameterize suitable PFTs for all imaginable locations and circumstances. I think the value of what you show in this study is to prove that the CSR-concept can work within a DGVM and is ecologically sound in many points. But to make it general, you will have to move away from the discrete parameterization of your PFT approach, for example by allowing an evolutionary approach that self-selects successfull strategies via environmental filtering from a pool of potential trait value combinations, where each trait is represented by a continuous range of allowed values.
line(s) 664/665: I do not really agree with this approach. The light extinction coefficient (as I know it) is a constant that describes how much light a respective layer of leaves will absorb and how much it will allow to transmit to the next lower leaf level. As such, it is a proxy associated with leaf characteristics such as leaf thickness or SLA more than overall plant stature. If anything, I'd deem LAI closer to stature than the light extinction coefficient, if you do not have height available as state variable.
line(s) 674: In rangelands, mechanical stress through trampling would be another important aspect to consider.
Minor editorial comments
line(s) 10: “… a temperate grassland, a hot and a cold steppe…” => “… a temperate grassland and a hot and a cold steppe…”
line(s) 13: at three grassland sites => at the three grassland sites
line(s) 17: Our results show, that => delete comma
line(s) 39: high carbon inputs => high carbon sequestration
line(s) 61: (examples) => delete, seems to be a leftover note from manuscript writing. Or alternatively replace with the examples you were thinking of…
line(s) 183: “recover slower” – “recover more slowly”
line(s) 184: “the SLA leaf longevity trade-off” – “the SLA v. leaf longevity trade-off”
line(s) 328: “While it remained similar…” – “However, it remained similar…”
line(s) 359 correct typo: resourCe
line(s) 420 contribute – contributed
line(s) 456: “we present results on above-ground biomass” – “we present results based on above-ground biomass”
line(s) 490: “this allows to assess” – “this allows assessment of”, or “this allows assessing”
line(s) 496: we only assess – we only assessed
line(s) 533: “this can be explained with” – “this can be explained by”
line(s) 535: “and contributed to the litter layer” –“and increased the input to the litter layer”.
line(s) 539: “In addition irrigation led to…” – “In addition, irrigation led to…”
line(s) 619: “…which constituted an additional investment.” – Rephrase? “…and therefore, a reduction of investment costs associated with N-fixation.”
Hyphenation of two-word combinations that are used in the function of an adjective:
line(s) 69: “disturbance prone environments” – “disturbance-prone environments”
line(s) 73: “multi species communities” – “multi-species communities”
line(s) 181 “stress prone ecosystems” – “stress-prone ecosystems”
l- 203: “age dependent individual mortality” – “age-dependent individual mortality”
line(s) 231 “plant specific resource availability” – “plant-specific resource availability”
line(s) 249 “site specific conditions” – “site-specific conditions”
line(s) 296: bias adjusted data” – “bias-adjusted data”
line(s) 374, 375 “water saving strategy” – “water-saving strategy”
line(s) 397 resource limited – resource-limited
line(s) 473: “S dominated community” – “S-dominated community”
line(s) 496: “neither water nor nutrient limited” – “neither water- nor nutrient-limited”
line(s) 542, line(s) 543, line(s) 579 “S dominated” – “S-dominated”
l- 580 “nutrient limited” – “nutrient-limited”
References
Scheiter, S., Langan, L. and Higgins, S.I., 2013. Next‐generation dynamic global vegetation models: learning from community ecology. New Phytologist, 198(3), pp.957-969.
Langan, L., Higgins, S.I. and Scheiter, S., 2017. Climate‐biomes, pedo‐biomes or pyro‐biomes: which world view explains the tropical forest–savanna boundary in South America?. Journal of Biogeography, 44(10), pp.2319-2330.
Citation: https://doi.org/10.5194/bg-2023-55-RC1 -
RC2: 'Comment on bg-2023-55', Anonymous Referee #2, 19 May 2023
This study is predicated on a novel way of quantifying CSR plant functional types (PFTs) for species, and comparing these with frameworks including the leaf economics spectrum. This forms the basis of the entire analysis, and so it is fundamental that the way the PFTs are derived represents CSR theory and can be compared against the LES. There are a number of basic problems with the approach used here, however.
Using the trait specific leaf area (SLA) to represent both leaf economics and also within the CSR calculations means that to two measures are very likely correlated, potentially leading to a Type 1 statistical error in which the conclusions are accepted despite the statistical test not being sufficient to assign a realistic probability.
With regard to stress, the authors state that “According to CSR theory, the stress gradient expresses the level of stress a species is exposed to in a certain habitat. It ranges from unstressed to severely stressed, but does not distinguish individual stress categories (e.g. temperature, water or nutrient)” thus “different strategies for water-resource use can be used to distinguish C- and R-strategists (low stress tolerance) from S-strategists”. Thus the traits used here are specific to water stress, and the definition of stress recognised in CSR theory (constrained metabolic efficiency and thus biomass production) is not cited nor considered. Any stress (including water stress - but also factors such as nutrient stress or ‘non-resource’ stressors such as temperature) limits metabolic performance and thus growth and biomass production. Internal, inherent metabolic traits (such as photosynthetic capacity and dark respiration rate) or growth traits (such as relative growth rate) would have been acceptable to demonstrate limitation, but the authors provide no evidence that, for instance, that specific adaptations determining canopy-level conductance can represent the extent of general tolerance to stress.
Line 233: the authors state that “plant stature … can be used to distinguish C- and S-strategists (low disturbance tolerance) from R-strategists”. No: S-selected species can be small (e.g. Salix herbacea) but some may become large over a long life-span (i.e. Sequoiadendron giganteum). What matters is the C-selected species get large quickly, S-selected species can become large eventually over a long life-span, and R-selected species cannot. This is more a reflection of longevity and how rapidly plants achieve adult size. In the present study only juveniles were investigated, so using the leaf area index of a sapling is not going to represent the strategy in the main vegetative phase (seedling CSR strategies are known to be different from adult CSR strategies; Dayrell et al. (2018) Functional Ecology 32, 2730-2741). Also, CSR strategies are phenotypic characters (i.e. attributes of the individual plant that are subject to natural selection), but establishment rate (kest) [line 237] is not a character of an individual (the units of measurement are stated in Table 2 as the number of individuals per metre squared per day – a population measure), and so cannot elucidate the individual phenotype or adaptations of the species (i.e. the plant strategy or PFT).
In Figure 4, the red, green, blue (RGB) color scheme is used both to represent the extent of C, S and R and the experimental treatments rainfed (red), irrigated (blue) and fertilised (green).
Citation: https://doi.org/10.5194/bg-2023-55-RC2
Stephen Björn Wirth et al.
Data sets
Code and data for Wirth et al., in discussion: Connecting CSR theory and LPJmL 5.3 to assess the role of environmental conditions, management and functional diversity for grassland ecosystem function Stephen Björn Wirth, Christoph Müller, and Susanne Rolinski https://zenodo.org/record/7727990
Stephen Björn Wirth et al.
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 174 | 51 | 9 | 234 | 24 | 2 | 6 |
- HTML: 174
- PDF: 51
- XML: 9
- Total: 234
- Supplement: 24
- BibTeX: 2
- EndNote: 6
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1