Meta-analysis reveals that the effects of precipitation change on soil and litter fauna in forests depend on body size

1 Title: Meta-analysis reveals that the effects of precipitation change on soil and litter fauna in 1 forests depend on body size 2 3 Running title: Precipitation change impacts on soil fauna 4 5 List of authors: Philip A. Martin1*, Leonora Fisher2*, Leticia Pérez-Izquierdo1, Charlotte 6 Biryol6, Bertrand Guenet3, Sebastiaan Luyssaert4, Stefano Manzoni5, Claire Menival6, 7 Mathieu Santonja6, Rebecca Spake7, Jan C. Axmacher2, Jorge Curiel Yuste1, 8 8 9 List of Authors’ ORCID IDs: Philip A. Martin: 0000-0002-5346-8868; Leticia Pérez10 Izquierdo: 0000-0002-5200-8157; Charlotte Biryol: 0000-0003-0467-593X; Bertrand Guenet: 11 0000-0002-4311-8645; Sebastiaan Luyssaert: 0000-0003-1121-1869; Stefano Manzoni: 12 0000-0002-5960-5712; Mathieu Santonja: 0000-0002-6322-6352; Rebecca Spake: 000013 0003-4671-2225; Jan C. Axmacher: 0000-0003-1406-928X; Jorge Curiel Yuste: 0000-000214 3221-6960 15 16 Joint First or Senior Authorship: Philip A. Martin and Leonora Fisher should be 17 considered joint first author. 18 19 Institutional affiliations: 1BC3 - Basque Centre for Climate Change, Scientific Campus of 20 the University of the Basque Country, 48940 Leioa, Spain; 2UCL Department of Geography, 21 University College London, Gower Street, London WC1E 6BT, UK.; 3 Laboratoire de 22 Géologie, Ecole normale supérieure, CNRS, IPSL, Université PSL, 24 Rue Lhomond, 75005 23 Paris, France; 4Amsterdam Institute for Life and Environment (A-LIFE), Section Systems 24 Ecology, Vrije Universiteit Amsterdam, the Netherlands; 5 Department of Physical 25 Geography and Bolin Centre for Climate Research, Stockholm University, Stockholm, 26 Sweden; 6Aix Marseille Univ, Avignon Univ, CNRS, IRD, IMBE, Marseille, France; 7School of 27 2 Biological Sciences, University of Reading, RG6 6EX, Reading, UK; 8IKERBASQUE, Basque 28 Foundation for Science, Bilbao, Bizkaia, Spain 29 Contact information: Philip Martin; telephone number: +34 944 014 690; email address: 30 [email protected] 31 Abstract 32 Anthropogenic climate change is altering precipitation regimes at a global scale. While 33 precipitation changes have been linked to changes in the abundance and diversity of soil 34 and litter invertebrate fauna in forests, general trends have remained elusive due to mixed 35 results from primary studies. We used a meta-analysis based on 430 comparisons from 38 36 primary studies to address associated knowledge gaps, (i) quantifying impacts of 37 precipitation change on forest soil and litter fauna abundance and diversity, (ii) exploring 38 reasons for variation in impacts, and (iii) examining biases affecting the realism and 39 accuracy of experimental studies. Precipitation reductions led to a decrease of 39% in soil 40 and litter fauna abundance, with a 35% increase in abundance under precipitation increases, 41 while diversity impacts were smaller. A statistical model containing an interaction between 42 body size and the magnitude of precipitation change showed that mesofauna (e.g. mites, 43 collembola) responded most to changes in precipitation. Changes in taxonomic richness 44 were related solely to the magnitude of precipitation change. Our results suggest that body 45 size is related to the ability of a taxon to survive under drought conditions, or to benefit from 46 high precipitation. We also found that most experiments manipulated precipitation in a way 47 that aligns better with predicted extreme climatic events than with predicted average annual 48 changes in precipitation and that the experimental plots used in experiments were likely too 49 small to accurately capture changes for mobile taxa. The relationship between body size and 50 response to precipitation found here has far-reaching implications for our ability to predict 51 future responses of soil biodiversity to climate change and will help to produce more realistic 52 mechanistic soil models which aim to simulate the responses of soils to global change. 53 3 Keywords: Precipitation change; drought; Soil fauna; meta-analysis; evidence synthesis; 54 climate change 55 56 4 Introduction 57 Anthropogenic climate change is altering global precipitation patterns (Seager et al., 2018) 58 and increasing the frequency and severity of extreme drought and precipitation events (Sun 59 et al., 2007). Understanding the consequences of precipitation changes is particularly vital 60 for forests, given their critical roles in the global carbon cycle (Walker et al., 2021) and in 61 supporting global biodiversity (Benton et al., 2022). Impacts of precipitation changes on 62 forests include increased tree mortality (Anderegg et al., 2019) and consequent increases in 63 CO2 emissions (Doughty et al., 2015; Yang et al., 2018), and mixed effects on aboveground 64 forest biodiversity (Fleming et al., 2021). However, the effects of disturbances on the 65 biodiversity of soil and litter invertebrate fauna in forests, remains poorly known (Pressler et 66 al., 2019; Winding et al., 2020) despite its importance in regulating organic matter 67 decomposition, nutrient cycling, and plant health among other ecosystem functions (Handa 68 et al., 2014; Nielsen et al., 2015). 69 Since soil moisture is a key limiting factor to the fitness and behaviour of many soil 70 and litter fauna, precipitation changes may threaten the processes to which they contribute 71 (Coyle et al., 2017). Precipitation changes and associated changes in soil moisture and 72 distribution of water within the soil matrix can alter the movement of microfauna such as 73 nematodes, and therefore their access to food sources, or the humidity in pores which 74 represent the habitat of mesofauna such as Collembola (Coyle et al., 2017; Deckmyn et al., 75 2020). These changes can alter reproduction and mortality of a wide range of soil and litter 76 fauna (Kardol et al., 2011; Singh et al., 2019; Wang et al., 2022). For example, extreme 77 drought conditions can increase mortality for taxa such as Collembola (Wang et al., 2022) 78 and Enchytraeidae (Maraldo et al., 2009). Nonetheless, while some studies have reported 79 biodiversity losses as a result of precipitation reduction (Aupic-Samain, Santonja, et al., 80 2021; Chikoski et al., 2006; Lindberg et al., 2002; Santonja et al., 2017) others have reported 81 increases (Homet et al., 2021; Lensing et al., 2005), with similarly mixed results for studies 82 5 of precipitation increases (Chikoski et al., 2006; Frew et al., 2013; Landesman et al., 2011) 83 making generalisation challenging. 84 One obvious reason for heterogeneity among studies of soil faunal responses to 85 precipitation change is the magnitude of the precipitation change itself. Most studies of 86 precipitation change represent manipulative experiments often using rain exclusion devices 87 for reduction treatments or irrigation for precipitation increases, with ambient conditions used 88 as a control. Meta-analyses have failed to find a consistent relationship between the 89 magnitude of precipitation changes and changes in either the abundance or taxonomic 90 richness of soil fauna (Peng et al., 2022). This could in part reflect diverging responses in 91 taxonomic groups to precipitation changes (Coyle et al., 2017). Functional traits, 92 morphological, physiological or phenological features measurable at the individual level 93 (Violle et al., 2007), might offer a tractable way to disentangle some of these differences. 94 There are many traits that could influence soil faunal responses to precipitation 95 change. Here we focus on three. First, taxa that inhabit the litter layer are likely to be more 96 exposed to extreme fluctuations in moisture, and thus to respond more strongly than taxa 97 inhabiting deeper, more buffered soil horizons (Fraser et al., 2012). Second, the presence of 98 an exoskeleton and a cuticle layer that helps to reduce water loss and may hence render 99 arthropods less prone to desiccation than soft-bodied annelids such as Enchytraeidae 100 (Evans, 2008; Singh et al., 2019). Third, body size relates to microhabitat preferences and 101 therefore dependence on water availability. For example, microfauna, such as nematodes, 102 inhabit water films, so may be particularly vulnerable as they are essentially aquatic 103 organisms (Vandegehuchte et al., 2015), mesofauna, such as Collembola, are sensitive to 104 changes in soil moisture because they are confined to existing air-filled pore spaces (Wang 105 et al., 2022) while macrofauna can create their own pore spaces (Lavelle et al., 2002) and 106 are more capable of avoidant behaviour such as burrowing to deeper depths (Gerard, 1967). 107 Therefore, increasing body size likely yields greater resistance to changes in precipitation. 108 In addition to functional traits, the characteristics of forest ecosystems could also 109 influence the responses of soil and litter fauna to precipitation changes. For example, in 110 6 relatively arid regions that have regularly been exposed to drought conditions in the past, soil 111 and litter fauna communities will have been subject to environmental filtering, which selects 112 for combinations of functional traits that govern species’ ability to persist in the local 113 environment (Balmford, 1996; Kraft et al., 2015). This means that we might expect greater 114 reductions in the abundance of soil and litter fauna in humid forests following precipitation 115 reductions than in relatively arid forests. Equally, we might expect greater increases in 116 abundance in arid forests following precipitation increases. Integrating information on 117 functional traits and ecosystem characteristics into research syntheses should allow for a 118 mechanistic understanding as to why soil faunal responses to precipitation are 119 heterogeneous. 120 Alongside a lack of understanding of between-study variability, it is also unclear how 121 study design impacts study results. Meta-research in ecology has shown that differences in 122 experimental and sampling designs can have large impacts on the accuracy of estimates of 123 biodiversity change (Christie et al., 2019, 2020; Spake et al., 2021). But methodological 124 robustness is rarely assessed in ecological meta-analyses (Pullin et al., 2022). As well as 125 methodological robustness, it is also unclear if experimental studies employ realistic future 126 precipitation scenarios. This is a particularly serious issue, given existing concerns about the 127 use of unrealistic precipitation manipulations in global change experiments which may result 128 in underor overestimations of impact (Korell et al., 2020; Kröel-Dulay et al., 2022). 129 To address these knowledge gaps, we carried out the first meta-analysis of the effects 130 of precipitation changes on soil and litter fauna in forests. In this study, we address three 131 questions: (1) What are the impacts of precipitation changes on the abundance and diversity 132 of forest soil and litter invertebrate fauna? (2) What are the major determinants of the impacts 133 of precipitation changes on abundance and diversity? (3) What are the major biases in studies 134 of the impacts of precipitation change on forest soil and litter invertebrate fauna? 135 For question 1, we hypothesised that precipitation reductions cause declines in the 136 abundance and diversity of soil and litter fauna, whereas additional precipitation causes an 137 increase in abundance and diversity (H1, Figure 1a). For question 2, we tested five 138 7 hypotheses: (i) an increased magnitude of changes in precipitation amplifies changes in 139 abundance and diversity (H2, Figure 1b); (ii) the effect of precipitation magnitude is further 140 amplified for organisms found in litter compared to soil dwellers (H3, Figure 1c), (iii) for 141 organisms without an exoskeleton compared to those with an exoskeleton (H4, Figure 1d), 142 (iv) for organisms with smaller body sizes (H5, Figure 1e), (v) or that organisms in moist forests 143 are more sensitive to reductions in precipitation, while organisms in drier forests are sensitive 144 to increases in precipitation (H6, Figure 1f). There were no hypotheses for question 3. 145 146 Materials and Methods 147 Searches and screening 148 This study focuses on the impacts of precipitation changes on forest soil and litter 149 invertebrate fauna in field settings. We formally defined these as PECOS elements (Table 1, 150 Grames et al., 2019). More precise definitions of these elements can be found in the 151 Supplementary methods. This study follows guidelines for synthesis in environmental 152 management (Collaboration for Environmental Evidence, 2018). We chose to focus on 153 precipitation change because although there are numerous primary studies investigating its 154 impact there is currently a lack of robust synthesis. We did not focus on the impacts of other 155 important global change drivers such as temperature change due to existing meta-analyses 156 on this topic (Goncharov et al., 2023; Peng et al., 2022). 157 As part of a previous systematic map (Martin et al., 2021), we screened a large 158 number of papers to identify those that assessed the impact of natural disturbances on soil 159 and litter invertebrate fauna in forests (see Supplementary methods for a summary of 160 systematic map methods). From the systematic map we identified 320 articles that related to 161 impacts of natural disturbances. On the 22nd of February 2024, we updated our search by 162 searching in three bibliographic platforms (Web of Science, Scopus, and Open Access 163 Theses and Dissertations) to find studies on the impact of precipitation changes on soil and 164 8 litter invertebrate fauna in forests published after our initial systematic map searches. Since 165 different bibliographic platforms have different rules for the formatting of searches, we 166 developed platform-specific searches (see Tables S2 and S3). By searching for unpublished 167 grey literature as well as published, peer-reviewed literature, we aimed to minimise the risk 168 of publication bias which could lead inaccurate estimates of disturbance impacts (Konno & 169 Pullin, 2020). In addition to formal searches, we contacted expert researchers to help identify 170 potentially relevant studies. 171 Once searches were complete, we downloaded all references found as .bib or .ris 172 files and used the R package synthesisr to remove duplicate articles (Westgate & Grames, 173 2020) . The bibfix package (Haddaway et al., 2021) was used to repair bibliographic files 174 with incomplete data. Files were then uploaded to sysrev (Bozada et al., 2021) - an online 175 tool that allows for screening and data extraction by review teams (see Martin, 2021). Article 176 titles and abstracts were screened for relevance, and articles that met inclusion criteria were 177 retained and their full text reviewed. To meet our eligibility criteria studies needed to: (1) 178 Relate to soil and litter fauna in forests; (2) Address the impact of changes in precipitation; 179 (3) Be field-based (i.e. not be carried out in greenhouses or mesocosms); (4) Quantitatively 180 assess soil fauna biomass, abundance, or diversity; (5) Have a comparison between sites 181 that vary in the intensity or frequency of the precipitation that they were exposed to; (6) Be 182 written in English; (7) Report measures of centrality (mean or median) for relevant litter or 183 soil fauna outcomes. 184 At the title and abstract screening stage, in order to be retained, articles needed to be 185 likely to meet criteria 1-3 and criterion 5. At the full-text stage criteria 1-7 needed to be met in 186 order for an article to be retained. At the full-text screening stage, we provided reasons for 187 the exclusion of all articles that did not meet our inclusion criteria in accordance with ROSES 188 guidelines (Haddaway et al., 2018; Figure S1). Despite being a multilingual team, we 189 focussed only on English-language literature because the inclusion of non-English language 190 literature would have made carrying out consistency checks between reviewers challenging. 191 9 We acknowledge that excluding literature written in non-English languages is a shortcoming 192 that may lead to biases (Amano et al., 2021; Konno et al., 2020). 193 To ensure consistency, a random sample of 10% of titles and abstracts were 194 screened by two team members, using our inclusion criteria. Any disagreements between 195 the two people were discussed, and eligibility criteria were revised where appropriate. 196 Cohen’s Kappa scores were calculated to test the agreement between the two people 197 (Cohen, 1960). If Kappa scores were below 0.6, another 10% of titles and abstracts were 198 screened by the same two team members with the process repeated until Kappa scores 199 were >0.6. The same process was repeated for the full texts of publications that met 200 inclusion criteria. After screening of titles and abstracts, inter-reviewer agreement was 98.7% 201 and the Kappa score was 0.93. For full text screening agreement was 100% and the Kappa 202 score was 1.0. We found 1965 papers during our updated searches, 82 of which were 203 retained after screening of titles and abstracts, and 38 of which were used for critical 204 appraisal and data extraction. We used 430 comparisons between control and treatment 205 groups extracted from these studies. The screening process is summarised in more detail in 206 Figure S1. 207 Critical appraisal 208 Critical appraisal of studies to assess their methodological robustness is a vital part 209 of synthesis (Collaboration for Environmental Evidence, 2018). We did this by assessing the 210 following threats to the internal validity of a study based on Martin et al. (2020) (i) selection 211 bias: when selection of study sites leads to a result that is systematically different to the 212 target population; (ii) confounding: where systematic distortion of the effect of a treatment 213 caused by mixing of the treatment of interest with other disturbances (e.g. plots where 214 precipitation was manipulated were in plantations while control plots were in natural forests); 215 and (iii) performance bias, differences that occur due to knowledge by researchers about 216 treatment allocation. We therefore determined whether studies (i) consisted of both spatial 217 (i.e. comparisons between control and treatment groups) and temporal comparisons (i.e. 218 16 The most parsimonious models for changes in taxonomic richness included only the 381 magnitude of precipitation change or the year of publication (Table S11). Model-averaging 382 showed a significant positive relationship with precipitation change (coefficient = 0.002, SE = 383 0.001, p-value = 0.0473, Figure 4, Table S12), indicating support for the impact of 384 precipitation change magnitude. Again, more recent studies tended to have larger effect 385 sizes, although this trend was not statistically significant (Table S12). Similarly, for Shannon386 Wiener diversity the most parsimonious models included different combinations of the 387 magnitude of precipitation change and/or the year of publication (Table S13). Model 388 averaging showed a non-significant negative effect of precipitation magnitude on Shannon389 Wiener diversity (coefficient = -0.002, SE = 0.004, p-value = 0.703, Table S14), and small 390 non-statistically significant effects of study size and publication year (Figure S5). 391 392 Study biases 393 There are clear biases in the geographic distribution of studies, with a large number 394 of studies carried out in western Europe and the USA, but relatively few in South America 395 and Asia, and no studies found for Africa (Figure 5a). This translates to an 396 underrepresentation of tropical forest biomes, with most studies carried out in temperate 397 seasonal forests or woodland/shrubland biomes found in Mediterranean climates (Figure 398 5b). In addition to geographic biases, there were also a number of biases that could impact 399 the validity of study results. First, studies of the effects of precipitation reduction reduced 400 precipitation by 91% more than projected changes for the same location (Figure 6a, 401 coefficient = -2.37, SE = 0.65, p-value = 0.004), while studies of precipitation increase 402 increased precipitation by 58% more than projected changes, although this difference was 403 not statistically significant (Figure 6a, coefficient = 0.455, SE = 0.846, p-value = 0.603). For 404 more details of results for each of the 25 climate models used in this analysis see Figure S6. 405 Second, the plots used for experimental manipulations tended to be small for studies of 406 micro-, meso-, and macrofauna (Figure 6b), with median areas of 460 m2, 300 m2, and 36 m2 407 respectively. 408 17 Discussion 409 Our findings partially supported our hypothesis (H1), indicating that reductions in 410 precipitation generally cause large decreases in the abundance of soil and litter fauna in 411 forests, while precipitation increases have the opposite effect. However, impacts on 412 taxonomic richness and Shannon-Wiener diversity were typically less pronounced. Changes 413 in abundance depended on the magnitude of precipitation changes and taxa body size: 414 mesofauna abundance changes were positively correlated with changes in precipitation, but 415 there was little detectable effect of body size for either microor macrofauna. We found 416 weak support for a positive correlation between changes in precipitation and changes in 417 taxonomic richness but no support for this correlation for Shannon-Wiener diversity. Thus, 418 the best supported of our hypotheses regarding the variability in response to precipitation 419 changes was H5, that the impacts of precipitation change depended on taxa body size. 420 However, there was only weak support for H2, that increased magnitude of changes in 421 precipitation amplifies changes in abundance and diversity, and little support for the effects 422 of fauna occupying litter or soil (H3), possessing an exoskeleton (H4) or whether sites were 423 relatively humid or arid (H6) regarding the modification of impacts of precipitation changes. 424 Additionally, our analyses to account for publication biases relating to study size and 425 changes in effect sizes over time mean that these results are highly robust. 426 Impacts of precipitation change 427 Our results broadly agree with those of the meta-analysis by Peng et al. (2022), who found 428 that impacts of precipitation change on the abundance of soil fauna in forests were much 429 larger than for richness. However, our meta-analysis included more than three times as 430 many primary studies relating to forests, and over twice as many effect sizes, indicating that 431 our results represent an important advance in robustness. Our results also broadly mirrored 432 those found by the recent meta-analysis of Bristol et al. (2023) who focussed solely on 433 nematodes and showed a non-significant increase in abundance as a result of precipitation 434 18 increases and a non-significant decrease as a result of precipitation reductions. In addition, 435 unlike the synthesis of Goncharov et al. (2023) we found evidence for impacts of 436 precipitation change on mesofauna, while their study only found impacts of precipitation 437 change on nematodes. In contrast to all previous similar meta-analyses on this topic 438 (Blankinship et al., 2011; Goncharov et al., 2023; Peng et al., 2022), we found important 439 evidence for nuanced effects of precipitation change. 440 The lack of pronounced changes in either taxonomic richness or Shannon-Wiener 441 diversity was surprising, but one intriguing finding was that species richness changed little as 442 a result of precipitation change, while Shannon-Wiener diversity was significantly reduced. 443 This hints that, as suggested by others, some Oribatida and Collembola species become 444 increasingly dominant when soil moisture is altered, reducing evenness (Meehan et al., 445 2020). In contrast with our findings for changes in abundance, we found relatively little 446 support for the effect of changes in precipitation magnitude or species traits on taxonomic 447 richness or Shannon-Wiener diversity. This could result from changes in local diversity as a 448 result of perturbations often not reflecting those in community composition (Hillebrand et al., 449 2018; Zajicek et al., 2021). This occurs when there is turnover in the identity and abundance 450 of species but no systematic change in the number of species (Hillebrand et al., 2018) as 451 appears to be common for numerous human-impacted ecosystems (Dornelas et al., 2014; 452 Vellend et al., 2013). This response differs from that observed in soil microbial communities 453 under reduced precipitation, which not only show a reduced abundance of the active 454 community, but also lower diversity and markedly different composition (Metze et al., 2023; 455 Zhou et al., 2020). However, it is also possible that the apparent lack of effect is actually a 456 result of different responses to precipitation changes across taxonomic or functional groups 457 that we were unable to capture due to a lack of data. 458 Drivers of precipitation change impact 459 Our findings for changes in abundance suggest that water availability is a key 460 constraint for many forest soil taxa (Aupic-Samain, Baldy, et al., 2021), but that impacts of 461 19 soil moisture vary depending on organism body size. The effect of intense precipitation 462 changes on mesofauna abundance is consistent with previous studies that suggested that 463 this group can be particularly sensitive to environmental changes (Wu & Wang, 2019) and 464 that high-intensity disturbances can have an enduring effect on ecological processes and 465 hinder recovery (Nielsen & Ball, 2015). This could lead to reductions in the incorporation of 466 leaf litter into soil, given that litter forms a major part of the diet for taxa such as Collembola 467 and Oribatida (Potapov et al., 2022). The apparent lack of significant impact of precipitation 468 change on microand macrofauna, while contradicting our expectations, could have a 469 variety of causes. 470 Our results suggest that there is a hump-shaped relationship between the body size 471 of soil and litter fauna and their sensitivity to precipitation changes, with microand 472 macrofauna being relatively insensitive and mesofauna being highly sensitive. We 473 hypothesise that this sensitivity is caused by three factors. First, differences in the ability to 474 avoid predation. Under drier conditions, microfauna, such as nematodes, can become 475 restricted to small pores (Erktan et al., 2020) that act as refuges from predatory mesofauna 476 such as mites (Potapov et al., 2022) which are unable to access them. Mesofauna are 477 confined to larger, air-filled pores (Erktan et al., 2020). In contrast to nematodes, this 478 confinement does not protect them against predation, because mesoand macrofauna 479 predators can move between soil layers in search of prey (Potapov et al., 2022). Thus, 480 mesofauna remains subject to predation even in dry conditions. 481 Second, physical adaptations to dry conditions. Both microand macrofauna possess 482 physical adaptations which aid them in drier conditions. Microfauna, such as nematodes, can 483 go into anhydrobiosis when under drought stress (Landesman et al., 2011; Watanabe, 484 2006). Many macrofauna, such as spiders or millipedes, have thick exoskeletons which 485 protect them against desiccation as well as being highly mobile, thus allowing them to move 486 more easily to wetter soil patches. In contrast, many mesofauna, such as Collembola or 487 Protura, have few physical adaptations to drought conditions and have limited mobility within 488 20 the soil. As a consequence, they are likely subject to greater drought-induced mortality than 489 microor macrofauna. 490 Third, availability of food sources. Soil and litter fauna feed on a wide variety of food, 491 ranging from living or dead plant matter to soil bacteria and fungi, as well as other soil fauna 492 (Potapov et al., 2022). Changes in the biomass and palatability of these different food 493 sources as a result of precipitation reductions causes cascading bottom-up impacts 494 throughout soil food webs. While reductions in the food sources for all soil and litter fauna 495 seem likely in dry conditions, mesofauna may be more severely impacted than the other 496 groups. For example, many Collembola species are fungivores and drier conditions may 497 reduce the abundance and quality of fungi available to them. Drought conditions appear to 498 favour mycorrhizal fungi that contain melanin in their cell walls (Fernandez & Koide, 2014; 499 Pigott, 1982), making them resistant to decomposition thereby limiting food sources for 500 detritivores (Fernandez et al., 2013; Fernandez & Koide, 2014; Malik & Haider, 1982). In 501 addition, precipitation reductions can reduce saprotrophic fungi abundance, and important 502 source of food for Collembola (Sanders et al., 2024). However, many microand macrofauna 503 groups have relatively diverse diets, potentially providing a buffer when some sources of 504 food are scarce (Potapov et al., 2022). Similarly, reductions in precipitation can reduce litter 505 input, reducing the quantity of food for some detritivore mesofauna, and increase the carbon 506 content of litter (Deng et al., 2021), reducing decomposition rates, resulting in a shortage of 507 readily available nutrients for litter feeding fauna. 508 Under increased precipitation, the hump-shaped relationship between size and 509 abundance appears to be reversed. This implies that mesofauna is more sensitive to 510 increases in water resources than either microor macrofauna. Mesofauna appear to be 511 more easily affected by seasonal changes than macrofauna due to their smaller body size 512 and shorter life cycles (Wu & Wang, 2019), thus explaining their increase in abundance with 513 increased precipitation. Mesofauna may also be less affected by predation under wetter 514 conditions (Aupic-Samain, Baldy, et al., 2021). Meanwhile, for microfauna such as 515 21 nematodes, increased precipitation may reduce the overall fungal biomass, reducing 516 populations of fungivorous nematodes (Liu et al. 2020). At the same time some fungal 517 groups, such as saprotrophic fungi, on which mesofauna such as Collembola preferentially 518 feed, are expected to increase under wetter conditions (Sanders et al., 2024). 519 The effect of body size seen in our meta-analysis represents an advance in our 520 understanding of the responses of soil biota to changes in precipitation associated with 521 climate change. However, the mechanisms that regulate this response to changes in water 522 availability are currently unclear and further research could substantially improve our ability 523 to predict the future impact of climate change on the resilience of soils and their functioning 524 in the face of climate change. One such potential impact is that loss of mesofauna could 525 cause a reduction in the rate of litter decomposition (Song et al., 2020) resulting in a 526 reduction in the incorporation of organic matter into soils and a reduction in the complexity of 527 soil structure. Equally, such a reduction in soil mesofauna could lead to increases in the 528 abundance of taxa belonging to other size groups that also feed on litter (e.g. earthworms) 529 resulting in a change in the structure of soil food webs, which may potentially buffer the 530 impacts of precipitation changes on soil functioning. However, this replacement could entail 531 major changes in the physical structure of the soil, since earthworms are ecosystem 532 engineers which can alter soil porosity (Flores et al., 2021). 533 Although our explanations for the observed patterns are grounded in theory and 534 empirical evidence from the literature, new experiments and observations are needed to test 535 them. In particular, within broad taxonomic groups, there are large differences in key traits 536 that determine their response to precipitation change. For example, Collembola, vary in body 537 size, permeability of cuticle, and occupancy of different soil layers, all of which influence how 538 species respond to drought conditions (Ferrín et al., 2023). For Nematodes, body size and 539 trophic group are linked with fungivores and bacterivores being relatively small, and 540 omnivores and predators being relatively large (Sechi et al., 2018). Given these variations, 541 we urge researchers to identify soil and litter fauna to a higher taxonomic resolution, thus 542 22 allowing for more nuanced interpretations of how body size and other functional traits impact 543 responses to precipitation change. 544 Study biases and recommendations for future research 545 Our study identified a need for changes in studies of precipitation change impacts on 546 forest soil and litter fauna. The proposed changes may be difficult to implement, and we 547 acknowledge that decisions about study practicalities are the result of a mixture of factors 548 such as socioeconomics (Llorente-Culebras et al., 2023) and the obsession with academic 549 productivity (Fischer et al., 2012). First, linked to our finding that many experiments use 550 precipitation regime alterations that are much more extreme than projected future changes, 551 we advocate for researchers to clearly distinguish between experiments which aim to 552 simulate changes in mean annual precipitation and those that aim to simulate extreme 553 events such as droughts and extreme rainfall (Korell et al., 2020). Second, this study shows 554 that the scale of experimental manipulations in many studies may be too small to capture 555 changes in more mobile macrofauna taxa, and so larger-scale studies are needed that allow 556 for a wider range of organisms and processes to be studied (Hanson & Walker, 2020). Third, 557 we found strong geographic biases, with few studies found outside of temperate and 558 Mediterranean forest biomes, and thus suggest the greater need for studies outside of these 559 regions. 560 While there is a need for changes in how primary studies are conducted, the same is 561 true for syntheses relating to soil fauna. Our study represents one of most methodologically 562 robust meta-analyses to date in soil ecology, collating more studies on the impacts of 563 precipitation changes than previous similar meta-analyses (Blankinship et al., 2011; Peng et 564 al., 2022), despite our narrower focus on precipitation changes in forests, and thus providing 565 greater statistical power than previous efforts. We encourage more researchers to strive for 566 more robust evidence syntheses and familiarise themselves with existing guidance for 567 evidence synthesis in ecology (Collaboration for Environmental Evidence, 2018; Haddaway 568 et al., 2018; Figure S1, 2020). In our study we used the log response ratio as an effect size 569 23 metric, due to differences between studies in the units of abundance. Because the log 570 response ratio measures proportionate change in biodiversity relative to a control or baseline 571 value, there is a loss of information that can render meta-analyses vulnerable to possible 572 inferential errors when baselines vary across studies (Spake et al., 2023). 573 In addition, existing meta-analyses on the impacts of global change on soil fauna 574 (This study; Beaumelle et al., 2023; Blankinship et al., 2011; Bristol et al., 2023; Goncharov 575 et al., 2023; Peng et al., 2022; Phillips et al., 2023) use biodiversity metrics related to 576 abundance and alpha diversity, meaning we know little about impacts on more complex 577 aspects of biodiversity such as community composition and functional diversity. We 578 advocate for researchers to collate and use raw data from field studies to allow for more 579 nuanced ‘full data’ analyses which can avoid issues associated with the use of effect sizes 580 (Spake et al., 2023) and that are becoming the gold standard in other fields, such as 581 medicine (Culina et al., 2018; Spake et al., 2022). Finally, we recognise that we were unable 582 to explicitly examine the impact of study scale (e.g., grain, extent) on observed changes in 583 soil fauna biodiversity as a result of precipitation changes. However, given the general 584 importance of scale for observations in ecology (Spake et al., 2021) and the findings that 585 precipitation change experiments have scale-dependent effects on other taxa (Korell et al., 586 2021) we encourage researchers to address this topic in future syntheses. 587 We acknowledge that our meta-analysis focussed solely on the impacts of 588 precipitation changes and ignored the potential impacts of other drivers that could have 589 synergistic impacts on soil and litter fauna. In the real world, ecosystems are typically 590 affected by more than one of global change factor (e.g. temperature increases, nitrogen 591 deposition) at any given time (Bowler et al., 2020). Importantly, interactions between 592 different global change factors can cause unpredictable changes in soil biodiversity and 593 functioning (Eisenhauer et al., 2012; Rillig et al., 2019). In the case of precipitation 594 reductions there is a clear synergy with temperature increases as these can lead to 595 increased evapotranspiration and further reductions in soil moisture. Such conditions could 596 24 further stress soil fauna sensitive to changes in moisture by limiting their diets (Sanders et 597 al., 2024) or directly causing desiccation. However, the recent meta-analysis of Peng et al. 598 (2022) suggests there are currently few primary studies that have investigated this synergy, 599 especially in forests. We urge researchers to prioritise experiments investigating the impacts 600 of multiple global change factors to inform more realistic predictions of future change. 601 602 Conclusion 603 Overall, our results suggest that forest soil and litter fauna abundance is sensitive to 604 changes in precipitation, and that for mesofauna this impact depends on the magnitude of 605 precipitation change. Meanwhile, alpha diversity appeared to be relatively insensitive, with 606 little evidence that changes were related to the magnitude of precipitation change. Given soil 607 mesofauna affect soil functions, such as litter decomposition, changes in the abundance of 608 this group may result in changes in the soil physical structure, soil nutrients, and soil carbon. 609 In turn, changes in mesofauna abundance will also alter the trophic structure of belowground 610 food webs. Our results provide new insights into belowground biodiversity change in forests 611 that can inform more realistic soil models in the future (Deckmyn et al., 2020; Flores et al., 612 2021). In addition, we call on global change researchers to conduct more realistic studies of 613 changes in mean annual precipitation, droughts, and extreme precipitation in future, in line 614 with representative concentration pathways (RCPs; van Vuuren et al., 2011) as well as 615 larger scale experiments to capture impacts on soil fauna more fully. 616 Acknowledgements 617 Philip Martin, Leticia Pérez-Izquierdo, Charlotte Biryol, Bertrand Guenet, Sebastiaan 618 Luyssaert, Stefano Manzoni, Claire Menival, Mathieu Santonja, and Jorge Curiel Yuste were 619 funded by the grant Holistic management practices, modelling and monitoring for European 620 forest soils – HoliSoils (EU Horizon 2020 Grant Agreement No 101000289). JCY was also 621 funded by the coordinated project ATLANTIS (PID2020-113244GB-C21), the Basque 622 25 Government through the BERC 2022-2025 program, and the Spanish Ministry of Science 623 and Innovation through the BC3 María de Maeztu excellence accreditation (MDM-2017624 0714). We would also like to thank two anonymous referees for help in improving this article. 625 Author contributions 626 Philip A. Martin: Conceptualization; Data curation; Analysis; Experiments/Investigation; 627 Methodology; Project Administration; Supervision; Validation; Visualization; Writing – original 628 draft; Writing – review & editing. Leonora Fisher: Data curation; Analysis; 629 Experiments/Investigation; Methodology; Visualization; Writing – original draft; Writing – 630 review & editing. Leticia Pérez-Izquierdo: Data curation; Writing – review & editing. 631 Charlotte Biryol: Data curation; Visualization; Writing – review & editing; Bertrand Guenet: 632 Conceptualisation; Funding acquisition; Writing – review & editing. Sebastiaan Luyssaert: 633 Conceptualisation; Funding acquisition; Writing – review & editing. Stefano Manzoni: 634 Conceptualisation; Writing – review & editing. Claire Menival: Data curation; Visualization; 635 Writing – review & editing. Mathieu Santonja: Conceptualisation; Funding acquisition; 636 Visualization; Writing – review & editing. Rebecca Spake: Conceptualisation; Methodology; 637 Visualization; Writing – review & editing. Jan C. Axmacher: Conceptualisation; 638 Methodology; Supervision; Writing – review & editing. Jorge Curiel Yuste: 639 Conceptualization; Funding acquisition; Experiments/Investigation; Methodology; Project 640 Administration; Resources; Supervision; Validation; Writing – original draft; Writing – review 641 & editing. 642 643 References 644 Amano, T., Berdejo-Espinola, V., Christie, A. P., Willott, K., Akasaka, M., Báldi, A., 645 Berthinussen, A., Bertolino, S., Bladon, A. J., Chen, M., Choi, C.-Y., Bou Dagher 646 Kharrat, M., de Oliveira, L. G., Farhat, P., Golivets, M., Hidalgo Aranzamendi, N., 647 Jantke, K., Kajzer-Bonk, J., Kemahlı Aytekin, M. Ç., … Sutherland, W. J. (2021). 648 32 E. W., Van de Waal, D. B., & Ryabov, A. B. (2018). Biodiversity change is uncoupled 817 from species richness trends: Consequences for conservation and monitoring. The 818 Journal of Applied Ecology, 55(1), 169–184. https://doi.org/10.1111/1365-2664.12959 819 Homet, P., Gómez-Aparicio, L., Matías, L., & Godoy, O. (2021). Soil fauna modulates the 820 effect of experimental drought on litter decomposition in forests invaded by an exotic 821 pathogen. The Journal of Ecology, 109(8), 2963–2980. https://doi.org/10.1111/1365822 2745.13711 823 Kardol, P., Reynolds, W. N., Norby, R. J., & Classen, A. T. (2011). Climate change effects 824 on soil microarthropod abundance and community structure. Applied Soil Ecology: A 825 Section of Agriculture, Ecosystems & Environment, 47(1), 37–44. 826 https://doi.org/10.1016/j.apsoil.2010.11.001 827 Konno, K., Akasaka, M., Koshida, C., Katayama, N., Osada, N., Spake, R., & Amano, T. 828 (2020). Ignoring non-English-language studies may bias ecological meta-analyses. 829 Ecology and Evolution, 10(13), 6373–6384. https://doi.org/10.1002/ece3.6368 830 Konno, K., & Pullin, A. S. (2020). Assessing the risk of bias in choice of search sources for 831 environmental meta‐analyses. Research Synthesis Methods. 832 https://doi.org/10.1002/jrsm.1433 833 Korell, L., Auge, H., Chase, J. M., Harpole, W. S., & Knight, T. M. (2020). Understanding 834 plant communities of the future requires filling knowledge gaps [Review of 835 Understanding plant communities of the future requires filling knowledge gaps]. Global 836 Change Biology, 26(2), 328–329. https://doi.org/10.1111/gcb.14920 837 Korell, L., Auge, H., Chase, J. M., Harpole, W. S., & Knight, T. M. (2021). Responses of 838 plant diversity to precipitation change are strongest at local spatial scales and in 839 drylands. Nature Communications, 12(1), 2489. https://doi.org/10.1038/s41467-021840 22766-0 841 Koricheva, J., & Kulinskaya, E. (2019). Temporal Instability of Evidence Base: A Threat to 842 Policy Making? Trends in Ecology & Evolution, 34(10), 895–902. 843 https://doi.org/10.1016/j.tree.2019.05.006 844 33 Kraft, N. J. B., Adler, P. B., Godoy, O., James, E. C., Fuller, S., & Levine, J. M. (2015). 845 Community assembly, coexistence and the environmental filtering metaphor. Functional 846 Ecology, 29(5), 592–599. https://doi.org/10.1111/1365-2435.12345 847 Kröel-Dulay, G., Mojzes, A., Szitár, K., Bahn, M., Batáry, P., Beier, C., Bilton, M., De Boeck, 848 H. J., Dukes, J. S., Estiarte, M., Holub, P., Jentsch, A., Schmidt, I. K., Kreyling, J., 849 Reinsch, S., Larsen, K. S., Sternberg, M., Tielbörger, K., Tietema, A., … Peñuelas, J. 850 (2022). Field experiments underestimate aboveground biomass response to drought. 851 Nature Ecology & Evolution, 6(5), 540–545. https://doi.org/10.1038/s41559-022-01685-3 852 Landesman, W. J., Treonis, A. M., & Dighton, J. (2011). Effects of a one-year rainfall 853 manipulation on soil nematode abundances and community composition. Pedobiologia, 854 54(2), 87–91. https://doi.org/10.1016/j.pedobi.2010.10.002 855 Lavelle, P., Senapati, B., & Barros, E. (2002). Soil macrofauna. In Trees, crops and soil 856 fertility: concepts and research methods (pp. 303–323). CABI Publishing. 857 https://doi.org/10.1079/9780851995939.0303 858 Lensing, J. R., Todd, S., & Wise, D. H. (2005). The impact of altered precipitation on spatial 859 stratification and activity-densities of springtails (Collembola) and spiders (Araneae). 860 Ecological Entomology, 30(2), 194–200. https://doi.org/10.1111/j.0307861 6946.2005.00669.x 862 Lindberg, N., Engtsson, J. B., & Persson, T. (2002). Effects of experimental irrigation and 863 drought on the composition and diversity of soil fauna in a coniferous stand. The Journal 864 of Applied Ecology, 39(6), 924–936. https://doi.org/10.1046/j.1365-2664.2002.00769.x 865 Liu, M., Miao, X., & Hua, F. (2023). The perils of measuring biodiversity responses to habitat 866 change using mixed metrics. Conservation Letters, 16(4). 867 https://doi.org/10.1111/conl.12959 868 Liu, T., Mao, P., Shi, L., Wang, Z., Wang, X., He, X., Tao, L., Liu, Z., Zhou, L., Shao, Y., & 869 Fu, S. (2020). Contrasting effects of nitrogen deposition and increased precipitation on 870 soil nematode communities in a temperate forest. Soil Biology & Biochemistry, 148, 871 107869. https://doi.org/10.1016/j.soilbio.2020.107869 872 34 Llorente-Culebras, S., Ladle, R. J., & Santos, A. M. C. (2023). Publication trends in global 873 biodiversity research on protected areas. Biological Conservation, 281, 109988. 874 https://doi.org/10.1016/j.biocon.2023.109988 875 Malik, K. A., & Haider, K. (1982). Decomposition of 14C-labelled melanoid fungal residues in 876 a marginally sodic soil. Soil Biology & Biochemistry, 14(5), 457–460. 877 https://doi.org/10.1016/0038-0717(82)90104-3 878 Maraldo, K., Ravn, H. W., Slotsbo, S., & Holmstrup, M. (2009). Responses to acute and 879 chronic desiccation stress in Enchytraeus (Oligochaeta: Enchytraeidae). Journal of 880 Comparative Physiology. B, Biochemical, Systemic, and Environmental Physiology, 881 179(2), 113–123. https://doi.org/10.1007/s00360-008-0305-5 882 Martin, P. A. (2021). Soil_fauna_systematic_map. https://sysrev.com/p/98916 883 Martin, P. A., Perez Izquierdo, L., Luyssaert, S., Guenet, B., Manzoni, S., Spake, R., 884 Santonja, M., & Curiel Yuste, J. (2021). Effects of natural disturbances on soil fauna in 885 forests: a systematic map protocol. In OSF. https://doi.org/10.17605/OSF.IO/YQU4W 886 Martin, P. A., Shackelford, G. E., Bullock, J. M., & Sutherland, W. J. (2020). Management of 887 UK priority invasive alien plants: a systematic review protocol. Environmental Evidence. 888 https://link.springer.com/article/10.1186/s13750-020-0186-y 889 Meehan, M. L., Barreto, C., Turnbull, M. S., Bradley, R. L., Bellenger, J.-P., Darnajoux, R., & 890 Lindo, Z. (2020). Response of soil fauna to simulated global change factors depends on 891 ambient climate conditions. Pedobiologia, 83, 150672. 892 https://doi.org/10.1016/j.pedobi.2020.150672 893 Metze, D., Schnecker, J., Canarini, A., Fuchslueger, L., Koch, B. J., Stone, B. W., Hungate, 894 B. A., Hausmann, B., Schmidt, H., Schaumberger, A., Bahn, M., Kaiser, C., & Richter, 895 A. (2023). Microbial growth under drought is confined to distinct taxa and modified by 896 potential future climate conditions. Nature Communications, 14(1), 5895. 897 https://doi.org/10.1038/s41467-023-41524-y 898 Nakagawa, S., & Freckleton, R. P. (2008). Missing inaction: the dangers of ignoring missing 899 data. Trends in Ecology & Evolution, 23(11), 592–596. 900 35 https://doi.org/10.1016/j.tree.2008.06.014 901 Nakagawa, S., Lagisz, M., Jennions, M. D., Koricheva, J., Noble, D. W. A., Parker, T. H., 902 Sánchez-Tójar, A., Yang, Y., & O’Dea, R. E. (2021). Methods for testing publication bias 903 in ecological and evolutionary meta‐analyses. Methods in Ecology and Evolution / 904 British Ecological Society. https://doi.org/10.1111/2041-210x.13724 905 Nakagawa, S., Noble, D. W. A., Lagisz, M., Spake, R., Viechtbauer, W., & Senior, A. M. 906 (2023). A robust and readily implementable method for the meta-analysis of response 907 ratios with and without missing standard deviations. Ecology Letters, 26(2), 232–244. 908 https://doi.org/10.1111/ele.14144 909 Nakagawa, S., Yang, Y., Macartney, E. L., Spake, R., & Lagisz, M. (2023). Quantitative 910 evidence synthesis: a practical guide on meta-analysis, meta-regression, and 911 publication bias tests for environmental sciences. Environmental Evidence, 12(1), 1–19. 912 https://doi.org/10.1186/s13750-023-00301-6 913 Nielsen, U. N. (Ed.). (2019). Soil Fauna Assemblages: Global to Local Scales. In Soil Fauna 914 Assemblages: Global to Local Scales (pp. v – v). Cambridge University Press. 915 https://www.cambridge.org/core/books/soil-fauna-assemblages/soil-fauna916 assemblages/78C28F6983A0011E36D74EB272EC143E 917 Nielsen, U. N., & Ball, B. A. (2015). Impacts of altered precipitation regimes on soil 918 communities and biogeochemistry in arid and semi-arid ecosystems. Global Change 919 Biology, 21(4), 1407–1421. https://doi.org/10.1111/gcb.12789 920 Nielsen, U. N., Wall, D. H., & Six, J. (2015). Soil Biodiversity and the Environment. Annual 921 Review of Environment and Resources, 40(1), 63–90. https://doi.org/10.1146/annurev922 environ-102014-021257 923 Peng, Y., Peñuelas, J., Vesterdal, L., Yue, K., Peguero, G., Fornara, D. A., Heděnec, P., 924 Steffens, C., & Wu, F. (2022). Responses of soil fauna communities to the individual 925 and combined effects of multiple global change factors. Ecology Letters, 25(9), 1961– 926 1973. https://doi.org/10.1111/ele.14068 927 Phillips, H., Cameron, E. K., Eisenhauer, N., Burton, V., Ferlian, O., Jin, Y., Kanabar, S., 928 36 Malladi, S., Murphy, R., Peter, A., Petrocelli, I., Ristok, C., Tyndall, K., van der Putten, 929 W., & Beaumelle, L. (2023). Global change and their environmental stressors have a 930 significant impact on soil biodiversity -- a meta-analysis. Authorea Preprints. 931 https://doi.org/10.22541/au.167655684.49855023/v1 932 Pick, J. L., Nakagawa, S., & Noble, D. W. A. (2019). Reproducible, flexible and high‐ 933 throughput data extraction from primary literature: The metaDigitise r package. Methods 934 in Ecology and Evolution / British Ecological Society, 10(3), 426–431. 935 https://doi.org/10.1111/2041-210x.13118 936 Pigott, C. D. (1982). SURVIVAL OF MYCORRHIZA FORMED BY CENOCOCCUM 937 GEOPHILUM FR. IN DRY SOILS. The New Phytologist, 92(4), 513–517. 938 https://doi.org/10.1111/j.1469-8137.1982.tb03409.x 939 Potapov, A. M., Beaulieu, F., Birkhofer, K., Bluhm, S. L., Degtyarev, M. I., Devetter, M., 940 Goncharov, A. A., Gongalsky, K. B., Klarner, B., Korobushkin, D. I., Liebke, D. F., 941 Maraun, M., Mc Donnell, R. J., Pollierer, M. M., Schaefer, I., Shrubovych, J., Semenyuk, 942 I. I., Sendra, A., Tuma, J., … Scheu, S. (2022). Feeding habits and multifunctional 943 classification of soil-associated consumers from protists to vertebrates. Biological 944 Reviews of the Cambridge Philosophical Society. https://doi.org/10.1111/brv.12832 945 Pressler, Y., Moore, J. C., & Cotrufo, M. F. (2019). Belowground community responses to 946 fire: meta-analysis reveals contrasting responses of soil microorganisms and 947 mesofauna. Oikos , 128(3), 309–327. https://doi.org/10.1111/oik.05738 948 Pullin, A. S., Cheng, S. H., Jackson, J. D., Eales, J., Envall, I., Fada, S. J., Frampton, G. K., 949 Harper, M., Kadykalo, A. N., Kohl, C., Konno, K., Livoreil, B., Ouédraogo, D.-Y., 950 O’Leary, B. C., Pullin, G., Randall, N., Rees, R., Smith, A., Sordello, R., … Woodcock, 951 P. (2022). Standards of conduct and reporting in evidence syntheses that could inform 952 environmental policy and management decisions. Environmental Evidence, 11(1), 1–11. 953 https://doi.org/10.1186/s13750-022-00269-9 954 Rillig, M. C., Ryo, M., Lehmann, A., Aguilar-Trigueros, C. A., Buchert, S., Wulf, A., Iwasaki, 955 A., Roy, J., & Yang, G. (2019). The role of multiple global change factors in driving soil 956 37 functions and microbial biodiversity. Science, 366(6467), 886–890. 957 https://doi.org/10.1126/science.aay2832 958 Sanders, S. K. D., Martínez-De León, G., Formenti, L., & Thakur, M. P. (2024). How will 959 climate change affect the feeding biology of Collembola? Soil Biology & Biochemistry, 960 188, 109244. https://doi.org/10.1016/j.soilbio.2023.109244 961 Santonja, M., Fernandez, C., Proffit, M., Gers, C., Gauquelin, T., Reiter, I. M., Cramer, W., & 962 Baldy, V. (2017). Plant litter mixture partly mitigates the negative effects of extended 963 drought on soil biota and litter decomposition in a Mediterranean oak forest. The Journal 964 of Ecology, 105(3), 801–815. https://doi.org/10.1111/1365-2745.12711 965 Seager, R., Lis, N., Feldman, J., Ting, M., Park Williams, A., Nakamura, J., Liu, H., & 966 Henderson, N. (2018). Whither the 100th Meridian? The Once and Future Physical and 967 Human Geography of America’s Arid–Humid Divide. Part I: The Story So Far. Earth 968 Interactions, 22(5), 1–22. https://doi.org/10.1175/EI-D-17-0011.1 969 Sechi, V., De Goede, R. G. M., Rutgers, M., Brussaard, L., & Mulder, C. (2018). Functional 970 diversity in nematode communities across terrestrial ecosystems. Basic and Applied 971 Ecology, 30, 76–86. https://doi.org/10.1016/j.baae.2018.05.004 972 Singh, J., Schädler, M., Demetrio, W., Brown, G. G., & Eisenhauer, N. (2019). Climate 973 change effects on earthworms - a review. Soil Organisms, 91(3), 114–138. 974 https://doi.org/10.25674/so91iss3pp114 975 Song, X., Wang, Z., Tang, X., Xu, D., Liu, B., Mei, J., Huang, S., & Huang, G. (2020). The 976 contributions of soil mesofauna to leaf and root litter decomposition of dominant plant 977 species in grassland. Applied Soil Ecology: A Section of Agriculture, Ecosystems & 978 Environment, 155, 103651. https://doi.org/10.1016/j.apsoil.2020.103651 979 Spake, R., Bowler, D. E., Callaghan, C. T., Blowes, S. A., Doncaster, C. P., Antão, L. H., 980 Nakagawa, S., McElreath, R., & Chase, J. M. (2023). Understanding “it depends” in 981 ecology: a guide to hypothesising, visualising and interpreting statistical interactions. 982 Biological Reviews of the Cambridge Philosophical Society. 983 https://doi.org/10.1111/brv.12939 984 38 Spake, R., Mori, A. S., Beckmann, M., Martin, P. A., Christie, A. P., Duguid, M. C., & 985 Doncaster, C. P. (2021). Implications of scale dependence for cross-study syntheses of 986 biodiversity differences. Ecology Letters. https://doi.org/10.1111/ele.13641 987 Spake, R., O’Dea, R. E., Nakagawa, S., Patrick Doncaster, C., Ryo, M., Callaghan, T., & 988 Bullock, J. M. (2022). Improving quantitative synthesis to achieve generality in ecology. 989 Nature Ecology and Evolution, 6, 1818–1828. https://doi.org/10.1038/s41559-022990 01891-z 991 Ștefan, V., & Levin, S. (2018). plotbiomes: R package for plotting Whittaker biomes with 992 ggplot2. https://doi.org/10.5281/zenodo.7145245 993 Sun, Y., Solomon, S., Dai, A., & Portmann, R. W. (2007). How Often Will It Rain? Journal of 994 Climate, 20(19), 4801–4818. https://doi.org/10.1175/JCLI4263.1 995 Vandegehuchte, M. L., Sylvain, Z. A., Reichmann, L. G., de Tomasel, C. M., Nielsen, U. N., 996 Wall, D. H., & Sala, O. E. (2015). Responses of a desert nematode community to 997 changes in water availability. Ecosphere , 6(3), art44. https://doi.org/10.1890/es14998 00319.1 999 van Vuuren, D. P., Edmonds, J., Kainuma, M., Riahi, K., Thomson, A., Hibbard, K., Hurtt, G. 1000 C., Kram, T., Krey, V., Lamarque, J.-F., Masui, T., Meinshausen, M., Nakicenovic, N., 1001 Smith, S. J., & Rose, S. K. (2011). The representative concentration pathways: an 1002 overview. Climatic Change, 109(1), 5. https://doi.org/10.1007/s10584-011-0148-z 1003 Vellend, M., Baeten, L., Myers-Smith, I. H., Elmendorf, S. C., Beauséjour, R., Brown, C. D., 1004 De Frenne, P., Verheyen, K., & Wipf, S. (2013). Global meta-analysis reveals no net 1005 change in local-scale plant biodiversity over time. Proceedings of the National Academy 1006 of Sciences of the United States of America, 110(48), 19456–19459. 1007 https://doi.org/10.1073/pnas.1312779110 1008 Viechtbauer, W. (2010). Conducting meta-analyses in R with the metafor package. Journal 1009 of Statistical Software, 36(3), 1–48. https://lirias.kuleuven.be/1059637?limo=0 1010 Violle, C., Navas, M.-L., Vile, D., Kazakou, E., Fortunel, C., Hummel, I., & Garnier, E. (2007). 1011 Let the concept of trait be functional! Oikos , 116(5), 882–892. 1012 39 https://doi.org/10.1111/j.0030-1299.2007.15559.x 1013 Walker, A. P., De Kauwe, M. G., Bastos, A., Belmecheri, S., Georgiou, K., Keeling, R. F., 1014 McMahon, S. M., Medlyn, B. E., Moore, D. J. P., Norby, R. J., Zaehle, S., Anderson1015 Teixeira, K. J., Battipaglia, G., Brienen, R. J. W., Cabugao, K. G., Cailleret, M., 1016 Campbell, E., Canadell, J. G., Ciais, P., … Zuidema, P. A. (2021). Integrating the 1017 evidence for a terrestrial carbon sink caused by increasing atmospheric CO2. The New 1018 Phytologist, 229(5), 2413–2445. https://doi.org/10.1111/nph.16866 1019 Wang, Y., Slotsbo, S., & Holmstrup, M. (2022). Soil dwelling springtails are resilient to 1020 extreme drought in soil, but their reproduction is highly sensitive to small decreases in 1021 soil water potential. Geoderma, 421, 115913. 1022 https://doi.org/10.1016/j.geoderma.2022.115913 1023 Watanabe, M. (2006). Anhydrobiosis in invertebrates. Applied Entomology and Zoology, 1024 41(1), 15–31. https://doi.org/10.1303/aez.2006.15 1025 Westgate, M., & Grames, E. (2020). synthesisr: Import, Assemble, and Deduplicate 1026 Bibliographic Datasets. https://CRAN.R-project.org/package=synthesisr 1027 Winding, A., Singh, B. K., Bach, E., Brown, G., Zhang, J., Cooper, M., Dion, P., Mele, P., 1028 Eisenhauer, N., Pena-Neira, S., & Others. (2020). State of knowledge of soil 1029 biodiversity: status, challenges, and potentialities. 1030 https://researchdirect.westernsydney.edu.au/islandora/object/uws:62954/ 1031 Wu, P., & Wang, C. (2019). Differences in spatiotemporal dynamics between soil 1032 macrofauna and mesofauna communities in forest ecosystems: The significance for soil 1033 fauna diversity monitoring. Geoderma, 337, 266–272. 1034 https://doi.org/10.1016/j.geoderma.2018.09.031 1035 Yang, Y., Saatchi, S. S., Xu, L., Yu, Y., Choi, S., Phillips, N., Kennedy, R., Keller, M., 1036 Knyazikhin, Y., & Myneni, R. B. (2018). Post-drought decline of the Amazon carbon 1037 sink. Nature Communications, 9(1), 3172. https://doi.org/10.1038/s41467-018-05668-6 1038 Zajicek, P., Welti, E. A. R., Baker, N. J., Januschke, K., Brauner, O., & Haase, P. (2021). 1039 Long-term data reveal unimodal responses of ground beetle abundance to precipitation 1040 40 and land use but no changes in taxonomic and functional diversity. Scientific Reports, 1041 11(1), 17468. https://doi.org/10.1038/s41598-021-96910-7 1042 Zhou, Z., Wang, C., & Luo, Y. (2020). Meta-analysis of the impacts of global change factors 1043 on soil microbial diversity and functionality. Nature Communications, 11(1), 3072. 1044 https://doi.org/10.1038/s41467-020-16881-7 1045 Zomer, R. J., Xu, J., & Trabucco, A. (2022). Version 3 of the Global Aridity Index and 1046 Potential Evapotranspiration Database. Scientific Data, 9(1), 409. 1047 https://doi.org/10.1038/s41597-022-01493-1 1048 Data sources 1049 Ashton, L. A., Griffiths, H. M., Parr, C. L., Evans, T. A., Didham, R. K., Hasan, F., Teh, Y. A., 1050 Tin, H. S., Vairappan, C. S., & Eggleton, P. (2019). Termites mitigate the effects of drought 1051 in tropical rainforest. Science, 363(6423), 174–177. https://doi.org/10.1126/science.aau9565 1052 1053 Aslam, T. J., Benton, T. G., Nielsen, U. N., & Johnson, S. N. (2015). Impacts of eucalypt 1054 plantation management on soil faunal communities and nutrient bioavailability: trading 1055 function for dependence? Biology and Fertility of Soils, 51(5), 637–644. 1056 https://doi.org/10.1007/s00374-015-1003-6 1057 1058 Aupic-Samain, A., Santonja, M., Chomel, M., Pereira, S., Quer, E., Lecareux, C., Limousin, 1059 J.-M., Ourcival, J.-M., Simioni, G., Gauquelin, T., Fernandez, C., & Baldy, V. (2021). Soil 1060 biota response to experimental rainfall reduction depends on the dominant tree species in 1061 mature northern Mediterranean forests. Soil Biology & Biochemistry, 154, 108122. 1062 https://doi.org/10.1016/j.soilbio.2020.108122 1063 1064 Bakonyi, G., Nagy, P., Kovács-Láng, E., Kovács, E., Barabás, S., Répási, V., & Seres, A. 1065 (2007). Soil nematode community structure as affected by temperature and moisture in a 1066 temperate semiarid shrubland. Applied Soil Ecology: A Section of Agriculture, Ecosystems & 1067 41 Environment, 37(1), 31–40. https://doi.org/10.1016/j.apsoil.2007.03.008 1068 1069 Chikoski, J. M., Ferguson, S. H., & Meyer, L. (2006). Effects of water addition on soil 1070 arthropods and soil characteristics in a precipitation-limited environment. Acta Oecologica, 1071 30(2), 203–211. https://doi.org/10.1016/j.actao.2006.04.005 1072 1073 Delph, R. J., Clifford, M. J., Cobb, N. S., Ford, P. L., & Brantley, S. L. (2014). Pinyon Pine 1074 Mortality Alters Communities of Ground-Dwelling Arthropods. Western North American 1075 Naturalist / Brigham Young University, 74(2), 162–184. https://doi.org/10.3398/064.074.0203 1076 1077 Ferguson, S. H., & Joly, D. O. (2002). Dynamics of springtail and mite populations: the role 1078 of density dependence, predation, and weather. Ecological Entomology, 27(5), 565–573. 1079 https://doi.org/10.1046/j.1365-2311.2002.00441.x 1080 1081 Frew, A., Nielsen, U. N., Riegler, M., & Johnson, S. N. (2013). Do eucalypt plantation 1082 management practices create understory reservoirs of scarab beetle pests in the soil? 1083 Forest Ecology and Management, 306, 275–280. 1084 https://doi.org/10.1016/j.foreco.2013.06.051 1085 1086 Guidi, C., Frey, B., Brunner, I., Meusburger, K., Vogel, M. E., Chen, X., Stucky, T., 1087 Gwiazdowicz, D. J., Skubała, P., Bose, A. K., Schaub, M., Rigling, A., & Hagedorn, F. 1088 (2022). Soil fauna drives vertical redistribution of soil organic carbon in a long-term irrigated 1089 dry pine forest. Global Change Biology, 28(9), 3145–3160. https://doi.org/10.1111/gcb.16122 1090 1091 Homet, P., Gómez-Aparicio, L., Matías, L., & Godoy, O. (2021). Soil fauna modulates the 1092 effect of experimental drought on litter decomposition in forests invaded by an exotic 1093 pathogen. The Journal of Ecology, 109(8), 2963–2980. https://doi.org/10.1111/13651094 2745.13711 1095 48 have positive impacts; (b) abundance and diversity changes with respect to control plots are 1229 driven by the magnitude of precipitation changes; (c) the effects of changes in precipitation 1230 depend on whether invertebrate fauna are found in the soil or the litter; (d) the effects of 1231 changes in precipitation depend on whether invertebrate fauna have an exoskeleton or not; 1232 (e) the effects of changes in precipitation depend on the body size of invertebrate fauna; (f) 1233 the effects of changes in precipitation depend on the aridity of the forest ecosystem. Dashed 1234 lines represent points at which there is no change in precipitation or no change in effect 1235 sizes relating to soil and litter fauna biodiversity. Effect size refers to the differences in 1236 abundance or biodiversity between control and treatment groups, with positive changes 1237 representing increases in abundance or biodiversity and decreases representing a loss in 1238 abundance or biodiversity. For all of these hypotheses we assume that abundance and 1239 diversity did not change for the control groups. 1240 1241 49 Figure 2 - Changes in the (a) abundance (b) Shannon-Wiener diversity index, and (c) 1242 taxonomic richness of soil and litter fauna in forests as a result of precipitation change. Large 1243 points refer to the summary effect size, thicker bars around them representing the 95% 1244 confidence intervals, and the thinner bars the 95% prediction intervals. Smaller, semi1245 transparent points represent individual comparisons. Differences in their size refer to the 1246 weight they supply to each analysis. The vertical dashed line represents where the effect 1247 size is equal to zero (i.e. where there is no difference between control and treatment 1248 groups). Annotations on the left of the plot refer to the number of comparisons in each 1249 analysis (k) and, in parentheses, the number of studies they are taken from. Annotations on 1250 the right of the plot refer to the mean weighted percentage change for each analysis and 1251 asterisks (*) indicate when effect sizes are significantly different from zero. 1252 1253 Figure 3 - Changes in the abundance of soil and litter fauna in forests relative to changes in 1254 precipitation for different faunal size classes. Points represent individual comparisons with 1255 different point sizes representing the different weights of comparisons to the analysis. Solid 1256 lines represent predictions from the most parsimonious model (R2=0.13), with darker 1257 coloured bands representing the 95% confidence intervals, and the lighter bands the 95% 1258 prediction intervals. Dashed lines represent points at which there is no change in 1259 precipitation (x equal to zero) or in effect size (y equal to zero). Annotations on the plot refer 1260 to the number of comparisons in each analysis (k) and, in parentheses, the number of 1261 studies they are taken from. 1262 50 1263 Figure 4 - Changes in taxonomic richness relative to changes in precipitation. Points 1264 represent individual comparisons with different point sizes representing the different weights 1265 of points in the analysis. The solid line represents predictions from the most parsimonious 1266 model (R2=0.26), with darker coloured bands representing the 95% confidence intervals, and 1267 the lighter bands the 95% prediction intervals. Dashed lines represent points at which the x 1268 and y axes are equal to zero. Annotations on the plot refer to the number of comparisons in 1269 the analysis (k) and, in parentheses, the number of studies they are taken from. 1270 1271 51 Figure 5 - (a) The location of study sites and (b) distribution within biomes. Studies of 1272 precipitation increase are shown by blue upward-pointing triangle symbols, studies of 1273 precipitation decrease by red downward-pointing triangles, and studies that investigated both 1274 increases and decreases are shown by purple diamonds. The size of the symbols indicates 1275 the number of comparisons made at each location (minimum: 1, maximum: 88). In (b) the 1276 location of each site in a Whittaker biome diagram is defined by mean annual temperature and 1277 mean annual precipitation. Map lines delineate study areas and do not necessarily depict 1278 accepted national boundaries 1279 1280 Figure 6 - Biases that affect the validity of study results: (a) Differences in precipitation 1281 changes investigated in studies compared to projected precipitation changes based on 25 1282 climate model projections for 2041-2060; (b) Sizes of plots used in experimental studies of 1283 precipitation changes effects on the differing body size groups of invertebrate soil and litter 1284 fauna. In (a) points represent median values for each group and error bars the 95% 1285 confidence intervals. The dashed vertical line represents the point at which there is no 1286 difference between projected and studied level of precipitation change. Annotations on the 1287 right of the plot refer to the mean percentage change for each analysis and asterisks (*) 1288 indicate when effect sizes are significantly different from zero. 1289 1290