The handbook for standardized field and laboratory measurements in terrestrial climate change experiments and observational studies (ClimEx)

Climate change is a world‐wide threat to biodiversity and ecosystem structure, functioning and services. To understand the underlying drivers and mechanisms, and to predict the consequences for nature and people, we urgently need better understanding of the direction and magnitude of climate change impacts across the soil–plant–atmosphere continuum. An increasing number of climate change studies are creating new opportunities for meaningful and high‐quality generalizations and improved process understanding. However, significant challenges exist related to data availability and/or compatibility across studies, compromising opportunities for data re‐use, synthesis and upscaling. Many of these challenges relate to a lack of an established ‘best practice’ for measuring key impacts and responses. This restrains our current understanding of complex processes and mechanisms in terrestrial ecosystems related to climate change. To overcome these challenges, we collected best‐practice methods emerging from major ecological research networks and experiments, as synthesized by 115 experts from across a wide range of scientific disciplines. Our handbook contains guidance on the selection of response variables for different purposes, protocols for standardized measurements of 66 such response variables and advice on data management. Specifically, we recommend a minimum subset of variables that should be collected in all climate change studies to allow data re‐use and synthesis, and give guidance on additional variables critical for different types of synthesis and upscaling. The goal of this community effort is to facilitate awareness of the importance and broader application of standardized methods to promote data re‐use, availability, compatibility and transparency. We envision improved research practices that will increase returns on investments in individual research projects, facilitate second‐order research outputs and create opportunities for collaboration across scientific communities. Ultimately, this should significantly improve the quality and impact of the science, which is required to fulfil society's needs in a changing world.


| INTRODUC TI ON
Climate change is an increasing threat to biodiversity and ecosystem functions and services world-wide (IPBES, 2019;Pacifici et al., 2015;Schuur et al., 2015). Numerous studies, including experiments, long-term monitoring, resampling of historic data and various kinds of space-for-time approaches, have been and are being conducted.
Such studies address climate and other global change-related questions for various players and processes in ecosystems across the soil-plant-atmosphere continuum. Plants are key players in our ecosystems: they make up the majority of biomass globally, are the main source of photosynthetically fixed carbon and serve as habitat, food sources and hosts for other organismal groups, while also competing with them for resources (Bar-On, Phillips, & Milo, 2018;Fanin et al., 2019). Plants interact with other plants, microbes, invertebrates and other organisms in multiple ways and on different trophic levels (e.g. decomposers, mutualists, pathogens, herbivores, predators). Together, they play a major role in controlling important terrestrial ecosystem processes such as carbon and nutrient cycling (Clemmensen et al., 2013). Improved understanding of the magnitude of climate change impacts, the underlying drivers and mechanisms and the ecological consequences at the individual, population, community and ecosystem scales across the soil-plantcontinuum, with a focus on first-order plant-animal interactions in detritivore and herbivore food web is urgently needed.
As the number of such climate change studies is quickly increasing, new opportunities for meta-analyses and syntheses are emerging to explore broad-scale patterns and context dependencies in effects, responses and processes, as well as for model development and evaluation. However, there are major challenges for meaningful and high-quality syntheses, including geographic and climatic biases in the available studies (Beier et al., 2012;Vicca et al., 2014), a lack of coordinated measurements and standardized protocols (Denny et al., 2014;Firbank et al., 2017;Vicca, Gilgen, et al., 2012) and a lack of basic and well-structured study information (i.e. covariates, metadata) needed for comparisons (Gerstner et al., 2017;Haddaway & Verhoeven, 2015). Similarly, model-experiment interactions are hampered by methodological inconsistencies across experiments, and by the lack of specific ancillary data to characterize sites and experimental treatments (Medlyn et al., 2015).
The majority of climate change studies are local or regional, and single-to multiple-site studies that are designed to address experts from across a wide range of scientific disciplines. Our handbook contains guidance on the selection of response variables for different purposes, protocols for standardized measurements of 66 such response variables and advice on data management. Specifically, we recommend a minimum subset of variables that should be collected in all climate change studies to allow data re-use and synthesis, and give guidance on additional variables critical for different types of synthesis and upscaling. The goal of this community effort is to facilitate awareness of the importance and broader application of standardized methods to promote data re-use, availability, compatibility and transparency. We envision improved research practices that will increase returns on investments in individual research projects, facilitate second-order research outputs and create opportunities for collaboration across scientific communities. Ultimately, this should significantly improve the quality and impact of the science, which is required to fulfil society's needs in a changing world.

K E Y W O R D S
best practice, coordinated experiments, data management and documentation, ecosystem, experimental macroecology, methodology, open science, vegetation specific research questions. As a consequence, these studies often have unique experimental and sampling designs (e.g. Countryside survey, Emmett et al., 2010;ExpeEr, Bertora et al., 2013;INCREASE, Schmidt et al., 2014). Individual research projects and networks invest considerable resources in collecting data for a number of environmental and biotic variables and in developing protocols for field measurements. This leads to a diversity of similar but not quite identical protocols, and hence to a diversity of ways to measure and quantify the same underlying effects and responses. While some of this variability may be due to good scientific reasons, protocol selection is often based on traditions and habits. This methodological diversity can hinder backward compatibility, and thus, syntheses across studies, for example due to incompatible quantification of treatment effects, lack of necessary covariates or response variables, and/or incompatible scale or units of measurements (Estiarte et al., 2016;Lu et al., 2013;Vicca, Gilgen, et al., 2012). These issues largely stem from using a diversity of protocols, and we refer to this as the 'data compatibility filter' (Figure 1). Another issue that may hinder syntheses and meta-analyses is when key information is not available from the original studies. Data, covariates, metadata and detailed methodological information that are critical for the synthesis step may not be necessary for first-order publications and are hence not reported, structured well or stored in an accessible location and format. We refer to this issue as the 'data availability filter' (Figure 1), which is increasingly recognized in the scientific community and is, in part, addressed in recent guidelines on data management and open science practices (e.g. British Ecological Society, 2017;Nosek et al., 2015; and see protocol 1.5). Together, the data compatibility filter and the data availability filter significantly compromise second-order research outputs such as data re-use, synthesis and upscaling (Figure 1).
The handbook for plant trait measurements is an excellent example of the latter approach; a collection of standardized, documented, stateof-the-art protocols for trait-based plant ecological research, providing F I G U R E 1 Challenges and solutions to the integration of information from first-order outputs of individual experiments or observational studies (i.e. the original data and publications) to second-order research outputs resulting from, for example meta-analyses, reviews and modelling (i.e. data re-use, synthesis and upscaling). The figure illustrates the major challenges to achieve such second-order outputs, summarized as two filters (dashed lines) relating to data availability and data compatibility across studies. Two general approaches to solve these challenges exist; either using formally coordinated and distributed experiments or using standardized methods, sampling protocols and reporting across individual and independent studies. The aim of this paper is to contribute to the latter approach by offering guidance on selection of response variables, protocols for standardized measurements of these variables and advice on data reporting and management in climate change studies general methodological advice on how to select and measure traits, independent of the particularities of the individual experiments or study (Cornelissen et al., 2003;Pérez-Harguindeguy et al., 2013). This handbook has become a key resource for trait-based research as it has been hugely important in stimulating the use of standardized methods while at the same time providing an easily available resource for new researchers in the field. Its potential is evident from its high impact world-wide (the two editions of the handbook have been cited over 1,700 times in Web of Science to date, September 2019). Other similar efforts exist like the web resource PrometheusWiki, a handbook for standardized physiological protocols (Sack et al., 2010), or the International Biological Program Handbooks (e.g. Milner & Elfyn Hughes, 1968), as well as national or community standards such as the UK Countryside survey (Emmett et al., 2010) or the UK Environmental Change Network (Sier & Monteith, 2016).
The aim of this paper was to provide a handbook (ClimEx) of standardized field and laboratory methods across the soil-plant-atmosphere continuum applicable to a broad range of terrestrial ecosystem climate change studies (including both experiments and plot-based temporal and spatial gradient studies) world-wide. Because of their dominance in terrestrial ecosystems (Bar-On et al., 2018), we focus mainly on plant communities and the other organisms with which they interact. We discuss a variety of organisms (i.e. microbes, fungi, invertebrates and first-order plant-animal interactions) where the connection to vegetation and ecosystem functioning is apparent.
Through a community effort by experts across scientific disciplines (Box 1), we provide a comprehensive collection of state-of-the-art methods and sampling protocols on key variables of relevance for terrestrial climate change research (Table 1).
The protocols are grouped into five chapters: site characteristics and data management, carbon and nutrient cycling, water cycling, species and interactions, and stress physiology. In each protocol, we describe which response variables should be measured broadly across studies to facilitate data re-use, synthesis and upscaling, using a Gold standard (if possible) and, if applicable, we offer minimal requirement Bronze standards. We discuss Special cases, emerging issues and challenges that address adaptations needed for specific systems or situations and new methods under development. Finally, we give guidance on additional variables that should be measured more widely to improve inter-site and inter-study comparisons and model-experiment interactions.
Our intention is that the ClimEx handbook will be widely used and will stimulate standardized data collection and collaboration between projects within and across sub-disciplines. We acknowledge that different projects may have different specific needs and resources available, and therefore propose a 'minimal set' of variables and practices that, if more widely applied, would significantly increase each individual study's potential contribution to second-order outputs. Many of the methods described in this handbook are also applicable more generally in ecosystem ecology and for other global-change drivers (e.g. nitrogen deposition, invasion, pollution, land-use change). Where relevant, this is indicated in the protocols.

| How were the protocols produced and assessed?
We collected a large number of protocols from major terrestrial climate change experiments; ITEX, DroughtNet and NutNet to name a few, and complemented these with literature on methodology, method descriptions from publications and protocols from our own projects (see the appendix for full reference lists). From these sources, we extracted a list of response variables that are relevant and commonly used in terrestrial climate change research (Table 1).
A core group outlined the writing process, the relevant response variables, the protocol structure and the final editing, while expert teams on each topic wrote the protocols (Box 1).
We focussed on variables that are relevant for understanding climate change impacts on the biodiversity and functioning of the soilplant-atmosphere continuum, and/or which are particularly useful for data re-use, synthesis and upscaling. We do not comprehensively

| Content and structure of the protocols
For each protocol, we provide a short overview of the equipment and running costs, installation and maintenance effort, knowledgeneeds and the type of measurements conducted (Table 1). Note that this overview is to a certain extent country-, habitat-and project-dependent (i.e. costs differ between countries, knowledgeneeds for species identification might differ between a rainforest and a temperate bog). We give a short summary of the ecological background and its relevance to terrestrial climate change studies. We succinctly describe What and how to measure, providing sufficient detail to ensure reproducibility, and provide a reference list with more extensive literature on each method. We describe a Gold standard, that is the optimal measurement irrespective of economic, technical and practical constraints, and (if applicable) a Bronze standard, that is the minimal requirement for data of adequate quality, which can be advisable in specific situations. In the section on Special cases, emerging issues and challenges, we explain how the method can be adapted in specific cases and provide guidance on relevant challenges and opportunities. We further provide an extensive list of key references on the underlying theories, assumptions and applications of each method in the section on Theory, significance and large datasets, as well as under More on methods and existing protocols. These references may be especially useful for students, early-career scientists or research teams getting started in setting up new studies, and more generally for readers interested in exploring the wider literature related to a specific ecological response variable. Each chapter is available as a separate supplement for easier handling, but we encourage the climate change research community to be aware of aspects of other chapters outside their own scientific expertise.

| RE SULTS -OVERVIE W OF THE HANDBOOK CONTENT
We produced a collection of 66 protocols, describing response variables ranging from elements and organisms to the ecosystem scale, organized into five supporting information chapters (Table 1).
All protocols can be found in the online Supporting Information to this paper. In addition, the protocols are also available online on the ClimEx handbook webpage (clime xhand book.w.uib.no). To ensure that the handbook will also be a useful resource for the community in the future, comments and suggestions for updating the protocols can be made via the webpage. These comments and suggestions will be assessed by the authors of this paper and every update will be tracked. The protocols should be cited as appendices to this paper, see individual protocols for details. In the following, we present a summary of these chapters, guidance on their use and examples of their relevance to climate change research.

| Chapter 1: Site characteristics and data management
Systematic reporting of background information of the studied system, data analysis and results is necessary to ensure that studies are reproducible and to enable data syntheses and analyses across studies, meta-analyses, modelling and upscaling (Gerstner et al., 2017;Haddaway & Verhoeven, 2015). Such information includes a basic site description, such as geographic location and abiotic and biotic properties, pre-experimental baseline conditions (e.g. climate, soil properties), basic methodological descriptions (e.g. sample size, timing, duration), a proper characterization of treatments and results (e.g. units, effect sizes), and well structured and documented data management.
Surprisingly, this necessary basic information is often incomplete or missing in scientific publications (Hillebrand & Gurevitch, 2013).
The site characteristics and data management chapter (Supporting Information S1) therefore describes which key site, study system, and study design variables and information should be collected, and how this information is best reported. We first give practical guidance on how to design and set up a climate change study that may serve multiple uses beyond the needs of the particular project. Then, we describe basic site description parameters (e.g. coordinates, elevation, land-use history, vegetation) and physical (e.g. soil horizon, pH), chemical (e.g. nutrient availability) and meteorological variables. Although some of this information may not directly relate to the particular research question or hypothesis of the original project, reporting all relevant information is essential as it puts studies in a larger context and is key to making data and results useful beyond the particular research for which they were designed.
Arguably, improved collection and reporting of background information about the sites and study systems are probably the largest benefit of this handbook in terms of opportunities for second-order outputs (Figure 1). At the same time, these data are non-focus variables in most studies and therefore typically have low priority. To stimulate systematic and standardized collection and reporting of key background variables, we therefore provide an overview of the most critical variables, both overall and specifically for different kinds of data re-use, synthesis and upscaling (Table 2). Here, we also indicate the spatial scale (plot/treatment/site) at which the different variables should preferably be collected for different opportunities of data reuse. The first 'minimum requirement' column in Table 2   Alongside the planning of the study design and data collection, the 'data cycle' in a research project -creating, processing, documenting, sharing, storing and re-using data -should also be planned, follow standardized practice and be well structured and documented (British Ecological Society, 2018;Poisot, Bruneau, Gonzalez, Gravel, & Peres-Neto, 2019). Good data management practice is a key element of 'preproducible' research practice, a term describing a holistic approach to reproducible science (Stark, 2018). We argue that good data management practice is a key step in overcoming the challenges related to the 'data availability filter' (Figure 1) and thus in creating opportunities for data re-use, synthesis and upscaling. Funding bodies and publishers are increasingly recognizing this, and now often require data management plans and open science practice (British Ecological Society, 2018).
We therefore include guidance on open science practice, reproducible workflow and data management in this chapter.

| Chapter 2: Carbon and nutrient cycling
The cycles of carbon and nutrients involve many ecological processes that are all sensitive to climate change (Bai et al., 2013;Yue et al., 2017), including photosynthesis, above-and below-ground plant growth, autotrophic and heterotrophic respiration, organic matter decomposition and nutrient mineralization. Even minor changes in any of these processes, or in the balance between them, can have implications for biodiversity and ecosystems, which again can impact strongly on ecosystem goods and services such as provisioning of food and fibre, water regulation and carbon sequestration (Trumbore, Brando, & Hartmann, 2015).
In the carbon and nutrient cycling chapter (Supporting Information S2), we focus on the main components of the terrestrial carbon cycle, including above-and below-ground processes, pools and fluxes.
Nutrients are also included, but limited to pools and processes that are linked to carbon cycling and ecosystem feedbacks to climate. We stratify measurements into three thematic protocols (plants, soil and ecosystems) that are particularly relevant when considering carbon and nutrient cycling processes within terrestrial ecosystems.

| Chapter 3: Water cycling
Ecosystems play key roles in water and energy cycles, with feedbacks to climate. The understanding of the water and energy exchange between the soil, plants and the atmosphere is still a major research challenge in climate change research because of difficulties in some of the measurements, which are needed to complete the water and energy balances. Water that enters the ecosystem via precipitation will be separated into evaporation, infiltration, transpiration by plants, drainage to groundwater and (temporary) storage in the soil. All these water fluxes need to be determined to fully understand the water and energy exchange between the ecosystem and the atmosphere.
In the water cycling chapter (Supporting Information S3), we start with soil hydraulic measurements, which include soil moisture, hydraulic conductivity, water retention and water potential. The methods included are key to determining parameters that link to, or are used in, modelling the soil-plant-atmosphere continuum.
We also provide guidelines to quantify the ecosystem water stress aiming to facilitate comparison and syntheses across studies. We further include measurements used to track the progress of water through the plant and back to the atmosphere.

| Chapter 4: Species and interactions
As discussed above, plants and the interactions in which they participate play key roles in terrestrial ecosystems. Impacts of climate change on ecosystems and their functioning are therefore to a large extent, mediated through the plant-soil food web, both through changes in the abundance and diversity of organisms, and through changes in their interactions (competition, mutualism, grazing, predation, decomposition, etc.).
In the species and interactions chapter (Supporting Information S4), we provide guidance on quantifying the consequences of climate change for key organisms, biodiversity components and ecological processes across the plant-soil food web.
At the level of populations, we cover vital rates such as reproduction, recruitment, growth, mortality and phenology, as well as the overall impacts on the dynamics and growth rates of populations.
At the community level, we provide guidance on the assessment of impacts on plant-, invertebrate-and microbial-species composition, abundance and diversity. For plants, we consider both aboveand below-ground diversity, as well as community-level dynamics in the form of propagule rain. With regard to species interactions, we cover pollination, vertebrate and invertebrate herbivory, plant predation and pathogens, and decomposition. We also provide a short motivation for, and link to, the plant traits protocol (Pérez-Harguindeguy et al., 2013). This chapter does not cover organism responses at the individual level, which are dealt with in chapter 5 on stress physiology (see below).

| Chapter 5: Stress physiology
In uncovering how climate change will affect plants and ecosystems, ecophysiology provides the means to mechanistically predict when tolerance limits are exceeded, and therefore when changes in  (Neuner & Buchner, 2012) and scaling up from leaves to whole organisms or even ecosystems.
In the stress physiology chapter (Supporting Information S5), we describe a series of physiological or related measurements that are useful in climate change research. We focus mostly on their use as indicators of stress, attained through determination of compounds (e.g. chlorophyll and carotenoid content, non-structural carbohydrates), plant functional traits (e.g. reflectance, leaf hydraulic conductivity, leaf thermal properties, stable isotopes of carbon and water) and measurements that directly characterize or assess stress and tolerance.

| The case for standardized data collection and reporting
Variation in the reporting of basic site and experimental characteristics or in measurement methodologies substantially reduces the potential data (re-)use for analyses and comparisons across studies.
Comprehensive comparisons across studies are therefore often hampered by (a) unrecorded factors increasing the unexplained variance in the data, (b) incompatible reporting leading to loss of information in the data (e.g. necessitate a simplification from quantitative to factor or qualitative variables) and (c) failure to report key (co)variates reducing the sample size (e.g. if studies have to be excluded from a synthesis) (Gerstner et al., 2017;Haddaway & Verhoeven, 2015). Using standardized protocols and reporting -to the extent that these are appropriate for the individual study -will therefore substantially increase the potential downstream use of data and added value of individual studies (affecting both filters; Figure 1).
Another challenge relates to specific parameters that are important for syntheses and meta-analyses, but which are often not measured or reported in stand-alone studies (affecting the data availability filter; Figure 1). For example, nutrient availability and nutrient dynamics may not be important in all individual studies but are increasingly being recognized as key drivers of across-site variation in ecosystem properties, dynamics and responses to climate and global changes (De Graaff, Van Groeningen, Six, Hungate, & Van Kessel, 2006;Janssens et al., 2010;Ren et al., 2017;Stevens et al., 2015;Terrer, Vicca, Hungate, Phillips, & Prentice, 2016;Vicca, Luyssaert, et al., 2012). Carbon-cycle models have traditionally overlooked the role of nutrient availability, but the overwhelming evidence of its dominant role is now encouraging the modelling community to address the roles of nitrogen and, more recently, phosphorus (Goll et al., 2012;Peñuelas et al., 2013;Wang, Law, & Pak, 2010). While the scientific community is increasingly acknowledging the importance of nitrogen, phosphorus and other nutrients for understanding and projecting the carbon cycle, there is still a significant lack of informative and comparable datasets at regional and global scales (Vicca et al., 2018). In the Site characteristics and data management chapter, we therefore provide a section on what variables to measure to enable disentangling the role of nutrients in carbon synthesis studies (summarized in Table 2).

| The case for using standardized and relevant units and scales
Several of the key variables and covariates for inter-study comparisons (  Table 2; affecting the compatibility filter, Figure 1). For example, changes in soil organic carbon in response to a manipulation are typically reported either per unit area or as a weight percentage. These metrics are both valid, but they are not comparable unless the data necessary for conversion (bulk density and sampling depth) are provided. The necessary information for recalculation or conversion across reporting traditions should therefore be recorded and reported.
A second challenge relates to the need to quantify the treatments as experienced by the biota. For example, precipitation manipulation experiments typically report the amount of water added or removed, but the manipulation as experienced by the biota may deviate substantially from what is reflected in the absolute or percentage change in precipitation. Specifically, soil water availability is influenced by many factors, including soil water-holding capacity, run-off, hydrological legacy, rooting depth and drainage (Vicca, Luyssaert, et al., 2012). Hence, assessing water availability in a standardized way will substantially improve our understanding of the sensitivity of ecosystems to the manipulations and facilitate cross-experimental comparisons (see protocol 3.8 on Ecosystem water stress).

| The case for upscaling through models
Various types of mechanistic models can be used to generalize and extrapolate the data collected from in situ climate change studies in space and time. The potential of model-data interaction and its potential high impact, however, is often forgotten during experimental planning. Here, we want to illustrate the importance of early project planning for future data use (Table 2). For example, soil pH as an easy and low-cost measure may have been traditionally measured at the site level (lowest resolution), but in order to be a useful variable in meta-analyses, ecosystem models and temporal and spatial upscaling, soil pH data are more valuable if measured at a higher resolution (e.g. treatment or plot level). Thus, investing time in considering the aspired impact of the data-to-becollected already in the project planning phase can direct budget investments and will be beneficial for the wider experimental and modelling community.
Purely statistical models attempt to describe cause-and-effect based on independent measurements of key processes. On the other hand, process-based models are built on a theoretical  (Luo et al., 2012). Each of these strategies has its place, but the difference in objectives and data requirements needs to be recognized.
To give an example, sensitivity analyses often identify similar variables as important for similar models. The BIOME-BGC model, which describes biogeochemical processes, is especially sensitive to leaf area index, cumulative soil water-holding capacity, leaf mass to area ratio, temperature and some measure of nutrient status (White, Thornton, Running, & Nemani, 2000). with changes in temperature and precipitation, the model predicted, for example, that boreal forest, boreal steppe and tundra would lose significant area. By measuring and reporting the variables needed to meet the data requirements of different types of models (Table 2), the information flow between empirical studies and modelling will be increased.

| The case for considering new mechanisms and drivers under future climates
As the climate warms and rainfall patterns intensify in many regions across the globe, mechanisms and driving factors that are currently of minor importance could become more dominant in the future. Thus, new variables might need to be considered for measurement protocols to adequately track modifications in ecosystems under a changed climate. In particular, drought events are expected to occur more often, to last longer or to be more severe in many regions (Prudhomme et al., 2014;Touma, Ashfaq, Nayak, Kao, & Diffenbaugh, 2015). Conversely, increased rainfall intensities and flash flooding during the growing season are also predicted (Kendon et al., 2014), leading to a short-term rise in the availability of water. Drier conditions could have consequences for the drivers and mechanisms operating in biogeochemical cycles. For example, rain might periodically be absent, thus driving the ecosystem to pass a threshold and start relying on non-rainfall water inputs (occult precipitation, such as dew and fog) for plant, litter and soil processes (Gliksman et al., 2017;Hill, Dawson, Shelef, & Rachmilevitch, 2015;McHugh, Morrissey, Reed, Hungate, & Schwartz, 2015). Moreover, abiotic processes, such as photochemical and thermal processes, start to kick in after passing heat and drought thresholds (Austin & Vivanco, 2006;McCalley & Sparks, 2009). Methodologies need to be adapted to capture such changes in drivers and mechanisms. For example, litterbags to study decomposition should transmit UV radiation and should only minimally affect the energy balance of litter to prevent interfering with dew formation. In addition to the need of measurement devices to consider these mechanisms and allow their quantification, the relevant drivers have to be measured with affordable equipment, such as UV irradiance by UV sensors and duration of dew by leaf wetness sensors. In general, this illustrates how novel future climates and conditions will necessitate the consideration of new drivers and mechanisms.

| CON CLUS IONS
Climate change experiments and observational studies are increasingly being conducted across a broad range of habitats and regions.
Experiments are generally designed to disentangle complex and interactive drivers, causal relationships, response surfaces and underlying mechanisms, whereas observational studies document large-scale temporal and spatial patterns. The growing interest in data re-use, synthesis and upscaling within and across experiments and observations highlights challenges and pitfalls regarding the downstream use of knowledge from these studies, especially regarding data availability and data compatibility across studies (Denny et al., 2014;Firbank et al., 2017;Vicca, Gilgen, et al., 2012) and the lack of basic study information or covariates necessary for comparison (Gerstner et al., 2017;Haddaway & Verhoeven, 2015;Halbritter et al., 2018).
To overcome these challenges, we provide a handbook with guidance on the selection of response variables, protocols for standardized measurements of these variables and advice on data reporting and management. The ClimEx handbook summarizes best-practice methodologies emerging from major ecological research networks and studies, as synthesized by 115 experts from across relevant research fields. Our ambition is that this community effort will facilitate awareness by the next generation (and the older generation) climate-and global-change scientists of standardized methods to promote data availability, compatibility and transparency. We envision improved research practices that will not only result in better returns on the time and money invested in individual research projects, but will especially facilitate second-order research outputs across projects and scientific disciplines. More general, this should lead to a better connection of scientific communities and disciplines. The ultimate goal is to optimize the quality and impact of our science to fulfil society's needs in a changing world.

ACK N OWLED G EM ENTS
We thank the ClimMani core group for the opportunity to realize this handbook, Linn Vassvik, Christine Pötsch, William Erazo Garcia and Siri V. Haugum for assisting in the collection of existing protocols, and Cathy Jenks for reading all the protocols thrice, checking the language, flow, logic and all references. The authors were funded by different national and international research agencies while contributing to this work, but in the interest of the word count we are not mentioning these individually. This study was funded by the ClimMani COST Action ES1308 (2014-2018).

S U PP O RTI N G I N FO R M ATI O N
Additional supporting information may be found online in the