The role of field margins in microclimatic temperature buffering of temperate Canadian landscapes

The role of field margins in microclimatic temperature buffering of temperate Canadian landscapes | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 7 October 2025 V1 Latest version Share on The role of field margins in microclimatic temperature buffering of temperate Canadian landscapes Authors : Sarah Endicott 0000-0001-9644-5343 , Adam Blasl , Marie-Hélène Brice , and Ilona Naujokaitis-Lewis 0000-0001-9504-4484 [email protected] Authors Info & Affiliations https://doi.org/10.22541/au.175986071.14914813/v1 241 views 206 downloads Contents Abstract Introduction Methods Microclimate and macroclimate data Environmental variables Data analyses Results Discussion Acknowledgements Author contributions Data Statement Appendix Figures References Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Field margins in agricultural landscapes exhibit considerable structural variation, ranging from herbaceous strips to shrubby borders and tree-rich hedgerows, which can influence microclimatic conditions and wildlife habitat quality. Despite their ecological importance, the microclimatic properties of these diverse field margins remain poorly characterized. By generating fine-scale thermal heterogeneity, field margins may act as microrefugia, buffering organisms against temperature extremes and supporting climate adaptation. We investigated how local vegetation structure and landscape context influence microclimatic variation in a temperate Canadian agricultural landscape. From June to September 2018, we deployed 180 temperature data sensors across 30 fields in 20 1-km2 landscapes in eastern Ontario, Canada, recording conditions in both field margins and adjacent crop fields. We quantified daily maximum and minimum temperature offsets between margins and fields, compared these to regional weather station data, and assessed the sensitivity of measurements to solar radiation shielding. Linear mixed-effects models showed that tree-rich margins strongly reduced maximum temperatures and elevated minimum temperatures, while herbaceous and shrubby margins resulted in limited thermal buffering. Margin width and landscape-level natural vegetation cover had no detectable effects. Local microclimate measurements were consistently more extreme than regional weather data, highlighting the importance of fine-scale monitoring. Although radiation shielding reduced the magnitude of temperature offsets, the overall direction of effects remained consistent. Our findings underscore the ecological value of conserving and expanding treed field margins, which can provide microrefugia and enhance wildlife resilience under climate change. Introduction In an era of rapid climate change, identifying areas and landscape features that remain relatively buffered from temperature extremes may reduce risks associated with climate warming (Lenoir et al., 2017; Riddell et al., 2021; Suggitt et al., 2015). Forests, in particular, buffer temperatures, resulting in understory microclimates that are different from free-air macroclimates as measured by weather station data (Chen et al., 1999; De Frenne et al., 2021, 2019; Scheffers et al., 2014). Forest microclimates support microrefugia, areas which enable species to persist locally due to suitable microclimate conditions that remain buffered from climate change impacts (Ashcroft, 2010; Dobrowski, 2011; Kemppinen et al., 2024). The microrefugia potential of larger tracts of contiguous forests and forest dominant ecosystems is well documented (De Frenne et al., 2021; Greiser et al., 2018; Haesen et al., 2023; Morelli et al., 2020; Wei et al., 2024; Zellweger et al., 2020). However, with much of the global land mass consisting of agriculture (Foley et al., 2005; Ramankutty et al., 2008), understanding the contributions of individual trees, smaller forest remnants, such as hedgerows, and other non-crop vegetation within agricultural landscapes to creating buffered microclimates and their microrefugia potential remains a key priority to inform climate resilience in agricultural systems. Agricultural landscapes are often highly heterogeneous reflecting a diversity of land uses and land cover types, which can affect the availability of microclimates. There is substantial regional variation across agricultural systems, with these human-dominated landscapes varying from intensive agricultural systems with large fields of monoculture crops and few non-crop cover types to landscapes consisting of a diversity of crops and non-crop ‘natural’ covers, such as remnant forest patches, wetlands, and field margins (Foley et al., 2005; Tilman et al., 2001). Field margins, defined as relatively narrow linear strips of natural and semi-natural vegetation that separate crop or pasture fields, are common features of many temperate climate zones that contribute to a diversity of ecosystem services (Burel, 1996; Haddaway et al., 2018; Montgomery et al., 2020). Field margins, and treed hedgerows specifically, function as windbreaks and reduce soil erosion, resulting in improved crop yields (Bendall et al., 2022; Forman and Baudry, 1984; Haddaway et al., 2018). Other functions include pest regulation and pollination services (Montgomery et al., 2020; Poveda et al., 2012; Précigout and Robert, 2022), ecological corridors (Forman and Baudry 1984), and habitat for biodiversity (Montgomery et al., 2020; Poveda et al., 2012; Précigout and Robert, 2022). With accelerating climate change, benefits of non-crop natural vegetation in agricultural landscapes extends to their role in climate change mitigation as demonstrated by their carbon sequestration potential (Drexler et al., 2024). From a climate change adaptation perspective, field margins have the potential to introduce microclimate heterogeneity within agricultural landscapes thus contributing to microrefugia for biodiversity by providing thermally buffered habitats (Keppel et al., 2015) and contributing to local climate regulation (Burkhard et al., 2012). While the role of larger contiguous forests in creating unique microclimates has gained traction recently (De Frenne et al., 2021; Haesen et al., 2023; Jucker et al., 2020), the contribution of field margin vegetation to microclimate heterogeneity in agricultural landscapes remains less well studied (Haddaway et al., 2018). Local scale attributes of the field margin, such as vegetation structure and composition, can affect the microclimates it creates (Ghafarian et al., 2024; Sánchez et al., 2009; Vanneste et al., 2020). In a continent-wide study in Europe, Vanneste et al (2020) found that hedgerows with taller, wider, and more dense tree cover produced cooler microclimates that buffered temperature extremes, though buffering within hedgerows was less than in adjacent forest woodlots. The few studies assessing field margin microclimates in an agricultural context predominantly focus on woody vegetation (Ghafarian et al., 2024; Sánchez et al., 2009; Vanneste et al., 2020), yet, field margin vegetation can vary from herbaceous plants to shrubs to trees depending on factors including land conversion history and agricultural management practices. All studies to date have taken place in Europe, which has a long history of the use of hedgerows in particular (Collier, 2021; Forman and Baudry, 1984), however other agricultural landscapes reflect different land use conversion histories and different past and potential future climate changes. Thus, gaps remain in our understanding of the effect of the full range of field margin vegetation on microclimate and the applicability of previous studies to other regions. The mosaic nature and spatial complexity associated with agricultural landscapes necessitates understanding how landscape-scale factors influence microclimates in addition to local scale factors. The microclimate influence of vegetation associated with field margins can extend beyond the spatial boundaries of the margin itself. For example, small woody vegetation associated with field margins can cool land surface temperatures of adjacent crop fields (Ghafarian et al., 2024). At the regional scale, land use and land cover can affect mesoclimate and local temperatures. The impact of agriculture and forest land covers on surrounding landscape climate is a complex combination of cooling by evapotranspiration, warming by lower albedo and changes to wind speed (Mahmood et al., 2014). These processes are influenced by the type and structure of vegetation as well as moisture availability. Several recent studies have found that landscape scale forest land cover has a cooling effect (Borderieux et al., 2023; Peng et al., 2020; Zimmermann et al., 2024). For example, Peng et al. (2020) found that the amount and configuration of different land cover types affected temperature and that water, forestland and shrubland contributed to cooling. These landscape scale effects may modify the impact of field margin vegetation on temperature but have not been included in previous studies on this topic. Recent research on the measurement of microclimates has created some debate about the correct way to measure air temperatures and compare them across different locations (Ashcroft, 2018; Maclean et al., 2021; Terando et al., 2018, 2017). When determining whether a landscape feature contributes to a distinct microclimate, conditions within that feature are compared to a reference environment, which represents the macroclimate (De Frenne et al., 2021; Zellweger et al., 2024). Weather stations measure the temperature of the air 2 m above the ground, are shielded against solar radiation and placed to minimise impacts of local terrain or vegetation (Bramer et al., 2018; Maclean et al., 2021). This coarse scale weather station data (i.e. macroclimate) does not reflect the range of microclimate conditions that organisms experience in the near-surface, vegetation-mediated habitats where they typically reside (Bramer et al., 2018; Pincebourde et al., 2007; Potter et al., 2013). Interest in microclimates has led to the use of compact temperature sensors (e.g. HOBO loggers, iButtons, etc.), in place of weather station based measurements, to capture temperatures at the scale and in the microhabitat of organisms (De Frenne et al., 2024). These sensors have been used for a wide array of applications including measuring soil, air, surface and water temperature (see Table S4 in Terando et al., 2017), and have been frequently used to compare microclimate air temperature to air temperature measured by weather stations (Gilbert et al., 2022; Latimer and Zuckerberg, 2017; Vanneste et al., 2020). However, Maclean et al (2021) and Terando et al (2017) both raise concerns that temperatures recorded by unshielded or poorly shielded compact sensors, that are exposed to solar radiation, are biassed when compared to air temperature measured by weather stations. Maclean et al (2021) recommends ultrafine­wire thermocouple temperature sensors be adopted in order to measure air temperature without the influence of solar radiation. Conversely, in some contexts solar radiation is seen as an important component of the thermal environment that organisms experience and the availability of shade as an important factor in the ability of species to thermoregulate (Elmore et al., 2017; Kearney et al., 2009). Ashcroft (2018) argues, in response to Terando et al (2017), that since organisms are exposed to solar radiation, reducing its impact on temperature sensors through shielding could bias instruments towards cooler temperatures. This ongoing debate emphasises the importance of understanding the consequences of different methodological choices related to air temperature measurements on the interpretation of outcomes and ensuring that these methodological decisions reflect a study’s goal. Here, we use local fine-scale temperature monitoring to investigate the effects of local and landscape features on the temperature offsets produced by field margins in an agricultural landscape in Ontario, Canada. Our goal is to determine the amount of temperature buffering that hedgerows provide compared to conditions in adjacent agricultural fields. To that end we measure daily and hourly temperatures in fields and field margins using both shielded and unshielded microclimate temperature sensors and use them to calculate offsets. Since it is a common practice, we also calculate offsets relative to regional weather stations and discuss the implications of this decision. We then assess how the temperature offsets of field margins are influenced at the local level by field margin height and width as well as at the landscape level by proportion of natural vegetation. In addition, we compare measurements for shielded and unshielded sensors to assess how different methodological decisions influence our results. We expect that taller, wider field margins will be cooler than fields and weather stations located in open areas, and that higher proportions of natural vegetation in the surrounding landscape will have a cooling effect. We also expect that shielded and unshielded sensors will yield significantly different results when deployed in unshaded environments. Methods Study area and site selection Our study was located across a 10,000 km 2 area of southeastern Ontario, Canada, 114 m above sea level (45°19’ N, 75°40’ W) (Fig. 1a). The region is largely agricultural with corn and soybean dominating the landscape. These two crops are typically rotated in alternate years, with wheat, barley and forage crops, such as alfalfa, clover, and hay, occasionally included in rotations (Canada, 2017). Although the region has undergone significant conversion to agriculture, urban, and suburban uses, remaining forested vegetation consists of mixed deciduous and coniferous forests. The mean annual temperature in this region is 6.4 °C, and mean annual precipitation is 943.4 mm. During the summer months (June through to end of September) the mean temperature is 18.5 °C and mean total precipitation is 360.3 mm (https://climateatlas.ca/map/canada). We selected 20 1 km 2 landscapes that maximised variation in the average size of crop fields and captured variation in the amount of natural land-covers surrounding fields and across the landscape (Fig. 1b). Landscape site selection was constrained to include a subset of sites used in Fahrig et al. (2015) in part to take advantage of ongoing parallel studies on pollinators. Each landscape was a minimum distance of 3.5 km from any other landscape. Within the 20 sites, we selected one to two fields, resulting in a total of 30 fields to assess microclimate heterogeneity. Temperature sensors were deployed along transects located within a focal crop field and an adjacent field margin. At the field-scale we applied multiple criteria to select paired field-field margin sampling locations. First, we selected fields that were bordered by a minimum of three vegetated field margins, and where the adjacent land-use to the focal field was another crop field and where the field margin did not include a drainage ditch or any other type of waterway. Second, we selected field margins across all sites to represent a range of vegetation heights. Specifically, field margin vegetation varied from herbaceous vegetation to woody vegetation including shrubs and trees and captured both a range of vegetation composition and vegetation heights. To the extent possible we selected fields where all field margins surrounding the field were of similar height and broad vegetation type. Third, we aimed to select only soybean fields due to their lower crop height for deployment of field temperature sensors, however, when this was not possible other low-height crops (e.g. hay, wheat, barley, etc.) were chosen. Corn fields were avoided entirely due to their tall height potential, that would grow to cover the sensors and alter the field temperature measurements. Figure 1. a) Map of the study area in eastern Ontario, Canada including the locations of the 7 weather stations and 20 sampling sites. Landcover is based on the Southern Ontario Land Resource Information System Version 3.0. Source: Ontario Ministry of Natural Resources and Forestry © King’s Printer for Ontario, 2023 b) Examples of 1 by 1 km landscapes that sampling sites were located within, showing decreasing mean field size from left to right at the landscape scale and increasing proportions of natural area from left to right. Satellite image source: Imagery ©2022 Google, Imagery ©2022 CNES/Airbus, Maxar Technologies c) A diagram showing the placement of sensors along the margin and in the field. d) A diagram showing the sensor placement within the margin and the range of margin vegetation heights that were included. Microclimate and macroclimate data Within each focal field, we created two 50 m transects, one in the margin and another 50 m into the crop field. Along each transect we deployed three sensors every 25 m (Fig. 1c). When the field size was smaller than 100 m in length, the transect was placed in the centre of the field to avoid being too close to the opposite hedgerow. A total of 180 temperature sensors were installed within 30 paired transects, and were deployed in 2018 between June 11 th and June 26 th , and removed between September 14 th and September 21 st , generally reflecting the location’s summer season. We used HOBO temperature loggers (112 loggers of model MX2201 and 68 loggers of model UA-002-08; range -20°C to 70°C; accuracy ±0.5°C and ±0.53°C, respectively). All temperature sensors were assessed for accuracy using a water bath and an independent thermometer in the lab prior to deployment. Each temperature sensor was mounted 1 m above the soil surface on a wooden stake and was configured to record at 30 minute intervals. Temperature sensors in the field were oriented towards the margin and margin sensors were oriented towards the field to minimise variation, and field margins were selected to minimise variation in general orientation (i.e. transects were selected to run in a north-south direction, as best as possible). To test the effect of shielding sensors, along each transect two temperature sensors were unshielded and one was shielded using Onset RS1 Solar Radiation Shields, resulting in 60 shielded and 120 unshielded sensors (Fig. 1c). We obtained macroclimate temperature data for each study site using nearby weather stations from Environment and Climate Change Canada. We selected all weather stations that were located within a buffer of 30 km around the entire study area resulting in seven weather stations (Fig. 1a). Using the R package “weathercan” (LaZerte and Albers, 2018), we retrieved hourly temperature data from the four weather stations for which it was available and daily mean, minimum and maximum temperatures from all seven stations for the time period of the study (June 27, 2018 to September 13, 2018). We calculated the macroclimate temperature as the inverse distance weighted mean across the weather stations for each temperature measure at each study site. Environmental variables We measured field margin height using a clinometer from each of the three sensor locations in the field and then took the average. We took the average of the measured field margin width taken at each of the three sensors located within the field margin transect. The width of the margin extended from the edge of the sampled focal field, through the field margin, and to the edge of the adjacent field’s crop. The proportion of natural vegetation in each 1 km 2 site was based on field surveys of each focal landscape in 2018. We used maps based on aerial imagery at 40 cm resolution which was flown in 2011 and 2012 (see Fahrig et al., 2015 for detailed description) and updated manually based on our field observations. Our natural land cover layer consisted of non-cropped areas including field margins, as well as forests, wetlands and open water. Table 1 : Description of all local and landscape variables used in the linear mixed-effect models. Field margin height Height of each field margin was measured using a clinometer with the observer standing at each of the three field sensor locations and averaged (m). Field Field margin width Field margin width measured at the three sensor locations and averaged (m). Field % natural Proportion of natural vegetation within 1 km 2 landscape containing the site. Landscape Macroclimate temperature Inverse distance weighted mean of the temperature from weather stations within 30 km of the study area. Regional Data analyses We evaluated the differences between microclimates and (regional) macroclimate by comparing temperatures recorded by sensors located in fields and margins to each other and to the macroclimate temperature derived from weather station data. Temperature differences were quantified by calculating three temperature offsets: \begin{equation} T_{offset\ FM\ -\ F}=T_{\text{field\ margin}}-T_{\text{field}}\nonumber \\ \end{equation}\begin{equation} T_{offset\ FM\ -\ M}=T_{\text{field\ margin}}-T_{\text{macro}}\nonumber \\ \end{equation}\begin{equation} T_{offset\ F\ -\ M}=T_{\text{field}}-T_{\text{macro}}\nonumber \\ \end{equation} The temperature offset is the difference between temperatures measured by different sensors, while buffering refers to when conditions in a particular microclimate are less extreme than temperatures in a different microclimate or than the macroclimate. Of the 180 temperature sensors, seven failed and were removed. In addition, the time series of 19 others were truncated (15 after 2018-08-19, three after 2018-09-09 and one before 2018-07-13), thus the total number of temperature sensors varied between 155 and 173 at different points in the season. When there were two unshielded sensors available at the same site and location (which was the case for most fields), the temperature time series were averaged. Microclimate temperatures based on temperature sensors were truncated to the period of June 27 th to September 13 th , 2018 to capture the dates when all sensors were deployed. We summarised the 30 minute temperature readings for each sensor into daily minimum, mean and maximum temperature as well as the daily average absolute deviation. Average absolute deviation is the mean of the differences between each data point and the median, it is a measure of the variability in temperature throughout a day. To test for differences between temperatures recorded at different locations we performed paired Wilcoxon signed rank tests to determine if the difference between two sensors was symmetrical around zero and to estimate the median offset and 95% confidence intervals. For each daily temperature measure (minimum, mean and maximum) we grouped the data by the location of the sensor (field margin, field, or macroclimate) and whether the sensor was shielded. For sensors located in the field margin we also compared data grouped by whether the sensor was shielded and whether the field margin was short ( 7.5 m). A height of 7.5 m is considered the minimum height for shrubs and small trees (Petrides, 1972), and this value coincided with natural breaks in our vegetation height data (Fig. S1). We assessed how the temperature offset between field margins and fields changed as a function of local and landscape factors. For each paired transect, \(T_{offset\ FM\ -\ F}\) was calculated for each of the three temperature components: daily mean, minimum and maximum temperatures. Negative offset values reflect cooler temperatures in field margins compared to fields while positive values indicate warmer temperatures in field margins. As such we interpret negative offset values for maximum temperature and positive offset values for minimum temperature as indicative that field margins are providing buffering against extreme temperature values in the surrounding landscape. We fit linear mixed-effect models (LMMs) for each daily temperature metric with \(T_{offset\ FM\ -\ F}\) as the response variable and field margin height, field margin width, percent natural vegetation in the landscape, and macroclimate temperature as fixed effects (Table 1). We also included interactions between field margin height and field margin width, field margin height and percent natural vegetation, and field margin width and percent natural vegetation. We included site ID as a random effect to account for non-independence between repeated samples from the same site. Separate models were fitted for unshielded and shielded sensors. Based on exploratory data analysis, field margin height showed a stepped relationship to temperature offset. To explore this, we fit two versions of the model, one with a quadratic term for field margin height, and one where field margin height was a binary variable of whether the height was greater than or less than 7.5 m tall. We compared the two versions of the model based on the Akaike’s Information Criterion (AIC). To account for temporal auto-correlation in our data we added correlation structures to our models. We ran each model with (1) no correlation structure, (2) an auto-regressive model of order 1 (corAR1) auto-correlation structure, and (3) an autocorrelation-moving average correlation structure of order p = q = 1 (corARMA). We compared models using AIC (Zuur et al., 2009). We incorporated corARMA as a model parameter based on evidence that including corARMA auto-correlation structure improved model fits and reduced temporal auto-correlation (Fig. S4). We evaluated the models using the marginal R 2 for linear mixed effect models (Nakagawa and Schielzeth, 2013) and diagnostic plots. We verified multicollinearity among predictors in our models by calculating their variance inflation factors and ensuring they were less than 2, using the package “car” (Fox and Weisberg, 2019). The R package “nlme” was used to fit our LMMs (Pinheiro et al., 2023). All statistical analyses and visualisations were performed in the open source statistical software R version 4.4.1 (R Core Team, 2024). Results On average sensors in field margins recorded less extreme temperatures than sensors in fields across the surveyed agricultural landscape based on lower daily mean and maximum temperatures, higher daily minimum temperatures and less temperature variability over a day (ie lower diurnal range and average absolute deviation) (Fig. 2, Table S1). For example, the mean daily maximum temperature recorded by unshielded sensors in fields was 38.03 °C, while it was 32.59 °C for unshielded sensors in field margins, 28.73°C for shielded sensors in fields and 27.67 °C for shielded sensors in field margins. Macroclimate temperatures derived from weather stations were more similar to shielded sensors in field margins and less extreme than sensors in fields. The mean daily maximum temperature recorded by weather stations was 27.83 °C. Figure 2. Comparison of the distributions of daily temperatures between field margin divided into tall (>7.5 m) and short (< 7.5 m) vegetation height, field and macroclimate, as measured by weather stations, for unshielded (left) and shielded sensors (right). AAD is average absolute deviation. Note, the scale is the same across panels but different portions of the scale are shown in each. Hourly temperatures observed showed that the temperature offsets changed throughout the day (Fig. 3). Macroclimate temperature, derived from weather stations, was warmer than other sensors in the early morning and remained so until about noon for shielded sensors at which point field and field margin temperatures became slightly warmer. For unshielded sensors, temperatures measured in fields and field margins increased much faster, becoming warmer than macroclimate temperatures by 7h and returning to similar temperatures after 20h. Sensors in fields had the largest offset from the macroclimate while field margins showed smaller offsets on average, but much more variation among sites. Overall, the macroclimate temperature was the most stable throughout the day, which is also reflected by the lower average absolute deviation (Fig. 2, Table S1). Figure 3. Mean hourly temperature across all sites on July 1st for sensors in fields, field margins, and based on weather stations. Lines show plus and minus one standard deviation and indicate the variation among sites. Note that only four of the seven weather stations used in our study had hourly data available. Field margins provided significant buffering of extreme temperatures observed in fields, with tall hedgerows (> 7.5 m) providing the strongest buffering (Fig. 4a, Fig. S2, Table S2). Daily maximum temperatures in field margins were significantly cooler than in fields for both shielded (median offset = -0.987, CI = -1.051 - -0.922) and unshielded sensors (median offset = -5.491, CI= -5.738 - -5.248). This cooling effect was only present for tall field margins (Shielded: median offset = -1.652, CI = -1.716 - -1.587; Unshielded: median offset = -9.26, CI = -9.419 - -9.101) while temperatures in short field margins were slightly warmer or the same as in fields (Shielded: median offset = 0, CI = -0.043 - 0.043, Unshielded: median offset = 0.52, CI = 0.332 - 0.715). A similar pattern held for daily minimum temperatures, although the magnitude was smaller, where field margins were significantly warmer than fields for both shielded (median offset = 0.493, CI = 0.45 - 0.536) and unshielded (median offset = 0.956, CI = 0.908 - 1.005) sensors, but when tall and short hedgerows were considered separately only tall hedgerows were warmer (Shielded: median offset = 0.901, CI = 0.858 - 0.944; Unshielded: median offset = 1.578, CI = 1.531 - 1.628) and short hedgerows were the same or cooler (Shielded: median offset = -0.043, CI = -0.086 - -0.021; Unshielded: median offset = -0.024, CI = -0.052 - 0.003). Macroclimate temperatures derived from weather stations tended to be less extreme than temperatures recorded by sensors in fields and field margins (Fig. 4 b-c, Fig. S3, Table S2). For daily maximum temperature both fields and field margins were warmer than the macroclimate for unshielded sensors (Field margin: median offset = 4.64, CI = 4.374 - 4.891; Field: median offset = 10.206, CI = 10.075 - 10.337), while for shielded sensors fields were warmer (median offset = 0.789, CI = 0.742 - 0.838) and field margins were cooler (median offset = -0.236, CI = -0.296 - -0.174). When field margin were separated by height, tall field margins were only slightly warmer for unshielded sensors (median offset = 0.876, CI = 0.778 - 0.978) and cooler for shielded sensors (median offset = -0.886, CI = -0.939 - -0.833). On the other hand, short field margins were similar to fields with unshielded sensors being much warmer (median offset = 10.692, CI = 10.477 - 10.909) and shielded sensors only slightly warmer (median offset = 0.797, CI = 0.719 - 0.874) than the macroclimate. Unlike when compared to fields there was no buffering of minimum temperature in field margins compared to macroclimate temperatures (median offset CIs between -0.084 °C and -1.913 °C depending on shielding and field margin height). Figure 4. Comparison of daily temperature metrics by position of the sensor. a) Margin vs field, b) margin vs macroclimate, c) field vs macroclimate. In a) and b) the light green represents margins with vegetation 7.5 m tall. Coloured lines are univariate linear models of the relationship between the temperatures recorded by the two sensors. The 1:1 diagonal line indicates the expected value if the offset between the sensors is zero and points above the diagonal indicate a positive offset while points below have a negative offset. Temperature offsets ( \(T_{offset\ FM\ -\ F}\)) for daily minimum, mean and maximum temperature for both shielded and unshielded sensors were significantly affected by field margin height, with positive direction of effects for minimum temperature and negative effects for mean and maximum temperature, suggestive of a buffering effect where tall trees act as a thermal buffer against extreme temperatures (Fig. 5, Tables S3-4). Field margin width, landscape scale natural cover, and the three interaction terms (% natural vegetation:field margin width, % natural vegetation:tall field margin, tall field margin:field margin width) had negligible effects across all temperature metrics and shielding status (Fig. 5, Tables S3-4). Across all three temperature metrics and regardless of shield status, \(T_{offset\ FM\ -\ F}\) was significantly negatively affected by free-air macroclimate temperature, except for mean temperature of unshielded sensors where the effect of free-air macroclimate temperature was negligible (Fig. 5, Tables S3-4). Field margin height had a significant effect in both versions of the model but we present the results using the binary field margin height variable here since they are a better visual match for the raw data (Fig. 6). Results from the model with continuous field margin height are included in the Supplementary Material (Tables S5-6). The binary field margin height variable reflected the fact that\(\ T_{offset\ FM\ -\ F}\) appeared to have a threshold between 6.3 and 8.9 m and was close to 0 for field margin heights less than 7.5 m and was further from 0 for heights greater than 7.5 m (Fig. 6). Figure 5. Standardized regression coefficients and 95% confidence intervals from linear mixed-effect models estimating the effect of local and landscape variables on temperature offset metrics (minimum, mean, maximum) for both unshielded and shielded sensors. Temperature offsets represent the difference between field margins and fields ( \(T_{\text{field\ margin}}-T_{\text{field}}\) ). Positive regression coefficients (red) refer to covariates associated with warmer temperatures in the field margin, negative estimates (blue) refer to covariates associated with cooler field margin temperatures, and estimates in grey have confidence intervals that cross zero. Figure 6. Conditional effect of field margin height on \(\ T_{offset\ FM\ -\ F}\) ( = \(T_{\text{field\ margin}}-T_{\text{field}}\) ) with 95% confidence intervals for unshielded and shielded sensors based on the results of linear mixed-effect models. The effect shown is conditional on other variables in the model being held at their means. Points show the raw data and red lines indicate warmer temperatures in the hedgerow, while blue lines indicate cooler temperatures in the hedgerow. Grey lines indicate that the confidence interval crosses zero. Discussion Using a dataset of paired field and field margin summer temperatures across agricultural landscapes in Ontario, Canada, we show that treed field margins result in substantial buffering of high temperatures relative to herbaceous field margins or open agricultural fields. The difference in the mean across the sampling period of daily maximum temperature between hedgerows and fields, only 50 m apart, was 1.06°C for the shielded sensors and 5.44°C for unshielded sensors. Both the mean and maximum temperature were cooler in hedgerows compared to agricultural fields, while minimum temperatures were higher and AAD was lower, demonstrating that hedgerows, particularly tall hedgerows, provide a buffered thermal environment relative to agricultural fields. There were also temporal differences in temperature offsets with the largest differences occurring in the afternoon. In addition, our results demonstrate that methodological decisions about how to measure temperature, such as whether sensors are shielded, can have a large impact on results and conclusions. Overall, our results aligned with previous studies of forest and hedgerow microclimates. Our median maximum temperature offset for tall field margins, with macroclimate as the reference (-0.89 °C), was inline with Vanneste et al. (2020) (-0.72 °C) and Zelleweger (2024) (-0.7 °C) for hedgerows and, as expected was, smaller than their offsets in forests (-1.52 °C and -1.3 °C, respectively). On summer days trees intercept solar radiation and contribute to cooling by evapotranspiration, while also blocking the wind and preventing mixing with warmer air (Geiger et al., 1995). At night trees can act as insulation by intercepting the heat released from the ground (Geiger et al., 1995). This is consistent with our finding of the importance of the hedgerow height and other studies that have shown that taller vegetation provides greater temperature buffering (Carroll et al., 2016; Vanneste et al., 2020). Vanneste et al. (2020) also found hedgerow width was related to buffering capacity resulting in more stable temperatures but neither the field margin width nor the interaction of width and height were significant in our models. We included the interaction term because we expected width to be important for treed field margins with higher height but not for herbaceous field margins with low height. The fact that we did not detect an effect of width may be related to the low variation of widths available in our sites where 63% of hedgerows were between 5 and 15 m wide, meaning they are all likely subject to edge effects (Corbit et al., 1999). In addition to the effects of the vegetation structure in the field margin, we expected the proportion of natural cover in the surrounding landscape to affect the temperature offset between fields and field margins. However, the effect of landscape natural cover was negligible in our models. While temperate forests have cooler maximum daily surface temperatures (Hesslerová et al., 2013) and microclimates (De Frenne et al., 2019) during the growing season, it is less clear what their effect is on regional or global climate. Tropical forests help to cool the regional climate through evapotranspiration but boreal forests warm the climate due to lower albedo, temperate forests may also have an overall warming effect on regional climate (Lee et al., 2011; Mahmood et al., 2014). Recent work has shown that small woody features can influence surface temperatures in the surrounding landscape but that this effect is limited to the immediate surroundings (i.e. within 100 m) of the feature and is affected by the feature’s shape and orientation (Ghafarian et al., 2024). Thus, our variable of proportion of natural cover in the 1 km landscape may not have been detailed enough to pick up an effect. In general, all of our landscapes had fairly low natural cover (<40%) so it is possible an effect of proportion of natural cover would be detected across landscapes with a wider gradient. Solar radiation and wind speed play important roles in the way organisms experience their microclimate (Helmuth et al., 2010). However, weather stations are designed to limit the effect of these drivers, since their goal is to measure macroclimate air temperature (Lembrechts et al., 2019; Potter et al., 2013). Recent studies have raised concerns that temperature sensors and shielding techniques commonly used in ecology, and used here, are biassed by solar radiation and should not be directly compared to macroclimate air temperature (Gril et al., 2023; Maclean et al., 2021; Terando et al., 2017). Our results show that both shielded and unshielded sensors located in open areas, which one might expect to be similar to weather stations, recorded more extreme temperatures than weather stations. This indicates that these sensors are indeed biassed compared to weather stations, and, even when shielded, should not be used as a proxy for what a weather station would measure in the same location (but see Gril et al. (2023) about the potential to adjust biassed readings). However, when the goal is to determine the difference in how organisms will experience available microclimates (e.g. fields and hedgerows), our approach, using paired sensors with the same thermal properties and shielding, may give a more accurate reflection than a comparison to weather stations. Indeed, offsets were twice as large when shielded sensors in fields were used as the reference instead of weather stations (-1.66 °C and -0.89 °C, respectively) and ten times larger (-9.26 °C) when unshielded sensors in hedgerows and fields were compared. Thus, using weather stations as the reference would have ignored the importance of protection from solar radiation and underestimated the buffering provided by tall hedgerows. The availability of habitats that allow species to avoid overheating by seeking out shaded microsites can allow them to thermoregulate and persist in times of extreme heat (Elmore et al., 2017; Olsen et al., 2018). While the thermal properties of the shields and temperature sensors used here are not expected to reflect how any particular organism experiences solar radiation (Terando et al., 2018), at least one study has found that a compact temperature sensor had similar thermal properties to a life-like replica lizard (Vitt and Sartorius, 1999). Thus, an unshielded HOBO temperature logger may provide a more middle of the road estimate of temperatures experienced by organisms in fields and field margins than an ultra-fine thermocouple that is unaffected by solar radiation or a black bulb thermometer which likely overestimates the effect of solar radiation. Our results have important implications for the management of agricultural landscapes, indicating that maintaining existing treed hedgerows and planting trees in more field margins can provide refuge for organisms experiencing extreme heat. Conservation strategies for wildlife should consider vegetation structure not only in relation to nesting or foraging resources but also in terms of thermal habitats (Elmore et al., 2017). When managing for many species, or when the thermal niche of a species is not well known, managing for increased heterogeneity may be an effective strategy, as many species can modify their behaviour to find suitable thermal environments if they are available (Carroll et al., 2015; Elmore et al., 2017; Rakowski et al., 2019). Treed hedgerows provide structural heterogeneity leading to greater plant diversity with prairie species on the sunward/windward side and species adapted to cool and shade on shaded/leeward side (Précigout and Robert, 2022). While our results showed that herbaceous field margins do not provide as much thermal refuge as treed hedgerows at 1 m above the ground they may affect temperatures closer to the ground (Rakowski et al., 2019). In addition, both herbaceous and treed field margins provide many benefits in addition to temperature regulation, such as habitat for pest predators and pollinators (Albrecht et al., 2020), regulating temperature for adjacent crops (Bendall et al., 2022), providing habitat to help maintain biodiversity, and carbon storage (Drever et al., 2021). Thus, maintaining a mixture of treed and herbaceous hedgerows may be most beneficial to biodiversity. Forests are widely recognised as important potential climatic refugia and their protection and expansion are commonly recommended climate change adaptation and mitigation strategies (De Frenne et al., 2019; Drever et al., 2021; IPCC, 2023). Meanwhile, despite recent work highlighting their importance for biodiversity (Riva and Fahrig, 2022), the ability of hedgerows to buffer temperatures has received less attention. Given an expected increase in summer daily maximum temperature of 4.6 °C by 2080 under climate change for this region (https://climateatlas.ca/map/canada), the buffering provided by tall hedgerows of -1.7 °C for shielded sensors and -9.3 °C for unshielded sensors is significant. The additional benefits provided by hedgerows, including moderating temperatures for adjacent crops and sequestering carbon, mean that maintaining and increasing the cover of treed hedgerows is a climate change adaptation and mitigation strategy with potential benefits for both wildlife and agriculture. Our study builds on existing research and aims to clarify the influence of local field margin vegetation attributes relative to the landscape-scale amount of natural vegetation on the availability of microclimates. We expand the geographies studied to include field margins and hedgerows in North American agricultural landscapes, which have not previously been represented. Furthermore, our in-field microclimate logger deployments expand the data available from other studies to include human modified habitats rather than focusing solely on natural habitats (Zellweger et al. 2024). The outcomes of our research can be used to identify land management practices in agricultural regions that support a diversity of microclimates and potential refugia (i.e. thermally buffered habitats). Acknowledgements We would like to thank the landowners who allowed us to locate temperature loggers on their property. We would also like to thank Catherine Maclean for her assistance in the field work. Author contributions Ilona Naujokaitis-Lewis: Conceptualization, Methodology, Writing - Original Draft, Writing - Review & Editing, Supervision, Project administration, Resources, Funding acquisition. Sarah Endicott: Conceptualization, Writing - Original Draft, Writing - Review & Editing, Methodology, Formal analysis, Visualization. Marie-Helen Brice: Methodology, Formal analysis, Visualization. Adam Blasl: Conceptualization, Investigation, Methodology Data Statement All data and code required to reproduce this analysis is available at https://osf.io/4d9km/?view_only=432120fe42bf4d07b617357825f102d3 Appendix Tables Table S1 : Summary of temperatures recorded by unshielded and shielded sensors located in hedgerows and field centres, as well as inverse distance weighted mean temperatures for each field site based on weather stations within 30 km (n = 7). Mean, median, standard deviation (SD) and interquartile range (IQR) were calculated across all sites and days in the sample. Daily average absolute deviation (AAD) includes data from 4 weather stations because hourly data was not available for all stations. Daily min Margin - unshielded 14.64 15.06 4.41 6.70 Daily min Field - unshielded 13.66 13.92 4.62 7.28 Daily min Margin - shielded 14.79 15.18 4.34 6.61 Daily min Field - shielded 14.24 14.52 4.50 7.12 Daily min Weather station 15.44 15.83 3.97 5.76 Daily mean Margin - unshielded 22.70 22.93 3.67 4.29 Daily mean Field - unshielded 24.24 24.30 3.70 4.16 Daily mean Margin - shielded 21.26 21.47 3.31 3.93 Daily mean Field - shielded 21.48 21.68 3.38 4.06 Daily mean Weather station 21.65 21.83 3.49 4.32 Daily max Margin - unshielded 32.58 31.32 6.59 10.10 Daily max Field - unshielded 38.03 38.93 5.09 6.63 Daily max Margin - shielded 27.66 27.58 3.77 4.46 Daily max Field - shielded 28.73 28.87 3.65 4.33 Daily max Weather station 27.83 28.16 3.77 4.66 Daily AAD Margin - unshielded 6.90 6.19 3.92 4.80 Daily AAD Field - unshielded 9.35 9.04 4.26 6.77 Daily AAD Margin - shielded 5.51 5.18 2.66 3.82 Daily AAD Field - shielded 6.20 6.04 2.80 4.29 Daily AAD Weather station 3.62 3.36 1.49 2.36 Table S2 : Results of paired Wilcoxon signed rank tests for daily temperature metrics with shielded and unshielded sensors as well as tall hedgerows and short hedgerows for all combinations of field, hedgerow and macroclimate (Macro). Daily min No All Margin Field 0.956 0.909 1.006 2293 2312226.0 < 0.0001 < 0.0001 Daily min No All Margin Weather station -0.707 -0.765 -0.650 2293 545751.0 < 0.0001 < 0.0001 Daily min No All Field Weather station -1.740 -1.797 -1.683 2293 50632.0 < 0.0001 < 0.0001 Daily min Yes All Margin Field 0.493 0.450 0.536 2081 1695911.5 < 0.0001 < 0.0001 Daily min Yes All Margin Weather station -0.570 -0.621 -0.519 2081 486244.0 < 0.0001 < 0.0001 Daily min Yes All Field Weather station -1.139 -1.195 -1.082 2081 181393.0 < 0.0001 < 0.0001 Daily mean No All Margin Field -1.548 -1.614 -1.480 2293 258726.0 < 0.0001 < 0.0001 Daily mean No All Margin Weather station 0.895 0.810 0.980 2293 2038919.0 < 0.0001 < 0.0001 Daily mean No All Field Weather station 2.608 2.541 2.674 2293 2605325.0 < 0.0001 < 0.0001 Daily mean Yes All Margin Field -0.187 -0.205 -0.170 2081 504091.0 < 0.0001 < 0.0001 Daily mean Yes All Margin Weather station -0.416 -0.451 -0.381 2081 522472.0 < 0.0001 < 0.0001 Daily mean Yes All Field Weather station -0.194 -0.228 -0.159 2081 797597.0 < 0.0001 < 0.0001 Daily max No All Margin Field -5.495 -5.744 -5.253 2293 266066.5 < 0.0001 < 0.0001 Daily max No All Margin Weather station 4.638 4.372 4.890 2293 2399266.0 < 0.0001 < 0.0001 Daily max No All Field Weather station 10.209 10.077 10.339 2293 2630071.0 < 0.0001 < 0.0001 Daily max Yes All Margin Field -0.987 -1.051 -0.922 2081 276877.0 < 0.0001 < 0.0001 Daily max Yes All Margin Weather station -0.236 -0.296 -0.175 2081 881164.0 < 0.0001 < 0.0001 Daily max Yes All Field Weather station 0.790 0.743 0.839 2081 1907849.0 < 0.0001 < 0.0001 Daily min No < 7.5 m Margin Weather station -1.818 -1.913 -1.726 875 5123.0 < 0.0001 < 0.0001 Daily min No 7.5 m Margin Weather station -0.132 -0.179 -0.084 1418 419414.0 < 0.0001 7.5 m Margin Field 1.579 1.532 1.629 1418 1003385.0 < 0.0001 < 0.0001 Daily min Yes < 7.5 m Margin Weather station -1.232 -1.318 -1.146 821 18415.0 < 0.0001 < 0.0001 Daily min Yes 7.5 m Margin Weather station -0.192 -0.243 -0.142 1260 301231.0 < 0.0001 7.5 m Margin Field 0.901 0.858 0.944 1260 741719.0 < 0.0001 < 0.0001 Daily mean No < 7.5 m Margin Weather station 2.721 2.617 2.824 875 381347.0 < 0.0001 < 0.0001 Daily mean No 7.5 m Margin Weather station -0.017 -0.062 0.028 1418 491450.0 0.453 0.453 Daily mean No > 7.5 m Margin Field -2.580 -2.644 -2.516 1418 1128.0 < 0.0001 < 0.0001 Daily mean Yes < 7.5 m Margin Weather station -0.203 -0.259 -0.146 821 123105.0 < 0.0001 < 0.0001 Daily mean Yes 7.5 m Margin Weather station -0.553 -0.597 -0.510 1260 132813.0 < 0.0001 7.5 m Margin Field -0.333 -0.358 -0.309 1260 101965.0 < 0.0001 < 0.0001 Daily max No < 7.5 m Margin Weather station 10.692 10.477 10.909 875 383250.0 < 0.0001 < 0.0001 Daily max No < 7.5 m Margin Field 0.520 0.332 0.715 875 233316.0 < 0.0001 7.5 m Margin Weather station 0.877 0.779 0.979 1418 776244.0 < 0.0001 7.5 m Margin Field -9.265 -9.424 -9.106 1418 2.0 < 0.0001 < 0.0001 Daily max Yes < 7.5 m Margin Weather station 0.797 0.719 0.874 821 289611.0 < 0.0001 < 0.0001 Daily max Yes 7.5 m Margin Weather station -0.885 -0.938 -0.832 1260 100890.0 < 0.0001 7.5 m Margin Field -1.652 -1.718 -1.587 1260 10224.0 < 0.0001 < 0.0001 Table S3: Temperature offset model results for shielded sensors with hedgerow height included as a binary variable. Daily min Daily mean Daily max Margin height (Tall) 0.833 *** -0.316 *** -1.373 *** (0.369, 1.296) (-0.506, -0.126) (-2.206, -0.540) Margin width 0.207 0.058 -0.375 (-0.201, 0.616) (-0.109, 0.225) (-1.110, 0.360) % natural -0.066 -0.015 -0.253 (-0.590, 0.457) (-0.229, 0.199) (-1.194, 0.689) Macroclimate -0.225 *** -0.051 *** -0.087 ** (-0.262, -0.188) (-0.076, -0.027) (-0.156, -0.017) Margin height (Tall):Margin width -0.032 -0.064 0.270 (-0.504, 0.439) (-0.257, 0.128) (-0.579, 1.118) % natural:Margin height (Tall) -0.169 -0.135 0.082 (-0.769, 0.432) (-0.380, 0.110) (-0.998, 1.162) % natural:Margin width 0.073 0.012 -0.179 (-0.123, 0.269) (-0.067, 0.092) (-0.532, 0.175) Constant -0.262 -0.015 -0.168 (-0.657, 0.133) (-0.176, 0.146) (-0.875, 0.540) Random Effects: Residual variance 0.5 0.14 1.08 Intercept variance 0 0 0.4 ICC 0 0 0.27 N sites 27 27 27 Model statistics: Marginal R-squared 0.343 0.251 0.384 Observations 2,081 2,081 2,081 Log Likelihood -1,552.347 -256.242 -2,595.748 Akaike Inf. Crit. 3,128.694 536.484 5,215.497 Note: * p<0.1; ** p<0.05; *** p<0.01 Table S4: Temperature model results for unshielded sensors with hedgerow height included as a binary variable. Daily min Daily mean Daily max Margin height (Tall) 1.698 *** -2.669 *** -9.798 *** (1.330, 2.066) (-3.170, -2.167) (-11.516, -8.080) Margin width -0.028 -0.127 -0.677 (-0.368, 0.311) (-0.586, 0.332) (-2.258, 0.904) % natural -0.215 -0.052 -0.390 (-0.638, 0.209) (-0.627, 0.523) (-2.366, 1.585) Macroclimate -0.389 *** -0.015 -0.458 *** (-0.433, -0.345) (-0.082, 0.053) (-0.633, -0.282) Margin height (Tall):Margin width 0.170 0.105 0.740 (-0.220, 0.561) (-0.423, 0.633) (-1.079, 2.560) % natural:Margin height (Tall) 0.091 0.045 0.589 (-0.393, 0.575) (-0.611, 0.701) (-1.669, 2.847) % natural:Margin width -0.075 0.104 0.170 (-0.236, 0.087) (-0.113, 0.321) (-0.582, 0.923) Constant -0.491 *** 0.084 0.746 (-0.805, -0.176) (-0.342, 0.511) (-0.719, 2.211) Random Effects: Residual variance 0.42 1.1 7.66 Intercept variance 0.1 0 1.57 ICC 0.19 0 0.17 N sites 30 30 30 Model statistics: Marginal R-squared 0.564 0.615 0.726 Observations 2,293 2,293 2,293 Log Likelihood -1,981.438 -2,673.447 -5,033.326 Akaike Inf. Crit. 3,986.877 5,370.895 10,090.650 Bayesian Inf. Crit. 4,055.686 5,439.704 10,159.460 Note: * p<0.1; ** p<0.05; *** p<0.01 Table S5: Temperature offset model results for shielded sensors with hedgerow height included as a continuous variable. Daily min Daily mean Daily max Margin height 0.430 *** -0.101 ** -0.653 *** (0.248, 0.612) (-0.180, -0.022) (-0.955, -0.352) Margin height squared -0.223 * 0.082 0.522 ** (-0.454, 0.009) (-0.018, 0.183) (0.139, 0.905) Margin width 0.089 0.016 -0.092 (-0.114, 0.292) (-0.072, 0.103) (-0.428, 0.244) % natural -0.177 -0.127 ** -0.187 (-0.382, 0.028) (-0.216, -0.039) (-0.526, 0.152) Macroclimate -0.225 *** -0.051 *** -0.087 ** (-0.262, -0.188) (-0.076, -0.026) (-0.156, -0.017) Margin height:Margin width 0.114 -0.026 -0.060 (-0.091, 0.319) (-0.114, 0.062) (-0.399, 0.279) % natural:Margin height -0.106 0.0004 0.109 (-0.340, 0.128) (-0.101, 0.101) (-0.278, 0.496) % natural:Margin width 0.118 -0.022 -0.235 (-0.056, 0.292) (-0.097, 0.053) (-0.523, 0.052) Constant 0.409 *** -0.303 *** -1.467 *** (0.140, 0.677) (-0.418, -0.188) (-1.910, -1.024) Random Effects: Residual variance 0.5 0.14 1.09 Intercept variance 0 0 0.29 ICC 0 0 0.21 N sites 27 27 27 Model statistics: Marginal R-squared 0.355 0.235 0.425 Observations 2,081 2,081 2,081 Log Likelihood -1,555.488 -262.172 -2,596.686 Akaike Inf. Crit. 3,136.977 550.344 5,219.371 Note: * p<0.1; ** p<0.05; *** p<0.01 Table S6: Temperature offset model results for unshielded sensors with hedgerow height included as a continuous variable. Daily min Daily mean Daily max Margin height 0.720 *** -1.254 *** -4.701 *** (0.515, 0.925) (-1.556, -0.952) (-5.757, -3.645) Margin height squared -0.421 *** 0.312 1.774 ** (-0.676, -0.166) (-0.064, 0.689) (0.459, 3.090) Margin width 0.008 0.062 0.335 (-0.223, 0.239) (-0.279, 0.402) (-0.854, 1.525) % natural -0.147 0.019 0.226 (-0.377, 0.082) (-0.320, 0.357) (-0.958, 1.411) Macroclimate -0.389 *** -0.013 -0.456 *** (-0.433, -0.345) (-0.080, 0.054) (-0.631, -0.281) Margin height:Margin width 0.176 -0.054 -0.291 (-0.057, 0.409) (-0.397, 0.288) (-1.489, 0.908) % natural:Margin height -0.034 0.060 0.325 (-0.302, 0.233) (-0.333, 0.454) (-1.052, 1.701) % natural:Margin width 0.056 -0.087 -0.449 (-0.143, 0.256) (-0.381, 0.207) (-1.478, 0.579) Constant 0.888 *** -1.746 *** -6.493 *** (0.605, 1.171) (-2.160, -1.332) (-7.943, -5.042) Random Effects: Residual variance 0.42 1.19 7.67 Intercept variance 0.2 0.21 5.01 ICC 0.33 0.15 0.4 N sites 30 30 30 Model statistics: Marginal R-squared 0.49 0.522 0.636 Observations 2,293 2,293 2,293 Log Likelihood -1,991.952 -2,687.741 -5,044.579 Akaike Inf. Crit. 4,009.905 5,401.483 10,115.160 Bayesian Inf. Crit. 4,084.443 5,476.021 10,189.700 Note: * p<0.1; ** p<0.05; *** p<0.01 Figures Figure S1: Histograms showing the distribution of the covariates used in the model. Figure S2: a) Median \(T_{\text{offset}}\) ( \(T_{\text{hedgerow}}-T_{\text{field}}\) ) based on paired Wilcoxon signed rank tests for daily temperature metrics with shielded and unshielded sensors. b) Compares sensors located in tall hedgerows and short hedgerows. Figure S3: Median offsets between the microclimate and the macroclimate based on paired Wilcoxon signed rank tests for daily temperature metrics with shielded and unshielded sensors. a) Offsets for sensors located in fields and in field margins b) Offsets for sensors located in tall field margins and short field margins. Figure S4: Plots of the autocorrelation function for the normalized residuals of models fit with (1) no correlation structure, (2) an auto-regressive model of order 1 (corAR1) auto-correlation structure, and (3) an autocorrelation-moving average correlation structure of order p = q = 1 (corARMA). References 1. Albrecht, M., Kleijn, D., Williams, N.M., Tschumi, M., Blaauw, B.R., Bommarco, R., Campbell, A.J., Dainese, M., Drummond, F.A., Entling, M.H., Ganser, D., Arjen de Groot, G., Goulson, D., Grab, H., Hamilton, H., Herzog, F., Isaacs, R., Jacot, K., Jeanneret, P., Jonsson, M., Knop, E., Kremen, C., Landis, D.A., Loeb, G.M., Marini, L., McKerchar, M., Morandin, L., Pfister, S.C., Potts, S.G., Rundlöf, M., Sardiñas, H., Sciligo, A., Thies, C., Tscharntke, T., Venturini, E., Veromann, E., Vollhardt, I.M.G., Wäckers, F., Ward, K., Wilby, A., Woltz, M., Wratten, S., Sutter, L., 2020. The effectiveness of flower strips and hedgerows on pest control, pollination services and crop yield: a quantitative synthesis. Ecol. Lett. 23, 1488–1498. https://doi.org/10.1111/ele.13576 Ashcroft, M.B., 2018. Which is more biased: Standardized weather stations or microclimatic sensors? Ecol. Evol. 8, 5231–5232. https://doi.org/10.1002/ece3.3965 Ashcroft, M.B., 2010. Identifying refugia from climate change. J. Biogeogr. 37, 1407–1413. https://doi.org/10.1111/j.1365-2699.2010.02300.x Bendall, E., Westgate, M., Haddaway, N., Lindenmayer, D., 2022. Tracking research trends on the effects of vegetative strips within agricultural landscapes: A systematic map update. https://doi.org/10.22541/au.164410852.23208571/v1 Borderieux, J., Gégout, J., Serra‐Diaz, J.M., 2023. High landscape‐scale forest cover favours cold‐adapted plant communities in agriculture–forest mosaics. Glob. Ecol. Biogeogr. 32, 893–903. https://doi.org/10.1111/geb.13676 Bramer, I., Anderson, B.J., Bennie, J., Bladon, A.J., Frenne, P.D., Hemming, D., Hill, R.A., Kearney, M.R., Körner, C., Korstjens, A.H., Lenoir, J., Maclean, I.M.D., Marsh, C.D., Morecroft, M.D., Ohlemüller, R., Slater, H.D., Suggitt, A.J., Zellweger, F., Gillingham, P.K., 2018. Advances in monitoring and modelling climate at ecologically relevant scales. Adv. Ecol. Res. 58, 101–161. https://doi.org/10.1016/bs.aecr.2017.12.005 Burel, F., 1996. Hedgerows and their role in agricultural landscapes. Crit. Rev. Plant Sci. 15, 169–190. https://doi.org/10.1080/07352689.1996.10393185 Burkhard, B., Kroll, F., Nedkov, S., Müller, F., 2012. Mapping ecosystem service supply, demand and budgets. Ecol. Indic., Challenges of sustaining natural capital and ecosystem services 21, 17–29. https://doi.org/10.1016/j.ecolind.2011.06.019 Canada, S., 2017. Table 32-10-0416-01 field crops and hay, census of agriculture, 2011 and 2016, inactive. https://doi.org/10.25318/3210041601-eng Carroll, J.M., Davis, C.A., Elmore, R.D., Fuhlendorf, S.D., Thacker, E.T., Parmenter, R., 2015. Thermal patterns constrain diurnal behavior of a ground-dwelling bird. Ecosphere 6. https://doi.org/10.1890/ES15-00163.1 Carroll, J.M., Davis, C.A., Fuhlendorf, S.D., Elmore, R.D., 2016. Landscape pattern is critical for the moderation of thermal extremes. Ecosphere Wash. DC 7. https://doi.org/10.1002/ecs2.1403 Chen, J., Saunders, S.C., Crow, T.R., Naiman, R.J., Brosofske, K.D., Mroz, G.D., Brookshire, B.L., Franklin, J.F., 1999. Microclimate in forest ecosystem and landscape ecology: Variations in local climate can be used to monitor and compare the effects of different management regimes. BioScience 49, 288–297. Collier, M.J., 2021. Are field boundary hedgerows the earliest example of a nature-based solution? Environ. Sci. Policy 120, 73–80. https://doi.org/10.1016/j.envsci.2021.02.008 Corbit, M., Marks, P.L., Gardescu, S., 1999. Hedgerows as habitat corridors for forest herbs in central New York, USA. J. Ecol. 87, 220–232. https://doi.org/10.1046/j.1365-2745.1999.00339.x De Frenne, P., Beugnon, R., Klinges, D., Lenoir, J., Niittynen, P., Pincebourde, S., Senior, R.A., Aalto, J., Chytrý, K., Gillingham, P.K., Greiser, C., Gril, E., Haesen, S., Kearney, M., Kopecký, M., le Roux, P.C., Luoto, M., Maclean, I., Man, M., Penczykowski, R., van den Brink, L., Van de Vondel, S., De Pauw, K., Lembrechts, J.J., Kemppinen, J., Van Meerbeek, K., 2024. Ten practical guidelines for microclimate research in terrestrial ecosystems. Methods Ecol. Evol. https://doi.org/10.1111/2041-210X.14476 De Frenne, P., Lenoir, J., Luoto, M., Scheffers, B.R., Zellweger, F., Aalto, J., Ashcroft, M.B., Christiansen, D.M., Decocq, G., Pauw, K.D., Govaert, S., Greiser, C., Gril, E., Hampe, A., Jucker, T., Klinges, D.H., Koelemeijer, I.A., Lembrechts, J.J., Marrec, R., Meeussen, C., Ogée, J., Tyystjärvi, V., Vangansbeke, P., Hylander, K., 2021. Forest microclimates and climate change: Importance, drivers and future research agenda. Glob. Change Biol. 1–19. https://doi.org/10.1111/gcb.15569 De Frenne, P., Zellweger, F., Rodríguez-Sánchez, F., Scheffers, B.R., Hylander, K., Luoto, M., Vellend, M., Verheyen, K., Lenoir, J., 2019. Global buffering of temperatures under forest canopies. Nat. Ecol. Evol. 3, 744–749. https://doi.org/10.1038/s41559-019-0842-1 Dobrowski, S.Z., 2011. A climatic basis for microrefugia: The influence of terrain on climate. Glob. Change Biol. 17, 1022–1035. https://doi.org/10.1111/j.1365-2486.2010.02263.x Drever, C.R., Cook-Patton, S.C., Akhter, F., Badiou, P.H., Chmura, G.L., Davidson, S.J., Desjardins, R.L., Dyk, A., Fargione, J.E., Fellows, M., Filewod, B., Hessing-Lewis, M., Jayasundara, S., Keeton, W.S., Kroeger, T., Lark, T.J., Le, E., Leavitt, S.M., LeClerc, M.-E., Lemprière, T.C., Metsaranta, J., McConkey, B., Neilson, E., St-Laurent, G.P., Puric-Mladenovic, D., Rodrigue, S., Soolanayakanahally, R.Y., Spawn, S.A., Strack, M., Smyth, C., Thevathasan, N., Voicu, M., Williams, C.A., Woodbury, P.B., Worth, D.E., Xu, Z., Yeo, S., Kurz, W.A., 2021. Natural climate solutions for Canada. Sci. Adv. 7, 1–14. https://doi.org/10.1126/sciadv.abd6034 Drexler, S., Thiessen, E., Don, A., 2024. Carbon storage in old hedgerows: The importance of below-ground biomass. GCB Bioenergy 16, e13112. https://doi.org/10.1111/gcbb.13112 Elmore, R.D., Carroll, J.M., Tanner, E.P., Hovick, T.J., Grisham, B.A., Fuhlendorf, S.D., Windels, S.K., 2017. Implications of the thermal environment for terrestrial wildlife management. Wildl. Soc. Bull. 41, 183–193. https://doi.org/10.1002/wsb.772 Fahrig, L., Girard, J., Duro, D., Pasher, J., Smith, A., Javorek, S., King, D., Lindsay, K.F., Mitchell, S., Tischendorf, L., 2015. Farmlands with smaller crop fields have higher within-field biodiversity. Agric. Ecosyst. Environ. 200, 219–234. https://doi.org/10.1016/j.agee.2014.11.018 Foley, J.A., Defries, R., Asner, G.P., Barford, C., Bonan, G., Carpenter, S.R., Chapin, F.S., Coe, M.T., Daily, G.C., Gibbs, H.K., Helkowski, J.H., Holloway, T., Howard, E.A., Kucharik, C.J., Monfreda, C., Patz, J.A., Prentice, I.C., Ramankutty, N., Snyder, P.K., 2005. Global consequences of land use 8, 570–574. https://doi.org/10.1126/science.1111772 Forman, R.T.T., Baudry, J., 1984. Hedgerows and hedgerow networks in landscape ecology. Environ. Manage. 8, 495–510. https://doi.org/10.1007/BF01871575 Fox, J., Weisberg, S., 2019. An R Companion to Applied Regression, Third. ed. Sage, Thousand Oaks CA. Geiger, R., Aron, R.H., Todhunter, P., 1995. The Climate Near the Ground, 5th ed. Vieweg+Teubner Verlag, Wiesbaden. Ghafarian, F., Ghazaryan, G., Wieland, R., Nendel, C., 2024. The impact of small woody features on the land surface temperature in an agricultural landscape. Agric. For. Meteorol. 349, 109949. https://doi.org/10.1016/j.agrformet.2024.109949 Gilbert, N.A., Anich, N.M., Worland, M., Zuckerberg, B., 2022. Microclimate complexities at the trailing edge of the boreal forest. For. Ecol. Manag. 524, 120533. https://doi.org/10.1016/j.foreco.2022.120533 Greiser, C., Meineri, E., Luoto, M., Ehrlén, J., Hylander, K., 2018. Monthly microclimate models in a managed boreal forest landscape. Agric. For. Meteorol. 250–251, 147–158. https://doi.org/10.1016/j.agrformet.2017.12.252 Gril, E., Spicher, F., Greiser, C., Ashcroft, M.B., Pincebourde, S., Durrieu, S., Nicolas, M., Richard, B., Decocq, G., Marrec, R., Lenoir, J., 2023. Slope and equilibrium: A parsimonious and flexible approach to model microclimate. Methods Ecol. Evol. 2022, 1–13. https://doi.org/10.1111/2041-210X.14048 Haddaway, N.R., Brown, C., Eales, J., Eggers, S., Josefsson, J., Kronvang, B., Randall, N.P., Uusi-Kämppä, J., 2018. The multifunctional roles of vegetated strips around and within agricultural fields. Environ. Evid. 7, 1–43. https://doi.org/10.1186/s13750-018-0126-2 Haesen, S., Lenoir, J., Gril, E., De Frenne, P., Lembrechts, J.J., Kopecký, M., Macek, M., Man, M., Wild, J., Van Meerbeek, K., 2023. Microclimate reveals the true thermal niche of forest plant species. Ecol. Lett. 26, 2043–2055. https://doi.org/10.1111/ele.14312 Helmuth, B., Broitman, B.R., Yamane, L., Gilman, S.E., Mach, K., Mislan, K.A.S., Denny, M.W., 2010. Organismal climatology: analyzing environmental variability at scales relevant to physiological stress. J. Exp. Biol. 213, 995–1003. https://doi.org/10.1242/jeb.038463 Hesslerová, P., Pokorný, J., Brom, J., Rejšková-Procházková, A., 2013. Daily dynamics of radiation surface temperature of different land cover types in a temperate cultural landscape: Consequences for the local climate. Ecol. Eng. 54, 145–154. https://doi.org/10.1016/j.ecoleng.2013.01.036 IPCC, 2023. Climate Change 2023: Synthesis Report. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Core Writing Team, H. Lee and J. Romero (eds.)]. IPCC, Geneva, Switzerland., in: Team, C.W., Lee, H., Romero, J. (Eds.),. Intergovernmental Panel on Climate Change (IPCC). Jucker, T., Jackson, T.D., Zellweger, F., Swinfield, T., Gregory, N., Williamson, J., Slade, E.M., Phillips, J.W., Bittencourt, P.R.L., Blonder, B., Boyle, M.J.W., Ellwood, M.D.F., Hemprich-Bennett, D., Lewis, O.T., Matula, R., Senior, R.A., Shenkin, A., Svátek, M., Coomes, D.A., 2020. A research agenda for microclimate ecology in human-modified tropical forests. Front. For. Glob. Change 2. https://doi.org/10.3389/ffgc.2019.00092 Kearney, M., Shine, R., Porter, W.P., 2009. The potential for behavioral thermoregulation to buffer “cold-blooded” animals against climate warming, PNAS March. Kemppinen, J., Lembrechts, J.J., Van Meerbeek, K., Carnicer, J., Chardon, N.I., Kardol, P., Lenoir, J., Liu, D., Maclean, I., Pergl, J., Saccone, P., Senior, R.A., Shen, T., Słowińska, S., Vandvik, V., von Oppen, J., Aalto, J., Ayalew, B., Bates, O., Bertelsmeier, C., Bertrand, R., Beugnon, R., Borderieux, J., Brůna, J., Buckley, L., Bujan, J., Casanova-Katny, A., Christiansen, D.M., Collart, F., De Lombaerde, E., De Pauw, K., Depauw, L., Di Musciano, M., Díaz Borrego, R., Díaz-Calafat, J., Ellis-Soto, D., Esteban, R., de Jong, G.F., Gallois, E., Garcia, M.B., Gillerot, L., Greiser, C., Gril, E., Haesen, S., Hampe, A., Hedwall, P.-O., Hes, G., Hespanhol, H., Hoffrén, R., Hylander, K., Jiménez-Alfaro, B., Jucker, T., Klinges, D., Kolstela, J., Kopecký, M., Kovács, B., Maeda, E.E., Máliš, F., Man, M., Mathiak, C., Meineri, E., Naujokaitis-Lewis, I., Nijs, I., Normand, S., Nuñez, M., Orczewska, A., Peña-Aguilera, P., Pincebourde, S., Plichta, R., Quick, S., Renault, D., Ricci, L., Rissanen, T., Segura-Hernández, L., Selvi, F., Serra-Diaz, J.M., Soifer, L., Spicher, F., Svenning, J.-C., Tamian, A., Thomaes, A., Thoonen, M., Trew, B., Van de Vondel, S., van den Brink, L., Vangansbeke, P., Verdonck, S., Vitkova, M., Vives-Ingla, M., von Schmalensee, L., Wang, R., Wild, J., Williamson, J., Zellweger, F., Zhou, X., Zuza, E.J., De Frenne, P., 2024. Microclimate, an important part of ecology and biogeography. Glob. Ecol. Biogeogr. 33, e13834. https://doi.org/10.1111/geb.13834 Keppel, G., Mokany, K., Wardell-Johnson, G.W., Phillips, B.L., Welbergen, J.A., Reside, A.E., 2015. The capacity of refugia for conservation planning under climate change. Front. Ecol. Environ. 13, 106–112. https://doi.org/10.1890/140055 Latimer, C.E., Zuckerberg, B., 2017. Forest fragmentation alters winter microclimates and microrefugia in human-modified landscapes. Ecography 40, 158–170. https://doi.org/10.1111/ecog.02551 LaZerte, S.E., Albers, S., 2018. weathercan: Download and format weather data from Environment and Climate Change Canada. J. Open Source Softw. 3, 571. Lee, X., Goulden, M.L., Hollinger, D.Y., Barr, A., Black, T.A., Bohrer, G., Bracho, R., Drake, B., Goldstein, A., Gu, L., Katul, G., Kolb, T., Law, B.E., Margolis, H., Meyers, T., Monson, R., Munger, W., Oren, R., U, K.T.P., Richardson, A.D., Schmid, H.P., Staebler, R., Wofsy, S., Zhao, L., 2011. Observed increase in local cooling effect of deforestation at higher latitudes. Nature 479, 384–387. https://doi.org/10.1038/nature10588 Lembrechts, J.J., Nijs, I., Lenoir, J., 2019. Incorporating microclimate into species distribution models. Ecography 42, 1267–1279. https://doi.org/10.1111/ecog.03947 Lenoir, J., Hattab, T., Pierre, G., 2017. Climatic microrefugia under anthropogenic climate change: Implications for species redistribution. Ecography 40, 253–266. https://doi.org/10.1111/ecog.02788 Maclean, I.M.D., Duffy, J.P., Haesen, S., Govaert, S., De Frenne, P., Vanneste, T., Lenoir, J., Lembrechts, J.J., Rhodes, M.W., Meerbeek, K.V., 2021. On the measurement of microclimate. Methods Ecol. Evol. 12, 1397–1410. https://doi.org/10.1111/2041-210X.13627 Mahmood, R., Pielke, R.A., Hubbard, K.G., Niyogi, D., Dirmeyer, P.A., Mcalpine, C., Carleton, A.M., Hale, R., Gameda, S., Beltrán-Przekurat, A., Baker, B., Mcnider, R., Legates, D.R., Shepherd, M., Du, J., Blanken, P.D., Frauenfeld, O.W., Nair, U.S., Fall, S., 2014. Land cover changes and their biogeophysical effects on climate. Int. J. Climatol. 34, 929–953. https://doi.org/10.1002/JOC.3736 Montgomery, I., Caruso, T., Reid, N., 2020. Hedgerows as ecosystems: Service delivery, management, and restoration. Annu. Rev. Ecol. Evol. Syst. 51, 81–102. https://doi.org/10.1146/annurev-ecolsys-012120-100346 Morelli, T.L., Barrows, C.W., Ramirez, A.R., Cartwright, J.M., Ackerly, D.D., Eaves, T.D., Ebersole, J.L., Krawchuk, M.A., Letcher, B.H., Mahalovich, M.F., Meigs, G.W., Michalak, J.L., Millar, C.I., Quiñones, R.M., Stralberg, D., Thorne, J.H., 2020. Climate-change refugia: biodiversity in the slow lane. Front. Ecol. Environ. 18, 228–234. https://doi.org/10.1002/fee.2189 Nakagawa, S., Schielzeth, H., 2013. A general and simple method for obtaining R 2 from generalized linear mixed-effects models. Methods Ecol. Evol. 4, 133–142. https://doi.org/10.1111/j.2041-210x.2012.00261.x Olsen, B.R.L., Fulbright, T.E., Hernández, F., Grahmann, E.D., Wester, D.B., Hehman, M.W., 2018. Ground surface vs. black globe temperature in northern bobwhite resource selection. Ecosphere Wash. DC 9. https://doi.org/10.1002/ecs2.2441 Peng, Y., Wang, Q., Bai, L., 2020. Identification of the key landscape metrics indicating regional temperature at different spatial scales and vegetation transpiration. Ecol. Indic. 111, 106066. https://doi.org/10.1016/j.ecolind.2020.106066 Petrides, G., 1972. A field guide to trees and shrubs: northeastern and north-central United States and southeastern and south-central Canada. Pincebourde, S., Sinoquet, H., Combes, D., Casas, J., 2007. Regional climate modulates the canopy mosaic of favourable and risky microclimates for insects. J. Anim. Ecol. 76, 424–438. https://doi.org/10.1111/j.1365-2656.2007.01231.x Pinheiro, J., Bates, D., R Core Team, 2023. nlme: Linear and Nonlinear Mixed Effects Models. Potter, K.A., Arthur Woods, H., Pincebourde, S., 2013. Microclimatic challenges in global change biology. Glob. Change Biol. 19, 2932–2939. https://doi.org/10.1111/gcb.12257 Poveda, K., Martínez, E., Kersch-Becker, M.F., Bonilla, M.A., Tscharntke, T., 2012. Landscape simplification and altitude affect biodiversity, herbivory and Andean potato yield. J. Appl. Ecol. 49, 513–522. https://doi.org/10.1111/j.1365-2664.2012.02120.x Précigout, P.A., Robert, C., 2022. Effects of hedgerows on the preservation of spontaneous biodiversity and the promotion of biotic regulation services in agriculture: Towards a more constructive relationships between agriculture and biodiversity. Bot. Lett. 169, 176–204. https://doi.org/10.1080/23818107.2022.2053205 R Core Team, 2024. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Rakowski, A.E., Elmore, R.D., Davis, C.A., Fuhlendorf, S.D., Carroll, J.M., 2019. Thermal refuge affects space use and movement of a large-bodied galliform. J. Therm. Biol. 80, 37–44. https://doi.org/10.1016/j.jtherbio.2018.12.024 Ramankutty, N., Evan, A.T., Monfreda, C., Foley, J.A., 2008. Farming the planet: 1. Geographic distribution of global agricultural lands in the year 2000. Glob. Biogeochem. Cycles 22. https://doi.org/10.1029/2007GB002952 Riddell, E.A., Iknayan, K.J., Hargrove, L., Tremor, S., Patton, J.L., Ramirez, R., Wolf, B.O., Beissinger, S.R., 2021. Exposure to climate change drives stability or collapse of desert mammal and bird communities. Science 371, 633–638. https://doi.org/10.1126/SCIENCE.ABD4605 Riva, F., Fahrig, L., 2022. The disproportionately high value of small patches for biodiversity conservation. Conserv. Lett. 15, e12881. https://doi.org/10.1111/conl.12881 Sánchez, I.A., Lassaletta, L., McCollin, D., Bunce, R.G.H., 2009. The effect of hedgerow loss on microclimate in the Mediterranean region: An investigation in Central Spain. Agrofor. Syst. 78, 13–25. https://doi.org/10.1007/s10457-009-9224-z Scheffers, B.R., Edwards, D.P., Diesmos, A., Williams, S.E., Evans, T.A., 2014. Microhabitats reduce animal’s exposure to climate extremes. Glob. Change Biol. 20, 495–503. https://doi.org/10.1111/gcb.12439 Suggitt, A.J., Wilson, R.J., August, T.A., Fox, R., Isaac, N.J.B., Macgregor, N.A., Morecroft, M.D., Maclean, I.M.D., 2015. Microclimate affects landscape level persistence in the British Lepidoptera. J. Insect Conserv. 19, 237–253. https://doi.org/10.1007/s10841-014-9749-y Terando, A., Youngsteadt, E., Meineke, E., Prado, S., 2018. Accurate near surface air temperature measurements are necessary to gauge large-scale ecological responses to global climate change. Ecol. Evol. 3–4. https://doi.org/10.1002/ece3.3972 Terando, A.J., Youngsteadt, E., Meineke, E.K., Prado, S.G., 2017. Ad hoc instrumentation methods in ecological studies produce highly biased temperature measurements. Ecol. Evol. 7, 9890–9904. https://doi.org/10.1002/ece3.3499 Tilman, D., Fargione, J., Wolff, B., Antonio, C.D., Dobson, A., Howarth, R., Schindler, D., Schlesinger, W.H., Simberloff, D., Swackhamer, D., 2001. Forecasting agriculturally driven environmental change. Am. Assoc. Adv. Sci. 292, 281–284. https://doi.org/10.1126/science.1057544 Vanneste, T., Govaert, S., Spicher, F., Brunet, J., Cousins, S.A.O., Decocq, G., Diekmann, M., Graae, B.J., Hedwall, P.O., Kapás, R.E., Lenoir, J., Liira, J., Lindmo, S., Litza, K., Naaf, T., Orczewska, A., Plue, J., Wulf, M., Verheyen, K., De Frenne, P., 2020. Contrasting microclimates among hedgerows and woodlands across temperate Europe. Agric. For. Meteorol. 281, 107818. https://doi.org/10.1016/j.agrformet.2019.107818 Vitt, L.J., Sartorius, S.S., 1999. HOBOs, tidbits and lizard models: The utility of electronic devices in field studies of ectotherm thermoregulation. Funct. Ecol. 13, 670–674. Wei, L., Sanczuk, P., De Pauw, K., Caron, M.M., Selvi, F., Hedwall, P.-O., Brunet, J., Cousins, S.A.O., Plue, J., Spicher, F., Gasperini, C., Iacopetti, G., Orczewska, A., Uria-Diez, J., Lenoir, J., Vangansbeke, P., De Frenne, P., 2024. Using warming tolerances to predict understory plant responses to climate change. Glob. Change Biol. 30, e17064. https://doi.org/10.1111/gcb.17064 Zellweger, F., De Frenne, P., Lenoir, J., Vangansbeke, P., Verheyen, K., Bernhardt-Römermann, M., Baeten, L., Hédl, R., Berki, I., Brunet, J., Calster, H.V., Chudomelová, M., Decocq, G., Dirnböck, T., Durak, T., Heinken, T., Jaroszewicz, B., Kopecký, M., Máliš, F., Macek, M., Malicki, M., Naaf, T., Nagel, T.A., Ortmann-Ajkai, A., Petřík, P., Pielech, R., Reczyńska, K., Schmidt, W., Standovár, T., Świerkosz, K., Teleki, B., Vild, O., Wulf, M., Coomes, D., 2020. Forest microclimate dynamics drive plant responses to warming. Science 368, 772–775. https://doi.org/10.1126/science.aba6880 Zellweger, F., Sulmoni, E., Malle, J.T., Baltensweiler, A., Jonas, T., Zimmermann, N.E., Ginzler, C., Karger, D.N., Frenne, P.D., Frey, D., Webster, C., 2024. Microclimate mapping using novel radiative transfer modelling. Biogeosciences 21, 605–623. https://doi.org/10.5194/bg-21-605-2024 Zimmermann, B., Kruber, S., Nendel, C., Munack, H., Hildmann, C., 2024. Assessing the cooling potential of climate change adaptation measures in rural areas. J. Environ. Manage. 366, 121595. https://doi.org/10.1016/j.jenvman.2024.121595 Zuur, A.F., Ieno, E.N., Walker, N., Saveliev, A. a, Smith, G.M., 2009. Mixed Effects Models and Extension in Ecology with R, 1st ed, Statistics for Biology and Health. Springer, New York, NY. https://doi.org/10.1007/978-0-387-87458-6 Crossref Google Scholar Information & Authors Information Version history V1 Version 1 07 October 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords description ecosystem ecology ecosystem function statistical terrestrial Authors Affiliations Sarah Endicott 0000-0001-9644-5343 Environment and Climate Change Canada National Wildlife Research Centre View all articles by this author Adam Blasl WSP Canada Inc View all articles by this author Marie-Hélène Brice Jardin botanique de Montréal View all articles by this author Ilona Naujokaitis-Lewis 0000-0001-9504-4484 [email protected] Environment and Climate Change Canada National Wildlife Research Centre View all articles by this author Metrics & Citations Metrics Article Usage 241 views 206 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Sarah Endicott, Adam Blasl, Marie-Hélène Brice, et al. The role of field margins in microclimatic temperature buffering of temperate Canadian landscapes. Authorea . 07 October 2025. DOI: https://doi.org/10.22541/au.175986071.14914813/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.175986071.14914813/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9ffc6d565a6bc13d',t:'MTc3OTQ1OTI2NQ=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();