ThermStats: An R package for quantifying surface thermal heterogeneity in assessments of microclimates

Abstract

Variation in temperature at a fine spatial scale creates critically important microclimates for many organisms. Quantifying thermal heterogeneity at this scale is challenging and, until recently, has been largely restricted to the use of dataloggers to record air temperature. Thermography is becoming an increasingly viable alternative. A single photo from a thermal imaging camera contains thousands of spatially explicit surface temperature measurements, making thermal cameras ideal for rapidly assessing temperature variation at fine scale. Here, we present an R package—ThermStats—for processing thermal images and other gridded temperature data. The package addresses current constraints on applying thermography in ecology, by speeding up and simplifying the extraction of data from thermal images, and by facilitating the calculation of different metrics of thermal heterogeneity. The metrics capture both the frequency distribution and spatial patterns of temperature, and the package functions are designed to accommodate different sampling strategies and data in either matrix or raster format. We demonstrate how ThermStats can be used to capture temperature variation at fine spatial scales in structurally complex habitat, such as tropical rainforest. Using thermal images collected in the field (~0.5 cm2 resolution), we found that thermal hetereogeneity varied little between primary and logged forest, but did vary with time of day. Comparing temperature extremes in a microclimate layer estimated from LIght Detection And Ranging data (LIDAR; 2,500 m2 resolution) , we found that both hot and cold extremes were hotter inside oil palm plantations than in the neighbouring forest. Our package simplifies the processing of thermal data, and our metrics capture key spatiotemporal temperature trends that underpin physiological, behavioural and demographic responses to environmental change. As such, ThermStats can advance a wide range of studies requiring fine‐scale surface temperature data for microclimate investigations.

A Research Agenda for Microclimate Ecology in Human-Modified Tropical Forests

“…Mapping horizontal, vertical, and temporal variation in microclimate at landscape and regional scales (e.g., Jucker et al, 2018) would help identify dispersal corridors that are climatically suitable for a range of taxonomic groups (Milanesi et al, 2017). This would allow microclimate ecology to be integrated into classical landscape ecology and metacommunity dynamics (Hesselbarth et al, 2019;Senior et al, 2019), and also guide efforts to restore and increase connectivity in fragmented landscapes (e.g., through the rehabilitation of riparian buffer zones; Luke et al, 2019). It also provides an opportunity to assess the effectiveness of different forest and plantation management practices that aim to restore microclimate regimes by altering canopy structure and ecophysiology, such as natural regeneration, enrichment planting, and climber cutting (Rodríguez-Ronderos et al, 2016;Ichihashi et al, 2017;Guzmán et al, 2018).…”

“…(C) Captures variation in near-surface mean daily temperature across a 1-ha logged forest plot within the SAFE landscape generated by interpolating data from a grid of 72 microclimate data loggers (Blonder et al, 2018). (D) Is a thermal image obtained using a FLIR model E40 camera in a selectively logged forest in Sabah (Senior et al, 2019). tropical forests have been attributed to changes in microclimate . However, due to a lack of data in most cases this link between microclimate and ecosystem change has been assumed rather than directly observed.…”

Water-Stressed Plants Do Not Cool: Leaf Surface Temperature of Living Wall Plants under Drought Stress

“…Because the position of the plants changed slightly, mainly due to growth, the polygons were adjusted daily. We extracted the radiometric values (DN) with the software RStudio (Version 1.1.463 R Studio, Boston, Massachusetts, USA) from each pixel within the polygons and calculated the median and the 97.5 percentile [41]. Because the differences between the treatments were clearer with the 97.5 percentile, we subsequently used these data.…”

Challenges and opportunities of species distribution modelling of terrestrial arthropod predators

“…CHELSA (Karger et al, 2017) and WorldClim 2 Mammola et al EcoEvoRxiv (Fick and Hijmans, 2017)] achieve a maximum resolution of 30 arcsec (cell size c. 1 km 2 at the equator), which is excellent but might not be enough in the case of invertebrates that are known to respond to microclimatic characteristics over spatial scales of millimetres to metres (Potter et al 2013, Suggitt et al 2018. On a positive note, gigantic leaps forward are being made in the development of microclimatic databases (e.g., Kearney et al, 2014), as well as approaches for downscaling temperature data at high resolutions from thermal images (Senior et al, 2019) or airborne light detection and ranging data (George et al 2015). It is predicted that in the following years, the use of remote sensing derived data will become the standard for modelling and mapping the microclimate (Zellweger et al 2019), especially in invertebrate research where the use of similar high-resolution data has already proved useful to achieve realistic conservation prioritization (e.g.…”