Deploying multispectral remote sensing for multi‐temporal analysis of archaeological crop stress at Ravenshall, Fife, Scotland

Abstract

Diminishing returns of archaeological crop marks in lowland areas from traditional observer-directed visible spectrum aerial survey with standard photographic cameras highlights a need to explore alternative approaches to maintain the effectiveness of survey programmes. Developments in low-cost multispectral remote sensing have in part been driven by the growth of precision agriculture and, whilst contributing to the intensification of land use, these technologies may offer new spectral and temporal capacities for detecting, recording and monitoring historic landscapes. However, there are significant challenges to the deployment of such approaches, not least the costs of data acquisition and uncertainty about the best conditions for data collection. This study assesses the effectiveness of the Parrot Sequoia, a relatively low-cost multispectral sensor recently developed for agricultural applications, for the detection of crop marks to inform archaeological survey. A series of observations were taken with the sensor mounted on an unmanned aerial vehicle (UAV) at Ravenshall, Fife, Scotland, between April and July 2017. The resulting reflectance maps are compared to red, green and blue (RGB) photographs taken with a Nikon D800E digital camera during seven light aircraft surveys, with the aim of testing the sensors' comparative ability to record crop mark developments over time. The contrast in reflectance between vegetation samples growing over buried archaeological remains and the surrounding field was assessed through separability in regional histogram values across different image band combinations. Separable values indicative of crop marks were found in both the multispectral and RGB results from June 2017 onwards. Several vegetation index (VI) maps, particularly the Simple Ratio (SR) and Normalised Difference Vegetation Index (NDVI), were found to be effective for distinguishing crop marks in the multispectral results. The Sequoia is a cost-effective sensor offering improved spectral resolution over the RGB photographs, showing potential for subtle crop mark detection across compact study areas.

Multitemporal 2016-2018 Sentinel-2 Data Enhancement for Landscape Archaeology: The Case Study of the Foggia Province, Southern Italy

“…Normalized Difference Water Index (NDWI) [71][72][73][74][75][76] These spectral combinations are useful for the enhancement of crop-marks, soil-marks, and anomalies generated by the presence of buried structures of potential archaeological interest [60][61][62][76][77][78]. The NDVI (Normalised Difference Vegetation Index) is one of the most widely used indices also in archaeological applications.…”

“…NDVI is sensitive to the absorption of radiation in the Red band, caused by chlorophyll, and the reflection of light in the Near Infrared [49,50,64,76]. Recently, several other indices have been also used for the identification of archaeological features, allowing the discrimination of pixels according to their spectral signature [76,77]. For example: (i) Albedo is an indicator of some surface features, such as brightness [50]; (ii) SR is useful for bare soil analysis; (iii) Green NDVI is a variant of the NDVI index but more sensitive to chlorophyll concentration [55,56]; (iv) SAVI and EVI indices, calculated using the blue, red, and near infrared bands, are useful to reduce background influence [49,50]; (v) GEMI index reduces the influence of the atmosphere [79,80]; (vi) SWIR (NDWI, NVMI) band-based indices are useful to assess the moisture content of soil and vegetation [81,82]; (vii) NAI and RN indices, as well as crop-and soil-indices, use wavelengths from 0.6 to 0.9 to enhance archaeological proxy indicators [82,83].…”

Monitoring Plant Status and Fertilization Strategy through Multispectral Images

“…Table 2. Vegetation Indices (VIs) used to create spectral profiles of the tomato plants from the multispectral data with formulae and traditional applications (obtained from [36]).…”

“…Note: For use in this study, some closest Sequoia reflectance bands were substituted by the traditional narrowband wavelengths as described by [36] and showed in the Table 2.…”

Mapping Heterogeneous Buried Archaeological Features Using Multisensor Data from Unmanned Aerial Vehicles

“…Multispectral and thermal imaging are comparatively recent developments in the archaeological survey toolkit and the availability of UAVs has further enabled the capture of very high spatial resolution imagery-vital for the detection of subtle, man-made features such as buried walls and ditches [24][25][26][27].…”