Quantifying soil carbon in temperate peatlands using a mid-IR soil spectral library

Abstract

Abstract. Traditional laboratory methods for acquiring soil information remain important for assessing key soil properties, soil functions and ecosystem services over space and time. Infrared spectroscopic modeling can link and massively scale up these methods for many soil characteristics in a cost-effective and timely manner. In Switzerland, only 10 % to 15 % of agricultural soils have been mapped sufficiently to serve spatial decision support systems, presenting an urgent need for rapid quantitative soil characterization. The current Swiss soil spectral library (SSL; n = 4374) in the mid-infrared range includes soil samples from the Biodiversity Monitoring Program (BDM), arranged in a regularly spaced grid across Switzerland, and temporally resolved data from the Swiss Soil Monitoring Network (NABO). Given that less than 2 % of the samples in the SSL originate from organic soils, we aimed to develop both an efficient calibration sampling scheme and accurate modeling strategy to estimate the soil carbon (SC) contents of heterogeneous samples between 0 and 2 m depth from 26 locations within two drained peatland regions (School of Agricultural, Forest and Food Sciences (HAFL) data set; n = 116). The focus was on minimizing the need for new reference analyses by efficiently mining the spectral information of the SSL. We used partial least square regressions (PLSRs), together with five repetitions of a location-grouped, 10-fold cross-validation, to predict SC ranging from 1 % to 52 % in the local HAFL data set. We compared the validation performance of different calibration schemes involving local models (1), models using the entire SSL combined with local samples (2), commonly referred to as spiking, and subsets of local and SSL samples optimized for the peatland target sites using the resampling local (RS-LOCAL) algorithm (3). Using local and RS-LOCAL calibrations with at least five local samples, we achieved similar validation results for predictions of SC up to 52 % (R2 = 0.93 to 0.97; bias = -0.07 to 1.65; root mean square error (RMSE) = 2.71 % to 3.89 % total carbon; ratio of performance to deviation (RPD) = 3.38 to 4.86; and ratio of performance to interquartile range (RPIQ) = 4.93 to 7.09). However, calibrations using RS-LOCAL only required five or 10 local samples for very accurate models (RMSE = 3.16 % and 2.71 % total carbon, respectively), while purely local calibrations required 50 samples for similarly accurate results (RMSE < 3 % total carbon). Of the three approaches, the entire SSL spiked with local samples for model calibration led to validations with the lowest performance in terms of R2, bias, RMSE, RPD and RPIQ. Hence, we show that a simple and comprehensible modeling approach, using RS-LOCAL together with a SSL, is an efficient and accurate strategy when using infrared spectroscopy. It decreases field and laboratory work, the bias of SSL spiking approaches and the uncertainty of local models. If adequately mined, the information in the SSL is sufficient to predict SC in new and independent study regions, even if the local soil characteristics are very different from the ones in the SSL. This will help to efficiently scale up the acquisition of quantitative soil information over space and time.

Developing the Swiss mid-infrared soil spectral library for local estimation and monitoring

“…Although the range of total C measured was large (14-520 g kg −1 C) and the soils were diverse, as few as 5 or 10 site-specific tuning samples were sufficient to estimate the validation samples with reasonable accuracy (RMSE = < 30 g kg −1 C; RPIQ > 3.4); this was com-parable to a local-only calibration with 50 samples. Helfenstein et al (2021) found considerably lower conditional prediction errors (< 10 g kg −1 ) when considering measurements of < 100 g kg −1 ; this suggests that increasing the amount and compositional complexity of organic soils in the library has potential for more accurately characterizing diverse soil ecoregions with soils that have high organic matter contents.…”

“…Because organic soils can have up to 500 g kg −1 OC, and because more than 98 % of the samples are mineral soils, organic soils are underrepresented in the current Swiss SSL. For this reason, Helfenstein et al (2021) evaluated the present Swiss SSL for a regional transfer based on new organo-mineral soils from two peatland regions in Switzerland. Although the range of total C measured was large (14-520 g kg −1 C) and the soils were diverse, as few as 5 or 10 site-specific tuning samples were sufficient to estimate the validation samples with reasonable accuracy (RMSE = < 30 g kg −1 C; RPIQ > 3.4); this was com-parable to a local-only calibration with 50 samples.…”

“…Other studies calibrated separate models on partitions of training data that were derived from applying certain criteria (i.e., geographical region, terrain attributes, parent material, soil type, land use and spectra-based clustering; Sila et al, 2016;Ogen et al, 2019). Still others used memory-based learning based on spectral similarity, extracting useful information from compositional relatedness of soils (Ramirez-Lopez et al, 2013;Clairotte et al, 2016;Hong et al, 2019;Dangal et al, 2019) or addition-ally geographic proximity (Tziolas et al, 2019). These all produce individual models for each sample to be predicted.…”

BIS-4D: Mapping soil properties and their uncertainties at 25 m resolution in the Netherlands

“…Despite the many advantages of using QRF for DSM (Sect 3.2), predictions may be further improved using methods such as convolutional or recursive neural networks (deep learning; Behrens et al, 2005Behrens et al, , 2018aPadarian et al, 2019b;Wadoux, 2019;Wadoux et al, 2019) or transfer learning (Liu et al, 2018;Padarian et al, 2019a;Seidel et al, 2019;Helfenstein et al, 2021;Baumann et al, 2021), defined as the process of sharing intra-domain information and rules learned by general models to a local domain (Pan and Yang, 2010). We recommend future research to investigate the use of deep learning and transfer learning in the Netherlands for SOM, due to the large amount of SOM data and more opportunities in accounting for differences in observational quality (field estimates and laboratory measurements) using more complex models.…”

BIS-4D: mapping soil properties and their uncertainties at 25 m resolution in the Netherlands

“…Despite the many advantages of using QRF for DSM (Sect. 4.2), predictions may be further improved using methods such as convolutional or recursive neural networks (deep learning; Behrens et al, 2005Behrens et al, , 2018aPadarian et al, 2019b;Wadoux, 2019;Wadoux et al, 2019) or transfer learning (Liu et al, 2018;Padarian et al, 2019a;Seidel et al, 2019;Helfenstein et al, 2021;Baumann et al, 2021), defined as the process of sharing intra-domain information and rules learned by general models to a local domain (Pan and Yang, 2010). We recommend future research to investigate the use of deep learning and transfer learning in the Netherlands for SOM due to the large amount of SOM data and more opportunities in accounting for differences in observational quality (field estimates and laboratory measurements) using more complex models.…”