Geoderma Regional 34 (2023) e00671 Available online 14 June 2023 2352-0094/© 2023 Elsevier B.V. All rights reserved. Development of land-use-specific pedotransfer functions for predicting bulk density of acidic topsoil in eastern Himalayas (India) Burhan U. Choudhury a, * , Priyabatra Santra b , Naseeb Singh a , Poulamee Chakraborty c a ICAR Research Complex for NEH Region, Umiam, Meghalaya 793103, India b ICAR-Central Arid Zone Research Institute, Jodhpur, Rajasthan 342003, India c Department of Plant Science, South Dakota State University, Brookings, SD, USA A R T I C L E INFO Keywords: Pedotransfer functions Machine learning algorithms Soil bulk density Inceptisols Land use Eastern Himalayan India ABSTRACT Soil bulk density (BD), although a key input parameter in many soil functions is often overlooked in soil studies, including surveys. Direct measurements over a large area are slow and expensive, especially in resource-poor mountain ecosystems. For soils of eastern Himalayan ecosystems, cost-effective indirect methods such as pedotransfer functions (PTFs) are lacking. We developed seven landuse-specific PTFs for BD estimation using a soil particle size distribution (sand, and silt) and soil organic carbon database of 1206 sampling sites. Land uses were 3 uncultivated (dense forests-DF, open forests-OF, and grasslands-GL), 3 cultivated (shifting cultivation-SC, upland agriculture -UA, lowland paddy-LP), and a perennial plantation (PL). PTFs were developed using five machine learning algorithms (MLAs) and a multi-linear regression (MLR). The best PTFs were chosen based on cross-validation performance with the values of the greatest coefficient of determination (R 2 ) and the smallest root mean square error (RMSE). In addition, the relative performance of thirteen (13) documented PTFs was evaluated on soils in the region. Among the MLAs, the artificial neural network (ANN) showed the best BD prediction for the whole dataset (R 2 : 0.58, RMSE: 0.09 Mg m  3 ) and four (e.g. GL, UA, SC, and LP: R 2 of 0.65 to 0.69, RMSE of 0.06 to 0. 09 Mg m  3 ) of the seven land use classes. For the remaining three land uses (e.g., DF, OF and PL), based on the greatest R 2 values (0.55 to 0.75), PTFs based on random forests (RF) made the best predictions. However, considering the smallest RMSE values (0.06–0.08 Mg m  3 ), ANN-PTFs were another op- tion for these three land uses. Of the land uses, the best prediction for BD was obtained for PL (R 2 : 0.75), while the DF with variable soil properties had the lowest predictive performance (R 2 : 0.55). The performance of MLA- based PTFs such as the support vector machine (SVM), ridge regression (RR), and extreme gradient boosting (XGB) and MLR-PTFs were lower than the ANN-PTFs and RF-PTFs. The precision of BD prediction for 13 pub- lished PTFs was inconsistent and inferior (R 2 : 0.13 to 0.39) to the ANN-PTFs and RF-PTFs (R 2 : 0.55 to 0.75) developed in this study. Since PTFs are very region-specific, we suggest ANN-PTFs to estimate the BD of grasslands and cultivated soils while the RF-PTFs (based on the greatest R 2 ) for dense and open forests and plantation soils in the eastern Himalayas and similar mountain ecosystems. 1. Introduction In recent decades, studies on soil carbon (C) inventories have garnered worldwide attention and many of them illustrated C’s primary role as a sink in the global carbon cycle (Regnier et al., 2022). Therefore, successful implementation of a C-sequestration program could offset the adverse impacts of global warming on vegetation and land productivity. In C-sequestration studies, soil bulk density (BD) is a key parameter for carbon weight-volume conversion while estimating C-stocks and fluxes (Martin et al., 2009). Furthermore, BD is also an important parameter for many soil functions and is well documented (Katuwal et al., 2020; Obidike-Ugwu et al., 2022; Palladino et al., 2022). However, in most soil fertility studies and surveys, BD is not measured often, unlike other soil properties. Measuring soil BD is not complex, but expensive and labo- rious (Richard et al., 2022). The direct methods commonly used to determine the soil BD in the field are core, ball and excavation, while radiation, an indirect method, is also used (Al-Shammary et al., 2018). In the core method, the accuracy of the measurements depends on core * Corresponding author at: Principal Scientist, Soil Physics and Soil water conservation, Division of System Research and Engineering, ICAR Research Complex for NEH Region, Umiam, Meghalaya 793 103, India. E-mail addresses: burhan3i@yahoo.com, burhan.icar@gmail.com (B.U. Choudhury). Contents lists available at ScienceDirect Geoderma Regional journal homepage: www.elsevier.com/locate/geodrs https://doi.org/10.1016/j.geodrs.2023.e00671 Received 14 January 2023; Received in revised form 8 June 2023; Accepted 12 June 2023