Geostatistical modeling and soil fertility assessment for sugarcane crop in tropical region of Maharashtra

Introduction

Sugarcane (Saccharum officinarum L.) is one of India's most important commercial crops, forming the backbone of the rural economy and supporting millions of farming families. In Maharashtra and other major sugarcane-growing states, the crop occupies a central place in agricultural planning; however, the average productivity has not shown significant improvement over the past decade. This stagnation is closely linked to the progressive deterioration of soil fertility, which remains a key determinant of crop performance. Multiple interacting factorssuch as erratic climatic conditions, continuous monocropping, over-exploitation of soils, indiscriminate fertilizer use, declining soil organic carbon, excessive irrigation, poor-quality water, and inadequate drainage have collectively contributed to soil health degradation across sugarcane landscapes. Spatial variability is an inherent and dynamic feature of soil. It may be both vertical (within a pedon) and horizontal (across the landscape). Soil variability is the outcome of many processes acting and interacting across a continuum of spatial and temporal scales and is inherently scale dependent. Variability may also occur as a result of land use and management strategies. As other environmental variables, soil water and soil texture changes in space and time. For farmers and sugar industries alike, enhancing sugarcane yield is essential for improving economic returns. Yet, considerable field-to-field variation in soil fertility often restricts the potential to achieve uniform productivity. This spatial heterogeneity of soils underscores the need for sitespecific nutrient management (SSNM), a core principle of precision agriculture that ensures balanced and efficient nutrient application suited to local soil conditions (Sharma et al., 2020) [33] . Sugar mills generally operate over large command areas typically 5,000 to 7,000 hectares and although many have established soil testing laboratories, conventional soil sampling and analysis remain time-consuming, labour-intensive, and financially demanding. Despite the distribution of thousands of soil health cards each year, a large proportion of farmers still adopt blanket fertilizer doses, often resulting in nutrient imbalance, unnecessary input costs, and suboptimal yields. In this context, the ability to predict and map the spatial variability of soil fertility parameters at a finer scale becomes critically important. Geospatial technologies and geostatistical modelling offer efficient tools for generating continuous surface information on soil properties using limited point-based data. Studies across diverse agroecosystems have demonstrated that spatial variability maps derived using these techniques are invaluable for diagnosing nutrient constraints, delineating management zones, and formulating soil test-based fertilizer prescriptions (Khan et al., 2020 [14,31,29,34,7] . Such maps also support longterm soil fertility monitoring and enable farmers to adopt nutrient-smart practices tailored to the inherent variability within their fields (Prasad et al., 2020) [27] . Given that systematic assessments of spatial nutrient variability are limited for sugar mill command areas in semi-arid Maharashtra, there is a pressing need to document soil fertility patterns and understand their geostatistical behaviour for precision nutrient advisory development. Therefore, the present study was undertaken in the operational area of Samarth SSK, Ambad Taluka, Jalna District, Maharashtra, with the objectives to: (i) assess the spatial variability of key soil fertility parameters, (ii) identify suitable geostatistical models for predicting nutrient distribution, and (iii) generate highresolution, cadastral-level nutrient management advisories to support sustainable sugarcane production. This investigation provides a scientific basis for transitioning from blanket fertilizer recommendations to field-specific nutrient management within sugar mill zones, ultimately contributing to improved soil health and enhanced sugarcane productivity.

Materials and Methods

Study Area

The soil fertility survey was conducted in the operational area of Samarth SSK Sugar Mill, located in Ambad Taluka, Jalna District, Maharashtra, with the aim of generating detailed soil fertility maps and facilitating site-specific nutrient management. The study area lies geographically between 19.61°N latitude and 75.79°E longitude, covering a typical sugarcane-growing region of Marathwada. The region experiences a tropical climate, characterized by distinct wet and dry seasons. According to the Köppen-Geiger climate classification, Ambad Taluka falls under the Aw category, indicating a tropical savanna climate where winter months receive significantly lower rainfall compared to the summer monsoon season. The mean annual temperature of the region is approximately 26.1°C, providing a warm environment conducive for sugarcane cultivation. Annual precipitation averages around 824 mm, with most of the rainfall occurring during the southwest monsoon months (June-September). The length of growing period in the area is 152 days. The soils have a Typic Tropoustic moisture regime. The soil temperature regime is hyperthermic. The soils of the area are Vertisols and associated blck soils typically developed on alluvium derived from weathering of Deccan Basalt, which influences the chemical and physical properties, including pH, nutrient availability, and cation exchange capacity. Majority percentage of cultivated land in kharif is under cotton (Gossypium spp.), Soybean (Glycin max), pigeon pea (Cajanous cajan), greengram (Phaseolus aurens) as principal crops. Chickpea (Cicer arietinum) is dominant crop in rabi season under residual soil moisture and/or with some protective irrigations. The natural vegetation of the area comprises of dry deciduous tree species and grasses. The dominant tree species are babul (Accacia arabica), ber (Ziziphus jujube), palas (Butea frondosa), neem (Azadiracta indica), rui (Calotropic giganda), kans (Saccharum spontaneum) and dub (Cynadon dactylon). This combination of climatic conditions, soil parent material, and cropping patterns makes Ambad Taluka a representative site for studying spatial variability in soil fertility and assessing the potential for precision nutrient management in sugarcane systems. Understanding these characteristics is crucial for optimizing fertilizer use, improving crop productivity, and sustaining soil health under intensive sugarcane cultivation.

Soil Sampling and Analysis

Soil samples were collected systematically from the operational area of Samarth SSK Sugar Mill using a survey-number-wise approach, targeting the 0 -22 cm soil depth, which represents the plough layer and is most relevant for nutrient availability to sugarcane. A total of 3,500 composite soil samples were collected, with each composite representing the overall fertility status of the corresponding plot. This approach ensures that spatial variability within individual fields is adequately captured while maintaining efficiency in sample processing. Collected samples were processed following standard laboratory procedures. Soil pH, a critical parameter influencing nutrient availability and chemical reactions, was determined in a 1:2.5 soil-to-water suspension, while electrical conductivity (EC), an indicator of soil salinity, was measured in the soil supernatant (Jackson, 1993) [11] . Soil organic carbon (OC), which plays a key role in soil fertility, structure, and nutrient retention, was analyzed using the wet digestion method (Walkley and Black, 1934). Available nitrogen (N) was estimated by the alkaline potassium permanganate method (Subbaiah and Asija, 1956) [40] , available phosphorus (P) by the Olsen bicarbonate method (Olsen et al., 1954) [23] , and available potassium (K) by the neutral normal ammonium acetate extraction method (Jackson, 1973) [11] . The measured soil test values were then interpreted to derive fertilizer recommendations based on the soil fertility rating and guidelines provided by the State Agriculture University (Table 1). This approach enables site-specific nutrient management, facilitating optimal fertilizer application tailored to the specific soil conditions of each field, and ultimately contributing to improved sugarcane productivity and sustainable soil management.

Geostatistical Analysis of Soil Parameters

To assess the spatial variability of soil fertility parameters across Ambad Taluka, thematic maps of soil chemical properties were developed using Geographic Information System (GIS) techniques in ArcGIS version 10.1 (ESRI, Redlands, USA). Digital shapefiles of Ambad Taluka, obtained from the Maharashtra Remote Sensing Application Centre (MRSAC), Nagpur, served as the base maps. These shapefiles were derived from IRS P6 LISS IV satellite imagery with a high spatial resolution of 5.8 meters, ensuring precise delineation of cadastral boundaries. Additionally, ancillary data sources, including Google Earth imagery, were integrated to enhance mapping accuracy and spatial reference. Survey-number-wise soil analytical data were overlaid on the base maps to create a point map, representing the exact locations of collected soil samples. Using this georeferenced dataset, surface thematic maps for key soil properties including soil pH, electrical conductivity (EC), organic carbon (OC), and macronutrients (N, P, K) were generated.

Semivariogram Modeling

Ordinary Kriging, a widely used geostatistical interpolation technique, was employed to predict soil property values at unsampled locations. Kriging leverages the semivariogram model to account for spatial dependence, assigning weights to known sample points based on their spatial correlation to provide accurate estimates across the landscape. This approach enabled the generation of continuous, high-resolution soil fertility maps, facilitating site-specific nutrient management in sugarcane cultivation. The geostatistical analysis provides valuable insights into spatial heterogeneity, supports precision fertilization, and helps optimize overall crop management strategies. This study employed ordinary kriging, a linear geostatistical interpolation technique, to estimate soil property values at unsampled locations by computing a weighted average of nearby observed values (Krige, 1951) [15] . The kriging estimate at an unknown location Z(k) is given by: where Zi_represents known values at sampled points, and yi are the weights assigned to each sample, determined based on spatial correlation patterns. In geostatistics, the semivariogram quantifies spatial dependence, measuring how similarity between paired observations decreases with distance (Matheron, 1963; Webster and Oliver, 2001) [18,43] . The semi variance γ (h) is calculated as: (1) where z(xi) is the value at location xi, h is the lag distance, and N(h) is the number of sample pairs separated by distance h. Semivariogram models, such as spherical and exponential, were fitted to the data to describe the spatial structure and optimize kriging weights. Key parameters-nugget (C₀), sill (C + C₀), and range (a) were derived, representing measurement error or micro-scale variability, total variance, and the distance over which spatial correlation persists, respectively (Lopez-Granados et al., 2002) [17] . Statistical analyses of soil parameters included computation of mean, standard deviation, coefficient of variation, skewness, and kurtosis, alongside evaluation of normality using histograms and QQ plots. Prediction accuracy of kriging was validated using error metrics such as mean error (ME), root mean square error (RMSE), root mean square standardized error (RMSSE), and average standard error (ASE). The formulae for ME and RMSE are:

The semivariogram parameters and kriging weights provided a robust framework for spatial interpolation of soil fertility parameters, allowing the generation of thematic maps for sitespecific nutrient management in the sugarcane-growing areas of Ambad Taluka. Across the soil parameter, the spherical and exponential models were found to be best fitted. The semivariogram models used are shown in Equations ( 4) and ( 5

Otherwise Exponential model:

Using the model semivariogram, basic spatial parameters such as nugget (C 0 ), sill (C+C 0 ) and range (a) was calculated which describes overall structure and inputs used for interpolation. Nugget is the variance at zero distance, sill (g(h)) is the lag distance between measurement at which one value for a variable does not influence neighboring values and range is the lag distance between measurements at which one value for a variable does not influence neighbouring values and range is the distance at which values of one variable become spatially independent of another (Lopez-Granadose et al. 2002) [17] Soil dependancy of soil parameters (the ratio of nuggets to sill variances) is expressed as percentage.

Nutrient Index

The Nutrient Index (NI) was calculated to assess the overall fertility status of surface soils in the study area, following the method proposed by Ramamoorthy and Bajaj (1969) [28] . This index integrates the distribution of nutrient levels across soil samples, providing a quantitative measure of soil nutrient sufficiency. The formula used is:

Where N L , N m and N H are numbers of samples falling in low, medium and high classes of nutrient status and N T is total no. of samples analyzed for a given area. [5,9] . It also reflects long-term pedogenic factors such as the nature of parent material, climatic conditions, and leaching intensity, which together shape soil development (Wild, 1993; Xu et al., 2020) [44,45] . In Ambad Taluka, the spatial distribution pattern of pH (Fig. 1) revealed that the soils range from slightly alkaline to strongly alkaline, with values between 7.1 and 9.1, and an average pH of 7.86 ( [4,26] . The pH data exhibited a normal distribution, and the coefficient of variation (CV < 15%) indicated low variability across the landscape, suggesting relative uniformity in soil reaction likely governed by the region's homogeneous parent material and limited leaching.

The reclassified soil pH map, based on AIS & LUS (1971) [2] guidelines, further showed that 81.9% (57,858.4 ha) of the area falls under the alkaline category. An additional 13.5% (9,540.5 ha) is moderately alkaline, while only 4.6% (3,253.3 ha) exhibits slightly alkaline characteristics ( [3] and Panhwar et al. (2003) [25] who reported dominant alkaline conditions influenced by parent material and regional climatic patterns.

Electrical Conductivity (EC)

Soil electrical conductivity (EC) is an important indicator of soil salinity and overall soil health, as it influences nutrient availability and crop productivity. In Ambad Taluka, the measured EC values ranged from 0.10 to 1.0 dS m⁻¹, with a mean value of 0.43 dS m⁻¹ (Table 3), which is well below the threshold for salinity risk (<1 dS m⁻¹), indicating that salinity is not a limiting factor for crop growth in the study area.

For clarity in mapping, the EC values were grouped into three categories: <0.4, 0.4-0.6, and 0.6-1.0 dS m⁻¹. The spatial analysis revealed that the majority of the area (44,114.99 ha; 62.43% of total geographical area, TGA) falls within the 0.4-0.6 dS m⁻¹ range. Approximately 25,661.18 ha (36.32% of TGA) were classified in the <0.4 dS m⁻¹ category, while very small pockets (876.07 ha; 1.24% of TGA) located mainly in the central and southern parts of the study area exhibited slightly higher EC values (0.6-1.0 dS m⁻¹) (Fig. 1, Table 4). These findings are in agreement with earlier studies by Prasad et al. (2020) [27] , which reported low to moderate EC levels in soils of similar agro-climatic conditions. The overall low EC levels suggest that soil salinity does not pose a constraint to sugarcane cultivation in Ambad Taluka.

Soil Organic Carbon (OC)

Soil organic carbon (SOC) is one of the most important indicators of soil fertility and soil health because it regulates the cycling, storage, and availability of key nutrients, especially nitrogen, phosphorus, sulfur, and various micronutrients (Stevenson, 1982) [39] . Adequate SOC improves soil aggregate stability, enhances water infiltration and retention, reduces bulk density, and strengthens the soil's resilience against erosion and degradation (Smith & Elliott, 1990; Lal, 2004) [37,16] . In addition, SOC plays a vital role in sustaining microbial activity and enzymatic processes that drive nutrient mineralization and soil biological functioning.

In the present study, SOC values in Ambad Taluka ranged from 0.11 to 1.49%, with a mean value of 0.32%, and exhibited a normal distribution pattern (Table 3). These values indicate overall poor soil organic matter status, which is a common feature in semi-arid tropical regions of India. One of the major reasons for low SOC in this region is the limited use of organic manures, such as farmyard manure, compost, and green manures. Due to increasing fertilizer dependency, most farmers apply chemical fertilizers without supplementing organic sources, leading to gradual depletion of soil carbon stocks over time (Ghosh et al., 2012) [10] . Another key factor is trash burning, particularly in sugarcane-based production systems. Farmers often burn sugarcane trash to ease land preparation, which results in the loss of large quantities of organic residues that could otherwise contribute to carbon buildup in soils (Sandhu et al., 2019) [30] . Burning residues accelerates carbon oxidation and eliminates potential additions to soil organic matter pools, thereby contributing significantly to SOC decline. Spatial analysis of SOC distribution revealed distinct carbon deficiency zones across the Taluka. The majority of the area 51,566.9 ha (73% of TGA) fell under the low SOC category. Soils with moderate SOC occupied 13,022. [35,36,19,20] .

Available Nitrogen (N)

Nitrogen is a fundamental macronutrient that controls vegetative growth, tillering, chlorophyll formation, and overall biomass production, making it one of the most yield-limiting nutrients in sugarcane. In the present study, available nitrogen in the soils of Ambad Taluka ranged from 13.5 to 195.5 kg ha⁻¹, with a mean of 58.86 kg ha⁻¹ (Table 3), and exhibited very low variability (CV = 0.42%), indicating that nitrogen deficiency is widespread across the region. Spatial mapping showed that 94.7% of the total geographical area (66,936.9 ha) fell under the very low nitrogen category, while 5.21% (3,685.5 ha) was classified as low, and only a negligible area (29.8 ha; 0.04%) recorded moderate nitrogen levels (Table 4; Fig. 2). The dominance of low nitrogen is largely attributed to heavy nitrogen losses through volatilization under alkaline soil conditions, leaching of nitrate beyond the root zone during irrigation or rainfall, and denitrification in intermittently waterlogged microsites. Another major reason is the common practice of applying nitrogen fertilizers in a single large dose, which leads to severe mismatch between fertilizer supply and crop demand, thereby reducing nitrogen use efficiency. These findings agree with the observations of Sharma et al. (2020) [33] and Khalid et al. (2012) [13] , who reported similar nitrogen depletion trends under tropical agricultural conditions. In sugarcane, nitrogen requirement is highly stage-specific; therefore, for efficient utilization, 10% of the recommended dose should be applied at planting, followed by 40% at 6-8 weeks to promote rapid tiller formation, another 10% at 12-14 weeks during early internode development, and the remaining 40% at the grand growth stage (around 120 days after planting), along with 50% of phosphorus and potassium. Such split application ensures synchronized nutrient release with crop demand, minimizes losses, and improves nitrogen use efficiency in sugarcane production systems.

Available Phosphorus (P)

Phosphorus is a vital macronutrient in sugarcane cultivation, particularly due to its role in promoting early root development, enhancing tillering, and improving overall vigour of the crop. In the soils of Ambad Taluka, the available phosphorus content ranged from 4.35 to 54.5 kg ha⁻¹, with a mean of 13.87 kg ha⁻¹ (Table 3), and the data exhibited low variability (CV = 0.73%), indicating relatively uniform phosphorus deficiency across the study area. Spatial distribution analysis showed that more than half of the total geographical area (35,580.9 ha) fell under the low phosphorus category, while an additional 17.9% (12,632.2 ha) was classified as very low, reflecting widespread scarcity of plant-available P (Fig. 3). Moderate and very high phosphorus categories each occupied substantial portions of land (7,147.6 ha and 7,089.8 ha, respectively), whereas moderately high (5,321.6 ha) and high P areas (2,880.4 ha) represented smaller but notable pockets of phosphorus sufficiency (Table 4). The predominance of low phosphorus levels can be attributed to the high calcium carbonate content of the region's soils, which leads to fixation of phosphorus in the form of insoluble calcium phosphate, significantly reducing its availability to plants. Furthermore, limited use of phosphorus fertilizers, poor adoption of soil testbased recommendations, and inadequate application of organic manures exacerbate P deficiency, as organic inputs help solubilize fixed phosphorus and improve P availability through microbial activity. These observations align with the findings of Memon et al. (1992) [21] and Sharma et al. (2020) [33] , who also reported phosphorus scarcity in calcareous soils due to fixation and suboptimal fertilizer management practices.

Available Potassium (K)

Potassium is the third major nutrient essential for sugarcane growth, contributing to enzyme activation, water regulation, stress tolerance, and sugar translocation within the stalks. In the soils of Ambad Taluka, available potassium ranged from 202 to 995 kg ha⁻¹, with a mean value of 446.5 kg ha⁻¹ (Table 3), and showed low variability (CV = 0.30%), indicating uniformly high potassium status across the study area. Spatial distribution revealed that a vast majority of the area (61,846.4 ha; 87.53% of TGA) belonged to the very high potassium category, while the high class accounted for 6,998.6 ha (9.9%). Only a small fraction of the land 1,806.9 ha (2.55%) fell under the moderately high category, and none of the soils were classified as low or very low in potassium availability (Table 4). The abundance of available potassium can be attributed to the dominance of Kbearing minerals such as mica (biotite), illite, and K-feldspars in the basaltic parent material, which undergo weathering and continuously release potassium into the soil. Moreover, the recycling of crop residues and upward movement of potassium through capillary rise during dry spells further enrich the surface soil with exchangeable K. Frequent drought occurrences in Marathwada, typically every three years, may also enhance the importance of potassium, as it plays a pivotal role in osmotic regulation and improves plant resilience under moisture stress. Similar trends of elevated potassium in black soils derived from basalt have been reported by Spark et al. (1987) and Khadka et al. (2019) [12] . Despite the high natural K reserves, balanced fertilization is still necessary to avoid nutrient imbalance, particularly in intensive sugarcane cultivation systems.

Soil Nutrient Index (SNI)

The Soil Nutrient Index (SNI) is a reliable indicator used to assess the overall fertility status of an area by integrating the distribution of soil samples across low, medium, and high nutrient categories. It provides a composite value that reflects the nutrient sufficiency or deficiency at the landscape level. In the context of Ambad Taluka, the SNI helps classify the status of available N, P, K, and other nutrients, thereby identifying areas requiring corrective nutrient management. The index is calculated using a weighted approach, where low, medium, and high nutrient classes are assigned scores of 1, 2, and 3, respectively. The resulting index value categorizes the nutrient status as low (<1.67), medium (1.67-2.33), or high (>2.33). This quantitative classification is useful for recommending sitespecific nutrient management strategies and aligning fertilizer application with soil fertility needs. Moreover, the SNI serves as a guiding tool for identifying nutrient-deficient zones, prioritizing extension activities, and planning balanced fertilization programs. Earlier studies by Ramamoorthy and Bajaj (1969) [28] and Kanwar (1976) emphasized the effectiveness of nutrient indexing for large-scale fertility evaluation, which further supports its application in the present study area. In contrast, available nitrogen followed the spherical model, implying a clearer sill and a well-defined spatial continuity structure (Table 5).

The range values representing the distance beyond which samples become spatially independent varied considerably among soil properties. Larger ranges were observed for pH (7859.3 m) and available phosphorus (7590.5 m), suggesting these properties are mainly governed by intrinsic soil-forming factors such as parent material, mineralogical composition, and long-term pedogenic processes. Since these characteristics do not change abruptly, their spatial influence extends over larger distances.

On the other hand, relatively shorter ranges were recorded for EC (6189.08 m), organic carbon (6643.18 m), available nitrogen (132.41 m), and available potassium (6647.1 m). These shorter ranges reflect a smaller zone of influence, implying that these properties vary more abruptly across the landscape. Such shortdistance variability is commonly associated with extrinsic factors such as heterogeneous fertilizer application, residue management, tillage operations, and localized soil degradation processes. For example, available nitrogen showed an extremely short range, likely due to the highly mobile and dynamic nature of N in the soil, coupled with farmers' practice of applying nitrogenous fertilizers in large quantities at a single time, leading to patchy distribution. The nugget-to-sill ratio, used to categorize spatial dependence (Cambardella et al., 1994) [6] , revealed that available nitrogen and phosphorus exhibited moderate spatial dependence, indicating that both intrinsic and extrinsic factors influence their spatial patterns. Strong spatial dependence was observed for parameters with nugget-sill ratios below 0.25, implying that their variability is largely governed by natural soil-forming processes rather than management-induced variability. Interpretation of the spatial variability maps further supported these findings. The pH map indicated that about 81.89% of the area falls under alkaline conditions (pH 8.4 -9.0), which is consistent with soils derived from basaltic parent material, known for their basic reaction due to the dominance of Ca-and Mg-bearing minerals (Bhaskar et al., 2014) [4] . Electrical conductivity showed that 62.43% of the area lies in the medium range (0.4 -0.6 dS m⁻¹), suggesting moderate salt accumulation influenced by irrigation practices and soil texture. Organic carbon displayed a strong predominance of low values (72.98% of the area), reflecting low biomass return, trash burning in sugarcane fields, and accelerated carbon oxidation due to high temperature. Available nitrogen showed extremely low levels across 94.07% of the landscape, in agreement with its short range and moderate spatial dependence, driven by episodic fertilizer use and rapid N transformations. Available phosphorus mapped to 50.55% of the area falling under the very low category, likely due to phosphorus fixation in calcareous soils and insufficient application strategies. Conversely, available potassium was very high in 87.53% of the area, attributed to the abundance of K-rich primary minerals such as biotite mica and K-feldspars in the basaltic parent material, consistent with observations by Vasu et al. (2017) [41] . Overall, the semivariogram analysis highlights that soil chemical properties in Ambad Taluka are shaped by a combination of natural pedogenic factors and anthropogenic interventions. Understanding these spatial patterns is crucial for designing sitespecific nutrient management strategies and enhancing fertilizeruse efficiency in the region. (Cambardella et al., 1994) [6] In contrast, soil organic carbon and major nutrients displayed high variability over short distances, largely influenced by extrinsic factors such as fertilizer management, tillage, and crop residue handling. Geostatistical modeling of these parameters provides a robust approach to assess spatial variability and facilitates site-specific nutrient management, which can enhance sugarcane productivity while sustaining soil fertility. Cross-validation of soil properties using the kriging interpolation method demonstrated reliable predictions with minimal errors. The application of GIS techniques for spatial interpolation of unsampled locations offers an effective, time-efficient, and costeffective tool for managing soil fertility over extensive sugarcane-growing areas. This approach can also support the development of soil health cards for operational areas of sugar mills, enabling more informed and targeted nutrient management strategies.

Acknowledgement

AcknowledgementsThe authors are grateful to the Director General of the Institute for the support and also Samarth SSK, sugar mill.

References

1. Abbas G, Khan M, Ahmed R, Ali S. Soil fertility variability and GIS mapping in relation to crop productivity. Pak J Agric Sci. 2011;48(3):123-30.
2. AIS & LUS. Soil Survey Manual. New Delhi: All India Soil & Land Use Survey; 1971.
3. Arain MS, Khan AB, Lee CD. Soil characterization and classification in relation to parent material. Soil Environ. 2011;30(2):85-96.
4. Bhaskar BP, Sarkar D, Mandal C, Bobade SV, Gaikwad MS, Gaikwad SS, et al. Reconnaissance soil survey of Yavatmal district, Maharashtra, India. Nagpur: NBSS and LUP; 2014. NBSS Publ. no. 1059. p. 208.
5. Brady NC, Weil RR. The Nature and Properties of Soils. 15th ed. Pearson; 2017.
6. Cambardella CA, Moorman TB, Novak JM, Parkin TB, Karlen DL, Turco RF, et al. Field scale variability of soil properties in central Iowa soils. Soil Sci Soc Am J. 1994;58(5):1501-11.
7. Dayani M, Mohammadi J. Geostatistical assessment of Pb, Zn and Cd contamination in near surface soils of the urban mining transitional region of Isfahan, Iran. Pedosphere. 2010;20(5):568-77.
8. ESRI. The principles of geostatistics analysis. New York: Environmental Systems Research Institute; 2003. p. 54.
9. Fageria NK, Baligar VC, Jones CA. Growth and Mineral Nutrition of Field Crops. CRC Press; 2011.
10. Ghosh A, Singh R, Sharma P. Long-term fertility management and its impact on soil organic carbon. J Indian Soc Soil Sci. 2012;60(3):245-52.
11. Jackson ML. Soil Chemical Analysis. New Delhi: Prentice Hall of India Pvt Ltd; 1973.
12. Khadka D, Lamichhane S, Amgain R, Joshi S, Vista S, Sah K, et al. Soil fertility assessment and mapping spatial distribution of Agricultural Research Station, Bijayanagar, Jumla, Nepal. Eurasian J Soil Sci. 2019;8(3):237-48.
13. Khalid R, Mahmood T, Bibi R, Siddique MT, Alvi S, Naz SY. Distribution and indexation of plant available nutrients of rainfed calcareous soils of Pakistan. Soil Environ. 2012;31(2):146-51.
14. Khan MZ, Islam R, Salam ABA, Tama Ray. Spatial variability and geostatistical analysis of soil properties in the diversified cropping regions of Bangladesh using GIS techniques. Appl Environ Soil Sci. 2020;2020:1-19.
15. Krige DG. A statistical approach to some basic mine valuation problems on the Witwatersrand. J Chem Metall Min Soc S Afr. 1951;52(6):119-39.
16. Lal R. Soil carbon sequestration impacts on global climate change and food security. Science. 2004;304:1623-7.
17. Lopez-Granados F, Jurado-Exposito M, Pena-Barragan JM, Garcia-Torres L. Using geostatistical and remote sensing approaches for mapping soil properties. Eur J Agron. 2005;23(3):279-89.
18. Matheron G. Principles of geostatistics. Econ Geol. 1963;58:1246-66. http://dx.doi.org/10.2113/gsecongeo.58.8.1246
19. Meena HB, Sharma RP, Rawat US. Status of macro and micronutrients in some soils of Tonk district of Rajasthan. Indian J Soil Sci. 2006;54:508-12.
20. Meena RS, et al. Organic carbon status of Indian soils and factors affecting it. Agropedology. 2006.
21. Memon KS, Rashid A, Puno HK. Phosphorus deficiency diagnosis and P soil test calibration in Pakistan. In: Proceeding seminar on phosphorus decision support system. Texas: USA; 1992. p. 125-47.
22. Muller TG, Pierce FJ. Soil carbon maps enhancing spatial estimates with simple terrain attributes at multiple scales. Soil Sci Soc Am J. 2003;67:258-67.
23. Olsen S, Cole CV, Watanabe FS, Dean LA. Estimation of available phosphorus in soil by extraction with sodium bicarbonate. Washington, DC: United States Department of Agriculture Circular; 1954. p. 939.
24. Panday D, Maharjan B, Chalise D, Shrestha RK, Twanabasu B. Digital soil mapping in the Bara district of Nepal using kriging tool in ArcGIS. PLoS One. 2018.
25. Panhwar RN, Keerio HK, Memon YM, Junejo S, Arain MY, Chohan M, et al. Response of Thatta-10 sugarcane variety to soil and foliar application of zinc sulphate under half and full doses of NPK fertilizers. J Appl Sci. 2003;3(4):266-9.
26. Patil RD, Dasog GS. Soil properties of basaltic landscapes in semi-arid tropics. J Indian Soc Soil Sci. 2018.
27. Prasad PNS, Subbarayappa CT, Sathish A. Spatial distribution and mapping of available micronutrient status in relation to soil properties of tomato grown soils of Chikkaballapura District, Karnataka. Int J Plant Soil Sci. 2020;32(10):1-10.
28. Ramamoorthy B, Bajaj JC. Available nitrogen, phosphorus and potassium status of Indian soils. Fertil News. 1969;14:25-36.
29. Reza SK, Nayak DC, Chattopadhyay T, Mukhopadhyay S, Singh SK, Srinivasan R. Spatial distribution of soil physical properties of alluvial soils: a geostatistical approach. Arch Agron Soil Sci. 2016;62(7):972-81.
30. Sandhu H, Dhillon PS, Rupana MS, Sandhu SK. Sugarcane trash burning effects on soil carbon and nutrient dynamics. Environ Sci Pollut Res. 2019.
31. Santra P, Kumar M, Panwar N. Digital soil mapping of sand content in arid western India through geostatistical approaches. Geoderma Reg. 2017;9:56-72.
32. Sharma KL, Singh AK. Alkaline soils and their management. Indian J Soil Conserv. 2017.
33. Sharma RP, Chattaraj S, Vasu D, Karthikeyan K, Tiwary P, Naitam RK, et al. Spatial variability assessment of soil fertility in black soils of central India using geostatistical modeling. Arch Agron Soil Sci. 2020;67:876-88. https://doi.org/10.1080/03650340.2020.1766678
34. Shit PK, Bhunia GS, Maiti R. Spatial analysis of soil properties using GIS based geostatistics models. Model Earth Syst Environ. 2016;2:2-6.
35. Singh RP, Singh GP, Sharma AK. Spatial variability of soil organic carbon under semi-arid conditions. Catena. 2014;113:251-9.
36. Singh YP, Verma SK, Dubey SR. Soil fertility status and correlation of available macro and micronutrients in Chambal region of Madhya Pradesh. Indian J Soil Sci. 2014;62(4):369-75.
37. Smith JL, Elliott ET. Tillage and residue management effects on soil organic matter dynamics. Soil Sci Soc Am J. 1990.
38. Sparks DL. Potassium dynamics in soils. In: Stewart BA, editor. Advances in Soil Science. Vol. 6. New York: Springer; 1987. p. 1-63.
39. Stevenson FJ. Humus Chemistry: Genesis, Composition, Reactions. New York: Wiley; 1982.
40. Subbaiah BV, Asija CL. A rapid procedure for the estimation of available nitrogen in soils. Curr Sci. 1956;25:259-60.
41. Vasu D, Singh SK, Sahu N, Tiwary P, Chandran P, Duraisami VP, et al. Assessment of spatial variability of soil properties using geospatial techniques for farm level nutrient management. Soil Tillage Res. 2017;169:25-34.
42. Walkey A, Black IA. An examination of soil organic carbon by the chromic acid titration method. Soil Sci. 1934;37(1):29-38.
43. Webster R, Oliver MA. Geostatistics for Environmental Scientists. New York: Wiley; 2001.
44. Wild A. Soils and the Environment. Cambridge: Cambridge University Press; 1993.
45. Xu X, et al. Soil pH as a key regulator of soil chemical and biological processes. Soil Biol Biochem. 2020.