Local and regional processes have considerable impact on the tropical deciduous forest (TDF) vegetation dynamics. To determine association between the soil physico-chemical properties and woody species assemblage in TDF, we applied an indirect ordination method. We randomly selected five sites in TDF of Vindhyan highlands, India and established nine plots (50×20m) at each site, and measurement was done for woody species assemblage in each plot. The soil physico-chemical properties measured at each site include soil moisture content (SMC), soil depth, clay, silt, sand, rockiness, bulk density, pH, organic carbon, total nitrogen and total phosphorus. We observed considerable impact of soil physico-chemical properties on the woody species assemblage in TDFs. We found that the detrended correspondence analysis (DCA) ordination divided the woody species into three major communities, i.e., moist, semi-dry and dry. Our study suggests extensive investigation of vegetation parameters in TDFs, and their association with local and regional environmental processes with the application of advanced ordination techniques.

Keywords: tropical deciduous forest; woody species assembly; soil physico-chemical properties; detrended correspondence analysis

Recently, several modeling approaches have been started which integrate the influence of local as well as regional environmental processes on the local species assemblage and diversity.^1‒4 Studies have reported that the local and regional processes have considerable impact on the tropical deciduous forest (TDF) vegetation dynamics.⁵ The TDFs are characterized by severe climatic conditions and experience drought period of several months each year.⁵ It has been emphasized that TDFs exhibit remarkable patchiness in the plant species assemblage as a result of small differences in the soil physico-chemical properties.⁶ The diversity⁷ and seasonal growth^8‒10 of woody species in TDFs are highly influenced by soil physico-chemical properties,^5,11 therefore it is important to understand the effect of soil conditions on the woody species assemblage.

To determine association between the soil physico-chemical properties and woody species assemblage in TDF, we applied an indirect ordination method, which is considered a powerful tool for extracting hypothetical soil physico-chemical properties from the species assemblage data.^12‒14 For this study, we identified five unmanaged forest fragments in Vindhyan highlands, Uttar Pradesh, India (24º18´07˝ - 25º00´17˝ N, 82º37´38˝ - 83º23´05˝ E). Forest area selected in our study experiences a typical monsoonal climate exhibiting three important seasons each year

1. Summer, which starts from April and ends in mid June,
2. Rainy, which starts from mid June and ends in September, and
3. Winter which starts from November and ends in February.¹⁵

Based on the meteorological data taken from the regional weather stations established in the regional forest department of Uttar Pradesh for the year 1980–2010, the average rainfall per year varies from 865mm to 1,196mm.¹⁶ Generally, around 85% of the rainfall, each year is recorded in the monsoon season, especially from the monsoonal winds coming from south west direction, while the remaining rainfall is commonly registered in December as well as in May–June. The forest region experience a long dry period ranging from eight to nine months, which starts from October or November and ends in mid June.⁵ The monthly rainfall in the forest region ranges from the lowest value of 6 mm in the month of April to highest value of 334 mm in August.¹⁷ Soils of the study region have been reported as residual ultisols, with texture of sandy-loam, light red to grayish color, and nutrient conditions very poor.¹⁸ The recent estimation of physicochemical soil properties of the forest region have been clearly described in Chaturvedi and Raghubanshi.¹⁹ Based on the density-dbh woody species distributions investigated in the study region, the forest registered the dominance of low dbh stems, and the mean densities of seedling, sapling, and adult stems were 9,261±1,511, 799±154, and 297±62, respectively.^7,20,21

The selected forest fragments in our study represented a range in soil water availability. At each selected site, we randomly established nine plots (50×20m) and measurement was done for woody species assemblage. The soil physico-chemical properties measured at each site include soil moisture content (SMC), soil depth, clay, silt, sand, rockiness, bulk density, pH, organic soil carbon, total soil nitrogen and total soil phosphorus. For detail protocol for the measurement of species assemblage and soil physico-chemical properties, see Chaturvedi,¹⁵ Chaturvedi & Raghubanshi.¹⁸ The ordination analyses were done by using the PC-ORD 5 program.²² Generally, a unimodal (non-negative) species response curve (i.e., Gaussian response curve) is suggested for datasets which cover wide range of environmental gradients, therefore we applied the DCA (detrended correspondence analysis) ordination.^12,14 We performed DECORANA to explore the relationship between the woody species assemblage from each habitat and the selected soil physico-chemical properties.¹² In this method, DCA develops a theoretical variable axis (DCA axis 1) which best describes the magnitude of variance in the woody species assemblage between different local communities, and also constructs second, third and further DCA axes by using the constraint which they need not to be correlated with the previous DCA axes.¹⁴ In the ordination diagram, relative positions or distance of the plots distributed along the DCA axes provide an estimate of the floristic similarity among the plots. Here, we also emphasize that the relative position of our plots situated inside the multidimensional DCA ordination space is primarily on the basis of similarities in the woody species assemblage.

The DCA ordination which we developed for our plots indicated strong correlations between the plot positions along the DCA axes (on the basis of species assemblage) and soil physico-chemical properties (Figure 1). This suggests that the variations in the plot positions within the DCA ordination space may be explained majorly on the basis of the soil physico-chemical properties. Moreover, the first ordination axis had high eigenvalue, indicating a high resolution. Results showed that the first ordination axis have strong negative correlations with the mean soil moisture content (SMC) (R=-0.614), clay (R=-0.550), soil depth (R=-0.399) organic carbon (R =-0.288), total nitrogen (R=-0.390) and total phosphorus (R = -0.328), and positive correlations with soil pH (R= 0.637), sand (R=0.509) and rockiness (R=0.380) (Table 1), whereas the second DCA axis exhibited positive correlation with total phosphorus (R=0.457). The third DCA axis showed negative correlation with bulk density. It was also observed that soil pH (R=-0.74, P<0.001) and bulk density (R=-0.56, P<0.001) exhibited strong negative correlation and total phosphorus showed positive association (R=0.49, P<0.01) with SMC. We observed that the DCA ordination divided the woody species into three major communities, i.e., moist, semi-dry and dry. The species present in these communities are as follows:

Figure 1 DCA ordination of the woody species on the basis of soil physico-chemical properties. SMC, soil moisture content; C, soil organic carbon; 1, Hathinala; 2, Gaighat; 3, Harnakachar; 4, Ranitali; 5, Kotwa. Species abbreviations are given in Chaturvedi.¹⁵ Source: Chaturvedi.¹⁵

Moist

Adina cordifolia, Bauhinia racemosa, Bridelia retusa, Cassia fistula, Cissampelos pareira, Elaeodendron quinquelocularis, Gardenia latifolia, Gardenia turgida, Grewia serrulata, Miliusa tomentosa, Mitragyna parvifolia, Pterocarpus marsupium, Schleichera oleosa, Schrebera swietenioides, Semecarpus anacardium, Shorea robusta, Sterculia urens, Stereospermum suaveolens, Terminalia chebula and Terminalia tomentosa.

Semi-dry

Acacia auriculiformis, Aegle marmelos, Anogeissus latifolia, Azadirachta indica, Boswellia serrata, Buchanania lanzan, Cassia siamea, Chloroxylon swietenia, Diospyros melanoxylon, Elaeodendron glaucum, Grewia hirsuta, Hardwickia binata, Holarrhena integrifolia, Hymenodictyon excelsum, Indigofera cassioides, Lagerstroemia parviflora, Lantana camara, Madhuca longifolia, Soymeda febrifuga, Woodfordia fruticosa and Ziziphus nummularia.

Dry

Abrus precatorius, Acacia catechu, Albizia odoratissima, Carissa spinarum, Dendrocalamus strictus, Emblica officinalis, Flacourtia indica, Holarrhena antidysenterica, Lannea coromandelica, Nyctanthis arbortristis, Ziziphus glaberrima and Ziziphus oenoplea.

These communities differ from each other on the basis of variations in effects of different soil physico-chemical properties on them. On one hand, there were effects of SMC, clay, nitrogen and phosphorus on moist and semi-dry communities, while on the other hand, dry community was strongly affected by pH, sand and rockiness.

We observed considerable impact of soil physico-chemical properties on the woody species assemblage in TDFs. With the application of ordination methods, we categorized woody species of TDFs on the basis of their association with soil physico-chemical properties. Our study suggests extensive investigation of vegetation parameters in TDFs, and their association with local and regional environmental processes with the application of advanced ordination techniques.

RKC thanks Council of Scientific and Industrial Research, India (award no. 09/13(452)/2012-EMR-I) and National Natural Science Foundation of China (NSFC), Chinese Academy of Science, China (grant No. 31750110466) for financial support.

The author declares there is no conflict of interest.

1. Herrick JE, Lessard VC, Spaeth KE, et al. National ecosystem assessments supported by scientific and local knowledge. Frontiers in Ecology and the Environment. 2010;8:403–408.
2. Wisz MS, Pottier J, Kissling WD, et al. The role of biotic interactions in shaping distributions and realised assemblages of species: implications for species distribution modelling. Biological Reviews. 2013;88(1):15–30. 
3. Laliberté E, Zemunik G, Turner BL. Environmental filtering explains variation in plant diversity along resource gradients. Science. 2014;345:1602–1605.
4. Ash JD, Givnish TJ, Waller DM. Tracking lags in historical plant species’ shifts in relation to regional climate change. Global Change Biology. 2017;23:1305–1315.
5. Singh JS, Chaturvedi RK. Tropical Dry Deciduous Forest: Research Trends and Emerging Features. Springer Nature Singapore Pte Ltd., Singapore. 2018.
6. Chaturvedi RK, Raghubanshi AS, Singh JS. Effect of small scale variations in environmental factors on the distribution of woody species in tropical deciduous forests of Vindhyan Highlands, India. Journal of Botany. 2011;10.
7. Chaturvedi RK, Raghubanshi AS. Species Composition, Distribution and Diversity of Woody Species in tropical dry forest of India. Journal of Sustainable Forestry. 2014;33(8):729–756.
8. Chaturvedi RK, Raghubanshi AS, Singh JS. Leaf attributes and tree growth in a tropical dry forest. Journal of Vegetation Science. 2011;22(5):917–931.
9. Chaturvedi RK, Raghubanshi AS, Singh JS. Growth of tree seedlings in a dry tropical forest in relation to soil moisture and leaf traits. Journal of Plant Ecology. 2013;6(2):158–170.
10. Chaturvedi RK, Raghubanshi AS, Singh JS. Relative effects of different leaf attributes on sapling growth in tropical dry forest. Journal of Plant Ecology. 2014;7(6):544–558.
11. Chaturvedi RK, Raghubanshi AS, Singh JS. Plant functional traits with particular reference to dry deciduous forests: a review. Journal of Biosciences. 2011;36(5):963–981.
12. Hill MO, Gauch HG. Detrended correspondence analysis, an improved ordination technique. Vegetatio. 1980;42:47–58.
13. Peet RK, Knox RG, Case JS. Putting things in order: the advantages of detrended correspondence analysis. American Naturalist. 1988;131(6):924–934.
14. Jongman RHG, Ter Braak CJF, Van Tongeren OFR. Data analysis in community and landscape ecology. Pudoc, Wagening, The Netherlands: Cambridge University Press. 1995.
15. Chaturvedi RK. Plant functional traits in dry deciduous forests of India. PhD Thesis, Banaras Hindu University, Varanasi, India. 2010.
16. Chaturvedi RK, Raghubanshi AS, Singh JS. Carbon density and accumulation in woody species of tropical dry forest in India. Forest Ecology and Management. 2011;262(8):1576–1588.
17. Chaturvedi RK, Raghubanshi AS, Singh JS. Effect of grazing and harvesting on diversity, recruitment and carbon accumulation of juvenile trees in tropical dry forests. Forest Ecology and Management. 2012;284(2012):152–162.
18. Chaturvedi RK, Raghubanshi A. Plant Functional Traits in a Tropical Deciduous Forest: An analysis. Lambert Academic Publishing GmbH & Co KG, Berlin, Germany. 2011.
19. Chaturvedi RK, Raghubanshi AS. Assessment of carbon density and accumulation in mono- and multi-specific stands in tropical dry forests of India. Forest Ecology and Management. 2015;339(2015):11–21.
20. Chaturvedi RK, Raghubanshi AS, Singh JS. Sapling harvest: A predominant factor affecting future composition of tropical dry forests. Forest Ecology and Management. 2017;384:221–235.
21. Chaturvedi RK, Raghubanshi AS, Tomlinson KW, et al. Impacts of human disturbance in tropical dry forests increase with soil moisture stress. Journal of Vegetation Science. 2017;28(5):997–1007.
22. McCune B, Mefford MJ. Multivariate analysis on the PC-ORD system. MjM Software, Gleneden Beach, USA. 2005.