Determinants of phytolith occluded carbon in bamboo stands across forest types in the eastern Indian Himalayas.

Phytoliths are known to play a significant role in the global carbon cycle by sequestering atmospheric carbon dioxide as phytolith-occluded carbon (PhytOC) for a long time. Given the resistant nature of phytolith to decomposition, PhytOC can represent up to 82 % of total carbon in some soil and sediments even after 2000 years of litter decomposition. Hence, forests with high PhytOC sequestration rates could play a critical role in increasing terrestrial carbon storage. In this study, we quantified the variation in PhytOC concentrations in bamboo leaves, branches and culms with forest types in the Eastern Indian Himalayas as bamboos are efficient accumulator of phytolith and PhytOC due to their fast growth and high biomass accumulation rates. Using nine different machine learning techniques, we also investigated the determinants of PhytOC production in bamboo stands in the study area in India. The results revealed that the PhytOC concentrations in bamboo stands were in the order of leaf (3.0 g kg-1) > culm (1.0 g kg-1) > branch (0.2 g kg-1) across forest types. The highest PhytOC stock (53.8 kg ha-1) was found in bamboo stands in the subtropical pine forests (1900-3500 m elevation), while the lowest (28.0 kg ha-1) was in the tropical evergreen forests (<900 m elevation). Machine learning techniques established a positive correlation of PhytOC content in leaf and total PhytOC content with soil available phosphorus, elevation, total nitrogen, exchangeable potassium, atmospheric humidity, SOC content, CEC and pH. Numerical evaluation criteria and graphic methods identified artificial neural network (ANN) and support vector regression as the superior techniques with a root mean square error value of 0.52 kg ha-1 and 0.59 kg ha-1 respectively. The results of these two models were found to be better among all the nine machine learning algorithms used. The high PhytOC storage in the bamboo stands in the Indian Himalayan region suggests that forest management could secure a stable carbon sink on a millennial scale.

Assessment of sal (Shorea robusta) forest phenology and its response to climatic variables in India.

Remote sensing-based observation provides an opportunity to study the spatiotemporal variations of plant phenology across the landscapes. This study aims to examine the phenological variations of different types of sal (Shorea robusta) forests in India and also to explore the relationship between phenology metrics and climatic parameters. Sal, one of the main timber-producing species of India, can be categorized into dry, moist, and very moist sal. The phenological metrics of different types of sal forests were extracted from Moderate Resolution Imaging Spectroradiometer (MODIS)-derived Enhanced Vegetation Index (EVI) time series data (2002-2015). During the study period, the average start of season (SOS) was found to be 16 May, 17 July, and 29 June for very moist, moist, and dry sal forests, respectively. The spatial distribution of mean SOS was mapped as well as the impact of climatic variables (temperature and rainfall) on SOS was investigated during the study period. In relation to the rainfall, values of the coefficient of determination (R2) for very moist, moist, and dry sal forests were 0.69, 0.68, and 0.76, respectively. However, with temperature, R2 values were found higher (R2 = 0.97, 0.81, and 0.97 for very moist, moist, and dry sal, respectively). The present study concluded that MODIS EVI is well capable of capturing the phenological metrics of different types of sal forests across different biogeographic provinces of India. SOS and length of season (LOS) were found to be the key phenology metrics to distinguish the different types of sal forests in India and temperature has a greater influence on SOS than rainfall in sal forests of India.

Seasonal dynamics and impact factors of atmospheric CO2 concentration over subtropical forest canopies: observation from eddy covariance tower and OCO-2 satellite in Northwest Himalaya, India.

Carbon dioxide (CO2) is the key atmospheric gas that controls the earth's greenhouse effect, and forests play a major role in abating the atmospheric CO2 by storing carbon as biomass. Therefore, it is vital to understand the role of different forests in regulating the spatiotemporal dynamics of atmospheric CO2 concentration. In this study, we have used eddy covariance (EC) tower-based atmospheric CO2 concentration measurements and satellite-retrieved column average CO2 concentration of 2018 to understand the diurnal and seasonal dynamics of atmospheric CO2 concentration over the sub-tropical forest in the foothills of northwest Himalaya, Uttarakhand, India. EC study revealed that the CO2 concentration over the forest canopy peaks during mid-night to early morning and drop to a minimum during the afternoon. On a monthly scale, peak atmospheric CO2 concentration was observed during July in both the sites, which was a result of more release of CO2 by the forest ecosystem through ecosystem respiration and microbial decomposition. Enhanced photosynthetic activities during the late monsoon and post-monsoon resulted in the decrease of atmospheric CO2 concentration over the forest ecosystem. Among the meteorological variables, rainfall was found to have the highest control over the seasonal variability of the atmospheric CO2 concentration. Orbiting Carbon Observatory-2 (OCO-2) satellite-retrieved column average CO2 (XCO2) was also examined to comprehend its reliability on an ecosystem scale. The OCO-2 retrieved XCO2 value was higher than the EC carbon flux tower-measured atmospheric CO2 concentration, which might be due to differences in the vertical resolution of the CO2 column and scale difference. However, the monthly atmospheric XCO2 retrieved from OCO-2 strongly adheres with the ground-measured monthly pattern. Our study highlights that forests with varying functional traits within the same climatic conditions show variability in the regulation of atmospheric CO2 concentration.

Monitoring and predicting regional land use and land cover changes in an estuarine landscape of India.

Deciphering land use and land cover (LULC) change patterns, identifying the variables that act as the major driving forces of change, and predicting possible changes are necessary tools of decision support for policymakers. Estuarine landscapes world over are under extreme pressure of developmental activities because of their resources. The developmental activities lead to unforeseen changes in the traditional land use practices, making it necessary for investigation of the possible outcomes. The present study aims to study the changing pattern of LULC in the East Godavari River Estuarine Ecosystem (EGREE) landscape during 1977-2015 using temporal satellite data and to predict the possible LULC changes by 2029. Cellular Automata-Markov model (CAMM) with and without the multi-criteria evaluator (MCE) and the multi-layer perceptron (MLP) models were used for future LULC prediction. Between 1977 and 2015, mangroves were converted to aquaculture (5.81 km2) on the landward side and were also lost to submergence at the seaward side (15 km2). All of the coastal scrub (69 km2) was lost to beach clearing. Over this period, the aquaculture area rose to 177 km2. The CAMM with MCE was found to yield better predictions. A further rise was predicted in aquaculture (16%), built-up (30%), and Casuarina plantations (28%) by 2029. The study highlighted the LULC change patterns in EGREE, an important estuarine landscape of India. The information generated in this study can act as baseline information for the stakeholders and policy makers in decision-making of developmental projects, land acquisition, and diversion of agricultural land to non-agricultural purposes.

Are phenological variations in natural teak (Tectona grandis) forests of India governed by rainfall? A remote sensing based investigation.

Monitoring and assessment of vegetation phenology at the regional to global scale are essential to understand the characteristics of various biophysical parameters in terrestrial ecosystems. Passive optical remote sensing data have been used extensively in the recent past to study phenology of vegetation, also called land surface phenology, at diverse landscapes across the globe. In the present study, the moderate resolution imaging spectroradiometer (MODIS)-derived enhanced vegetation index (EVI) time series data (2000-2013) was used to study the phenology of dry and moist teak (Tectona grandis) forests of different biogeographic provinces of India. Four phenology metrics, viz., start of season (SOS), end of season (EOS), peak of season (POS) and length of season (LOS) were derived using the TIMESAT tool. The SOSs' of dry and moist teak were found during July-August. LOS of moist teak was found to be much longer (~ 48 days) than dry teak. Also, a significant difference of leaf area index (LAI) (~ 2.8) of dry and moist teak forests was noticed during peak season from MODIS LAI product (MOD15A2). Vegetation phenology is greatly responsive to the fluctuation of climatic parameters such as rainfall. Hence, pre-season cumulative rainfall data were analysed to understand the control of rainfall over phenological variations in natural teak forests of India. It was noticed that rainfall was reasonably well correlated with SOS (R2 = 0.57-0.72) for both types of teak forests. The study highlighted the efficacy of time series MODIS EVI data to study the phenological variations in different teak forest types of India in a data-limited situation.

Assessment of historical forest cover loss and fragmentation in Asian elephant ranges in India.

India is home of the largest remaining population of the Asian elephant (Elephas maximus L.) in the South and Southeast Asia. The forest loss and fragmentation is the main threat to the long-term survival of Asian elephants. In the present study, we assessed forest loss and fragmentation in the major elephant ranging provinces in India, viz., north-eastern, north-western, central, and southern since the 1930s. We quantified forest cover changes by generating and analyzing forest cover maps of 1930, 1975, and 2013, whereas fragmentation of contiguous forest areas was quantified by applying landscape metrics on the temporal forest cover maps. A total of 21.49% of the original forest cover was lost from 1930 to 1975, while another 3.19% forest cover was lost from 1975 to 2013 in the elephant ranges in India. The maximum forest loss occurred in the southern range (13,084 km2) followed by north-eastern (10,188 km2), central (5614 km2), and north-western (4030 km2) elephant ranges in the past eight decades. The forests in the central range were the most fragmented followed by southern, north-eastern, and north-western elephant ranges. The forest fragmentation in the southern range occurred at the fastest rate than central, north-eastern, and north-western ranges. The core forest areas shrunk by 39.6% from 1930 to 2013. The causative factors of forest change and situation of elephant-human conflict have been discussed. Study outcomes would be helpful in planning effective conservation strategies for Asian elephants in India.

Does spatial heterogeneity of landscape explain the process of plant invasion? A case study of Hyptis suaveolens from Indian Western Himalaya.

Biological invasion is probably one of the most serious threats to biodiversity after climate change. Landscape distinguished by the heterogeneity of structure, forms, human interferences, and environmental settings plays an important role in the establishment and spread of invasive species. We investigated the effect of the spatial heterogeneity for a selected landscape upon the invasion process through a case study of Hyptis (Hyptis suaveolens) in the Indian Western Himalayan region. The selected study site constitutes a heterogeneous landscape of 32,300 ha in the state of Uttarakhand, placed at the lower elevation of the Indian Himalaya. The landscape has varying levels and patterns of Hyptis invasion. We quantified the spatial heterogeneity in terms of elevation; distance from the canal, river, road, and settlement; and 18 landscape metrics (at the patch and land use class level) to investigate their influence on the invasion; for this purpose, a logistic regression model was developed. The invasion of Hyptis was found to be governed by spatial heterogeneity. The highest probability of invasion was found in the areas adjacent to rivers and roads. The analysis at patch level revealed that the invasion is largely governed by the perimeter-area ratio of patches and is positively correlated. This suggests for greater invasion chances in smaller patches as compared with larger ones. The analysis for the land use class metrics indicated a higher influence of edge density expressed as total edge length of patches per unit area, followed by patch density expresses as a total number of patches per unit area. Hence, the landscapes with larger edges and more number of patches are supposed to be more prone to invasion risks. The results of the study can be used by forest managers in designing a landscape-level system to control invasion.