Introduction

Forests ecosystems play a vital role in sequestering carbon from the atmosphere, which helps in mitigating carbon footprint and aid in moderating the global warming and consequent changes in the climate. Atmospheric carbon gets stored in the above- and below-ground biomass, dead organic matter and soil organic matter. Mismanagement of forests leading to deforestation and enhanced emissions during the post-industrial revolution have increased carbon dioxide concentration in the atmosphere to 400 ppm from 270 ppm (pre-industrial era). Forest ecosystems aid in capturing 45% of terrestrial carbon and are responsible for ˜ 50% net ecosystem production (McGarvey et al. 2015). Forests and soil play a vital role in the carbon cycle, evident from sequestration of about 30% of annual global anthropogenic CO₂emissions (2 petagrams (Pg) of carbon per year) from the atmosphere (Lal 2005; Achat et al. 2015) and hence are aptly regarded as moderators of climate (Bellassen and Luyssaert 2014). Quantification of carbon sequestered by forests across the globe was estimated as 861 ± 66 Pg C (Pan et al. 2011), with 383 ± 30 Pg C (44%) in soil (to 1-m depth), 363 ± 28 Pg C (42%) in live biomass (above and below-ground), 73 ± 6 Pg C (8%) in deadwood, and 43 ± 3 Pg C (5%) in litter, respectively. Soil carbon is the major pool of carbon terrestrial ecosystem, with major ecosystem services of advancing nutritional security, quality of water, improving biodiversity, and strengthening elemental recycling. The retention of carbon in soil depends on land-use, anthropogenic pressure, disturbance regimes and climate. The native and intact forests enhance mean residence time of carbon sinks with minimal re-emission (Lal et al. 2015). The landuse land-cover (LULC) changes in forest ecosystem have been altering forest structure with the increase in forest fragmentations (loss of contiguity), biodiversity loss, alteration in biogeochemical cycles and hydrological processes (Bharath et al. 2013; Vinay et al. 2013; Ramachandra et al. 2016; Armenteras et al. 2019; Ramachandra and Bharath 2019). Deforestation due to LULC changes is the prime causal parameters of global warming and increase in earth's ambient temperature due to carbon emissions and loss of ecosystem ability to sequester carbon.

Large-scale land-cover changes leading to deforestation with land degradation contribute to 20–25% of anthropogenic carbon emissions (Pachauri and Reisinger 2007) with the regional impacts on climate patterns altering the hydrological regime. The immediate effect of deforestation is the increase in carbon load in the atmosphere, and cumulative impacts are global warming, changes in the climate and ecosystem degradation. The loss of forest cover has modified local rainfall regime due to the changes in thermodynamic and mesoscale circulation processes (Lawrence and Vandecar 2015), with consequences of extreme weather conditions. Large and intact forest areas are responsible for transforming sensible heat to latent heat through the leaves; leaf area and forest canopy thereby an increase in dynamics of wind and increases precipitation events. Lower evapotranspiration with deforestation across the region has consequences of delay in the onset of the rainy season and decline in the number of rainy days with higher dry conditions (Debortoli et al. 2017). The delay in onset rainfall is positively correlated with forest cover change and extended length of the dry season (Funatsu et al. 2012). It will influence the adaptive capacity of populations and cause a shift in their diversity. Additionally, persistent drier conditions in a region can aggravate the probability of tropical evergreen forest to transform dry forest and savannah (Malhi et al. 2009). Plant metabolic activity and respiration are intensified with higher temperatures resulting in vegetation dieoff due to unavailability of water. Unfavorable conditions with loss of moisture releases the soil carbon escape to the atmosphere. The biophysical variations due to deforestation has significantly altered the microclimate conditions with rapidly increasing air and land surface temperature (Alkama and Cescatti 2016; Ramachandra et al. 2018). Higher temperature changes induce higher annual water demand resulting in the loss of canopy, microflora, microclimate alteration as well as higher incidences of forest fire.

Availability of the spatial data acquired at regular intervals through spaceborne sensors (remote sensing data) helps in analyzing LULC dynamics, deforestation, forest cover status and quantification of carbon stock. This helps in framing appropriate policies for sustainable management of forests and mitigation of degradation. Traditionally, forest inventories were based on ground data, which is time and labor intensive. Integration of geo-informatics with temporal remote sensing data and ground inventory data help in the cost-effective estimation of surface characteristics such as total biomass across various biomes (Gallaun et al. 2010;Rodrı´ guez-Veiga et al. 2019). Realizing the impacts of global warming with escalating greenhouse gas emissions, due to accelerated deforestation process necessitated measures toward adaptation and mitigation of effects of climate changes. In this regard, Kyoto Protocol was the first global initiative proposed at third Conference of Parties (COP) of the United Nations Framework Convention on Climate Change (UNFCCC) in 1997 to curb deforestation and promote forest conservation (Humphreys 2008). Reduced Emissions of Deforestation (RED) has emerged as an initiative for conservation in 2005 at 11th COP meeting to support developing countries. REDD + materialized at the 18th COP proposed to offer incentives for the conservation and enhancement of the forest carbon stock and the sustainable management of forests in 2012. It has been playing a significant role in forest conservation and helps in addressing challenges, supporting direct/indirect costs involved in forest management (Ghazoul et al. 2010). REDD + adaptation is a form of Payments for Ecosystem Services (PES) and, while providing economic benefits to the local communities, has improved natural resource management in developing countries (Agrawal et al. 2011). This necessitates a comprehensive understanding of the carbon stock and the dynamic drivers of LULC change, to devise effective policy measures to mitigate global deforestation.

The current study investigates landscape dynamics with climate trends and carbon sequestration potential in the ecologically fragile Western Ghats (WG). The investigation involved:
(i) analyses of land-use dynamics (using temporal remote sensing data), assessment of fragmentation of forests, modeling and visualization of land-use dynamics in the WG;
(ii) evaluation of rainfall and climate dynamics associated with deforestation;
(iii) quantification of carbon sequestration potential and productivity of forests; and
(iv) formulation of appropriate management strategies to arrest deforestation and ecological security of forests.