Papers & Publications

Flood-related disasters have far-reaching impacts on infrastructure and societal well-being. Though characterizing flood susceptibilities using state-of-the-art approaches and modelling socio-economic exposure to highlight vulnerabilities is essential to assess and manage flood-associated risks, current studies are usually regional/coarser resolutions neglecting localized situations. Here we developed an integrated machine learning, artificial intelligence, and geospatial modelling-based framework for high-resolution flood susceptibility (30 m) and socio-economic exposure estimations at a larger scale using Pakistan as a case. To do so, the data on flooding, elevation, drainage, rainfall, Landsat-8 imagery, and gridded socio-economic layers were used. We produced the first national-scale high-resolution susceptibility maps for Pakistan, pinpointing areas at higher risk of flooding, and assessing the potential impact on the population and the economy. Our findings suggest that ∼29 % of the total area of Pakistan falls under critical flood susceptibility levels, with Sindh and Punjab being the most at-risk provinces. Notably, ∼95 million people (47 %) in Pakistan are exposed to high flood susceptibility with 74 % population of Sindh, 56 % of Punjab, and 33 % of Balochistan residing in high susceptibility areas. We further pinpoint economic hotspots in Sindh and upper Punjab as particularly vulnerable to flood risks, which calls for proactive disaster preparedness measures. Through the presented characterization of flood susceptibility and socio-economic exposure, our findings are useful to devise targeted interventions in highly exposed regions to enhance resilience and reduce the risks/impact of future floods. By addressing vulnerabilities and fostering resilience, Pakistan can effectively mitigate flood risks and safeguard its population and infrastructure.

Land use changes profoundly affect hydrological processes and water quality at various scales, necessitating a comprehensive understanding of sustainable water resource management. This paper investigates the implications of land use alterations in the Gap-Cheon watershed, analyzing data from 2012 and 2022 and predicting changes up to 2052 using the Future Land Use Simulation ({FLUS}) model. The study employs the Hydrological Simulation Program-{FORTRAN} ({HSPF}) model to assess water quantity and quality dynamics. Seven land use classes were identified, and their evolution was examined, revealing significant shifts in urban, agricultural, grassland, wetland, and forested areas. The model performance across observed data was evaluated using coefficient of determination (R2), percent bias ({PBAIS}), and mean absolute error ({MAE}). Results show the dynamic nature of land use changes, highlighting shifts in urbanization, agriculture, and forested areas. Notably, the study explores the consequences of these changes on water quantity and quality, scrutinizing surface runoff, evapotranspiration, stream flow, and nutrient loads. Urban green spaces emerge as key mitigators, regulating runoff and enhancing water absorption. Forests (vegetation) also play a crucial role in maintaining water balance, while wetlands act as natural filters for flood mitigation and water quality improvement. The findings underscore the importance of informed land use planning, recognizing urban green spaces, forests, and wetlands as integral components for sustainable watershed management. As society navigates environmental challenges, this research contributes to a deeper understanding of the complex interactions between human activities and the natural environment emphasizing the need for nature-based solutions in land use planning for resilient and balanced ecosystems.

Flood susceptibility mapping (FSM) is crucial for effective flood risk management, particularly in flood-prone regions like Pakistan. This study addresses the need for accurate and scalable FSM by systematically evaluating the performance of 14 machine learning (ML) models in high-risk areas of Pakistan. The novelty lies in the comprehensive comparison of these models and the use of explainable artificial intelligence (XAI) techniques. We employed XAI to identify significant conditioning factors for flood susceptibility at both the model training and prediction stages. The models were assessed for both accuracy and scalability, with specific focus on computational efficiency. Our findings indicate that LGBM and XGBoost are the top performers in terms of accuracy, with XGBoost also excelling in scalability, achieving a prediction time of ~18 s compared to LGBM's 22 s and random forest's 31 s. The evaluation framework presented is applicable to other flood-prone regions and highlights that LGBM is superior for accuracy-focused applications, while XGBoost is optimal for scenarios with computational constraints. The findings of this study can assist in accurate FSM in different regions and can also assist in scaling up the analysis to a larger geographical region which could assist in better decision-making and informed policy production for flood risk management.

While carbon sequestration is significant to achieve the net zero and plays an important role in climate change mitigation, urbanization-led land use land cover (LULC) changes are causing significant impacts on the carbon stocks of terrestrial ecosystems. Despite rapid urbanization in Pakistan, previous studies focus solely on localized areas without documenting the influence of urbanization on carbon stock. Hence, insights regarding the carbon storage dynamics (CSD) in response to LULC changes become essential to drive informed decisions and policies. In this context, we leverage high-resolution (30 m) remote sensing data to evaluate and map the grid-level spatial-temporal interactions between urbanization and CSD at the national scale in Pakistan during 1990–2020. To do so, multi-sensor earth observation data are retrieved and processed using the Google Earth Engine and the Integrated Valuation of Ecosystem Services and Tradeoffs models. Our findings indicate that urban areas have expanded exponentially (an increase of ∼1040%), resulting in reduced carbon storage (a decrease of ∼ − 5%). Major cities (e.g., Karachi, Lahore, Faisalabad) showed less urban sprawl while emerging cities (e.g., Rawalpindi and Peshawar) demonstrated higher urban sprawl, primarily due to shifting patterns from rangeland (∼47%) and agriculture (∼35%) to built-up class. Though some afforestation projects have increased forest carbon stocks in the northern region, there is a large north-south spatial heterogeneity in carbon storage loss across Pakistan. The presented high-resolution mapping of CSD over the past three decades advances our understanding of where and how much urbanization has influenced carbon sequestration, nationally. Considering the results, this study emphasizes the need for policies and management approaches that support sustainable urbanization, which does not compromise carbon pools in the country.

With the global upsurge in climatic extremes, disasters are causing more significant damages. While disaster risk management (DRM) is a serious global challenge, governments, stakeholders, and practitioners among many other actors seek advanced solutions to reduce disaster-related costs. Recently, Google Earth Engine (GEE), a cloud platform used for planetary-scale geospatial analysis using big-data, has gained popularity due to its applications in various fields. While the availability of free satellite data has facilitated long-term spatial-temporal trends and patterns identification, cloud computing emerged as a reputable tool in geo-big data analyses. Yet nearly after ∼15 years of its launch, the impact of such cloud-computing platform on DRM (risk assessment, monitoring, and planning) has not been carefully explored. Hence, a systematic review regarding the current state and trends in GEE applications to DRM is needed, which could provide the community with the bigger picture of the subject matter. Therefore, this study aims to investigate the advancement in DRM with GEE being the primary platform used. For this, 547 peer-reviewed studies published in 208 different journals during 2010–2022 were assessed. The current spectrum of GEE applications is dominated by floods, drought, and wildfires. For data type, most of the studies used optical data (Landsat and Sentinel-2). In terms of geographical distribution, China, USA, and India dominate with highest articles published. Within this research domain, three emerging research themes (floods, forest fire, and classification) are observed. Our results signify the emergence of GEE applications in DRM, which will continue making substantive progress on DRM-related multi-scale challenges.

With ~57% of the world population living in cities, the global urban population is increasing at an alarming rate, which further stimulates the urbanization process. Consequently, the increasing impervious surfaces in cities and associated variabilities in local/regional climatic characteristics pose several challenges to citizens (i.e., heat-related health issues, higher energy demands, and flooding among many others). Currently, cities contribute 75% of Green House Gases emissions, which is further worsening climate change impacts through global warming. Pakistan, the 6th most populated country globally, with ~220 million people, is among the top 10 most-affected nations vulnerable to climate change. Hence, studies addressing climate variability in local geographical regions have important implications to address the adverse urbanization-associated challenges, such as sustainability of the land resource and mitigating urban heat island (UHI) impacts in the context of climate change mitigation/adaptation. Due to temperature differences between urban, suburban, and rural areas, mapping city zones prone to the UHI effect is essential to provide actionable references. In connection with this, the present study analyses 15 megacities in Pakistan regarding their temperature variability in response to built-up area increment and highlights heat stress zones using the Urban Thermal Field Variance Index (UTFVI). The cloud-computing-based Google Earth Engine platform is employed to explore spatial–temporal variation in Land Surface Temperature (LST), which further leads to the identification of top-15 cities in terms of LST increase and the further evaluation of UTFVI for each city. The findings of this study suggest that the strongest UTFVI zones are concentrated around city-core areas, which are pure impervious surfaces with little or no green space. Moreover, in the last three decades (1990–2020), most of the weak and strong-strength UTFVI areas have been converted into the strongest strength primarily because of a rapid increase in the built-up areas. The findings of this study can help urban policymakers to identify priority intervention areas and design/implement strategies to counter the UTFVI and associated challenges. With proper land-use planning and on-time policy implementation, people residing in higher UTFVI zone areas can be safeguarded from noxious heatstroke-like health consequences along with mitigating and adapting to changing environmental conditions in cities.

Soil erosion triggered by water and wind pose a great threat to the sustainable development of Pakistan. In this study, a combination of geographic information systems (GISs) and machine learning approaches were used to predict soil water erosion rates. The Revised Wind Erosion Equation (RWEQ) model was used to evaluate soil wind erosion, map erosion factors, and analyze the soil erosion rates for each land use type. Finally, the maps of soil water and wind erosion were spatially integrated to identify erosion risk regions and recommend land use management in Pakistan. According to our estimates, the Potohar Plateau and its surrounding regions were mostly impacted by water erosion and have a soil erosion rate of 2500–5000 t·km−2·a−1; on the other hand, wind erosion predominated the Kharan Desert and the Thar Desert, with a soil erosion rate exceeding 15,000 t·km−2·a−1. The Sulaiman and Kirthar Mountain Ranges were susceptible to wind–water compound erosion, which was more than 8000 t·km−2·a−1. This study offers new perspectives on the geographic pattern of individual and integrated water–wind erosion threats in Pakistan and provides high-precision data and a scientific foundation for designing rational soil and water conservation practices.

Wetlands are crucial ecosystems as they enhance the quality of groundwater, protect from natural hazards, control erosion, and provide habitat to rare species of flora and fauna. Despite being valuable ecosystems, wetlands worldwide are decreasing in many regions, making mapping and monitoring of wetlands crucial. Large-scale wetlands mapping is challenging but recent advancements in machine learning, time series earth observation data, and cloud computing have opened doors to new techniques to overcome such limitations. Through evaluating the effectiveness of different classification methods, this study provides a brief analysis of wetlands dynamics in Sylhet, Bangladesh. The analysis is carried out between 1985 and 2022 using Google Earth Engine, Landsat imagery, and several spectral indices. To obtain reliable results, four classification algorithms (Random Forest, Minimum Distance, Classification and Regression Trees, and Support Vector Machine) are evaluated. As a result, Random Forest proved to be the most efficient and accurate for wetlands mapping by producing 99% accuracy across all periods. Change detection shows a rapid decrease in the wetlands in Sylhet, which could have serious consequences to the aquatic and terrestrial species, water and soil quality, and wildlife population, if not addressed. Between 1985 and 2022, nearly 45% of the wetlands have been lost in the region due to shifting land-use patterns, especially the conversion of wetlands into vegetative land (∼82,000 km2) as a result of increased agricultural practices in the region. Four critical regions (i.e., Derai, Sulla, Jamalganj, and Ajmiriganj) have undergone ∼80% reduction in wetlands, requiring prompt interventions for conservation and restoration of wetlands given their diverse services.

Since the last two decades, Pakistan has been often cited among the top ten countries most vulnerable to climate change and disasters, such as intense flooding, extreme heat, and droughts, among others. However, the unavailability of nationwide administrative-scale assessments from a space–time perspective hinders disaster resilience building in Pakistan. In this context, the key purpose of this study is to evaluate the spatial and temporal disparities in community disaster resilience (CDR) in Pakistan during 2004–2014—the period covering two of the most devastating disasters in Pakistan in recent history. Eventually, the dynamic nature of resilience is empirically demonstrated through the catastrophe progression method, and regions, where resilience increased/decreased, are identified using geo-information models, such as the Moran’s Index and the local indicators of Spatial Association (LISA). It is evident that CDR in the earlier, middle and final periods during 2004–2014 vary significantly (95% confidence). With inconsistent resilience distribution across Pakistan during 2004–2014, some noteworthy regional disparities are also found. For instance, while the overall lowest resilience is found for the areas in Balochistan province, the regions that became less resilient during the studied period are spread across Pakistan with notable concentrations in southern districts. Such place-based information is a crucial stepping-stone to initiating and formulating effective plans and resilience enhancement strategies in Pakistan. Furthermore, based on the pioneering analysis presented here, this study acts as a baseline for disaster resilience in Pakistan in terms of spatial–temporal heterogeneities along with pinpointing the significant areas for gradual or immediate attention—facilitating priority intervention areas.

In Pakistan, wetlands are of primary focus as they withstand the effects of floods, recharge groundwater, and provide several services in the context of economic, cultural, and climate mitigation aspects. However, the lack of field data and huge monitoring costs hinder their sustainable management in Pakistan. In connection with this, the current study leverages Google Earth Engine (GEE), earth observation data, and machine learning-based Random Forest (RF) algorithm to evaluate spatial-temporal heterogeneities in wetlands in Pakistan between 1990 and 2020. Additionally, the first high-resolution long-term inventory of wetlands in Pakistan is presented to provide a baseline. Our results ascertain an increase in wetlands areas over the last 30 years. The swamps’ area increased from 1391.19 km2 in 1990 to 8510.43 km2 in 2020 (2.62% annual change rate). Similarly, the marshes area increased between 1990 and 2020 with a ∼1.04% annual change rate. Conversely, the water area decreased from 8371.97 km2 in 1990 to 7818.34 km2 in 2020. The increase in wetlands could be associated with good conservation and planting practices in Pakistan. While these results provide important insights to implement conservation practices in the context of wetland sustainability, the resultant data is essential to the national wetlands inventory database for future evaluations.

Urbanization-led changes in land use land cover (LULC), resulting in an increased impervious surface, significantly deteriorate urban meteorological conditions compromising long-term sustainability. In this context, we leverage machine learning, spatial modelling, and cloud computing to explore and predict the changing patterns in urban growth and associated thermal characteristics in Bahawalpur, Pakistan. Using multi-source earth observations (1990–2020), the urban thermal field variance index (UTFVI) is estimated to evaluate the urban heat island effect quantitatively. From 1990 to 2020, the urban area increased by ~90% at the expense of vegetation and barren land, which will further grow by 2050 (50%), as determined by the artificial neural network-based prediction. The land surface temperature in the summer and winter seasons has experienced an increase of 0.88 °C and ~5 °C, respectively. While there exists spatial heterogeneity in the UTFVI 1990–2020, the city is expected to experience a ~140% increase in areas with severe UTFVI in response to predicted LULC change by 2050. The study provides essential information on LULC change and UTFVI and puts forth useful insights to advance our understanding of the urban climate, which can progressively help in designing more livable and sustainable cities in the face of environmental changes.

Timely and accurate estimation of rice-growing areas and forecasting of production can provide crucial information for governments, planners, and decision-makers in formulating policies. While there exists studies focusing on paddy rice mapping, only few have compared multi-scale datasets performance in rice classification. Furthermore, rice mapping of large geographical areas with sufficient accuracy for planning purposes has been a challenge in Pakistan, but recent advancements in Google Earth Engine make it possible to analyze spatial and temporal variations within these areas. The study was carried out over southern Punjab (Pakistan)-a region with 380,400 hectares devoted to rice production in year 2020. Previous studies support the individual capabilities of Sentinel-2, Landsat-8, and Moderate Resolution Imaging Spectroradiometer (MODIS) for paddy rice classification. However, to our knowledge, no study has compared the efficiencies of these three datasets in rice crop classification. Thus, this study primarily focuses on comparing these satellites’ data by estimating their potential in rice crop classification using accuracy assessment methods and area estimation. The overall accuracies were found to be 96% for Sentinel-2, 91.7% for Landsat-8, and 82.6% for MODIS. The F1-Scores for derived rice class were 83.8%, 75.5%, and 65.5% for Sentinel-2, Landsat-8, and MODIS, respectively. The rice estimated area corresponded relatively well with the crop statistics report provided by the Department of Agriculture, Punjab, with a mean percentage difference of less than 20% for Sentinel-2 and MODIS and 33% for Landsat-8. The outcomes of this study highlight three points; (a) Rice mapping accuracy improves with increase in spatial resolution, (b) Sentinel-2 efficiently differentiated individual farm level paddy fields while Landsat-8 was not able to do so, and lastly (c) Increase in rice cultivated area was observed using satellite images compared to the government provided statistics.

Gully erosion is a serious threat to the state of ecosystems all around the world. As a result, safeguarding the soil for our own benefit and from our own actions is a must for guaranteeing the longterm viability of a variety of ecosystem services. As a result, developing gully erosion susceptibility maps (GESM) is both suggested and necessary. In this study, we compared the effectiveness of three hybrid machine learning (ML) algorithms with the bivariate statistical index frequency ratio (FR), named random forest-frequency ratio (RF-FR), support vector machine-frequency ratio (SVM-FR), and naïve Bayes-frequency ratio (NB-FR), in mapping gully erosion in the GHISS watershed in the northern part of Morocco. The models were implemented based on the inventory mapping of a total number of 178 gully erosion points randomly divided into 2 groups (70% of points were used for training the models and 30% of points were used for the validation process), and 12 conditioning variables (i.e., elevation, slope, aspect, plane curvature, topographic moisture index (TWI), stream power index (SPI), precipitation, distance to road, distance to stream, drainage density, land use, and lithology). Using the equal interval reclassification method, the spatial distribution of gully erosion was categorized into five different classes, including very high, high, moderate, low, and very low. Our results showed that the very high susceptibility classes derived using RF-FR, SVM-FR, and NB-FR models covered 25.98%, 22.62%, and 27.10% of the total area, respectively. The area under the receiver (AUC) operating characteristic curve, precision, and accuracy were employed to evaluate the performance of these models. Based on the receiver operating characteristic (ROC), the results showed that the RF-FR achieved the best performance (AUC = 0.91), followed by SVM-FR (AUC = 0.87), and then NB-FR (AUC = 0.82), respectively. Our contribution, in line with the Sustainable Development Goals (SDGs), plays a crucial role for understanding and identifying the issue of “where and why” gully erosion occurs, and hence it can serve as a first pathway to reducing gully erosion in this particular area.

Phytohormones (PHs) play crucial role in regulation of various physiological and biochemical processes that govern plant growth and yield under optimal and stress conditions. The interaction of these PHs is crucial for plant survival under stressful environments as they trigger signaling pathways. Hormonal cross regulation initiate a cascade of reactions which finely tune the physiological processes in plant architecture that help plant to grow under suboptimal growth conditions. Recently, various studies have highlighted the role of PHs such as abscisic acid, salicylic acid, ethylene, and jasmonates in the plant responses toward environmental stresses. The involvement of cytokinins, gibberellins, auxin, and relatively novel PHs such as strigolactones and brassinosteroids in plant growth and development has been documented under normal and stress conditions. The recent identification of the first plant melatonin receptor opened the door to this regulatory molecule being considered a new plant hormone. However, polyamines, which are not considered PHs, have been included in this chapter. Various microbes produce and secrete hormones which helped the plants in nutrient uptake such as N, P, and Fe. Exogenous use of such microbes help plants in correcting nutrient deficiency under abiotic stresses. This chapter focused on the recent developments in the knowledge related to PHs and their involvement in abiotic stresses of anticipation, signaling, cross-talk, and activation of response mechanisms. In view of role of hormones and capability of microbes in producing hormones, we propose the use of hormones and microbes as potential strategy for crop stress management.

Monitoring spatial-temporal land use land cover (LULC) patterns and related processes (e.g., land surface temperature—LST) is essential to sustainable development at local, regional, and national levels. In this context, the present study leverages cloud-computing-based Google Earth Engine and geo-information modelling techniques to provide spatial-temporal insights regarding LULC and LST over the past three decades (1990–2020) in Pakistan—a south Asian country with ∼212 million people. Additionally, using Punjab province (the most populous and developed in Pakistan) as the study area, we empirically evaluate the association between several LULC types (i.e., built-up, forests, agriculture, rangeland, barren, and water) and LST. Our results show that due to the transition from rangeland and agriculture LULC to built-up areas (contributing 38 and 37%, respectively), ∼250% increase is observed in the impervious surface in Punjab during 1990–2020. While the rapid urbanization has resulted in ∼8.5 percent annual increase in built-up area during the study period, the highest percent change (∼10.5%) occurred during the most recent decade (i.e., 2010–2020). This increase in built-up areas has led to LST rise with 1.4 °C increase in maximum annual LST in Punjab. In addition, among the evaluated top-20 cities, the most significant rise in LST is observed by Kasur city followed by Chiniot, Sheikhupura, Sahiwal, and Lahore—areas known for industrial development in Pakistan. While the results on LULC provide important references for rational and optimal utilization of land resource via policy implications, the association between LULC and LST ascertains why it is critical to design sustainable LULC planning and management practices for climate change mitigation and adaptation.

Crop plants are affected by biotic and abiotic stresses (including salinity) and such stresses may affect the growth and yield of these crop plants seriously. High temperature (due to climate change) has also changed the pattern of precipitation and caused rise in sea level. These two factors have impacted soil salinization. To address such problems naturally, the crop plants adapt themselves by different mechanisms including changes in morphological, physiological, and biochemical processes. Both ions including sodium and chloride are the main ions, that become the reason for many physio-biochemical modulations inside plant tissues, in a similar way, chloride ion is the most dangerous because NaCl releases around 60% more ions in soil comparatively with Na2SO4. An extra amount of such types of salts increases the osmotic potential in soil matrix consequently the water absorbance by plants is reduced that leads towards physiological stresses or drought. This increase of Cl− relates to salt tolerance that is linked to plant growth, water use efficiency, and transpiration. Increasing salinity in the nutrient solution reduces growth directly and restricts leaf and root mineral fixing. In this chapter, we have discussed insights into various kinds of morphological, physiological, anatomical, and biochemical modulations in plants caused by abiotic stresses especially salinity. In the era of climate change, plant scientists should focus on each shotgun approaches as well as long-term genomic techniques to enhance salt tolerance in commercially important crops to ensure food security and sustainable productivity.

The rapid increase in urbanization has an important effect on cropping pattern and land use/land cover (LULC) through replacing areas of vegetation with commercial and residential coverage, thereby increasing the land surface temperature (LST). The LST information is significant to understand the environmental changes, urban climatology, anthropogenic activities, and ecological interactions, etc. Using remote sensing (RS) data, the present research provides a comprehensive study of LULC and LST changes in water scarce and climate prone Southern Punjab (Multan region), Pakistan, for 30 years (from 1990 to 2020). For this research, Landsat images were processed through supervised classification with maps of the Multan region. The LULC changes showed that sugarcane and rice (decreased by 2.9 and 1.6%, respectively) had less volatility of variation in comparison with both wheat and cotton (decreased by 5.3 and 6.6%, respectively). The analysis of normalized difference vegetation index (NDVI) showed that the vegetation decreased in the region both in minimum value (−0.05 [1990] to −0.15 [2020]) and maximum value (0.6 [1990] to 0.54 [2020]). The results showed that the built-up area was increased 3.5% during 1990–2020, and these were some of the major changes which increased the LST (from 27.6 to 28.5°C) in the study area. The significant regression in our study clearly shows that NDVI and LST are negatively correlated with each other. The results suggested that increasing temperature in growing period had a greatest effect on all types of vegetation. Crop-based classification aids water policy managers and analysts to make a better policy with enhanced information based on the extent of the natural resources. So, the study of dynamics in major crops and surface temperature through satellite RS can play an important role in the rural development and planning for food security in the study area.