Temporal Dynamics of Vegetative Cover and Surface Water Bodies in the Indus Delta, Pakistan

Under the current scenario of diminishing Indus River flows and changing the climate, the Indus Delta, the world’s 5 th largest delta which has undergone rapid changes in water bodies and vegetative cover since last few decades, is under serious risk of losing its ecological functions. Assessing the temporal variation in vegetative cover and water bodies of the Indus Delta is essential for the future planning and ecosystem management in this region. The present study quantified the temporal patterns of the surface water bodies and vegetation cover, including crops, mangroves and other natural vegetation in the Indus Delta, by using field survey and remote sensing technique during the last 27 years. Results showed that the area covered by vegetation declined from 3002.35 km 2 (22.98% of the entire delta) to 2817.03 km 2 (21.56%) from 1990 to 2017, within which the area covered by mangrove forests declined from 1032.49 km 2 (7.90%) to 812.55 km 2 (6.22%). However, the area of water bodies increased from 1611.67 km 2 (12.39%) to 3007.15 km 2 (23.8%) in the same period. The reduction in freshwater flow to the delta, surface and subsurface seawater intrusion from the Arabian Sea and irrigation waters are the potential causes. The study would be helpful for policymakers to mitigate negative impacts and protect the ecosystem of the Indus Delta.


INTRODUCTION
T he surface of the earth is experiencing rapid changes in land-use/cover due to natural and anthropogenic activities [1]. Land use and land cover are two separate terms which are often used interchangeably [2][3][4]. The land cover defines the earth's physical features such as vegetation, water, soil, etc., while land use denotes to land utilized by the people for human activities [2]. Rapid growth in population, expansion in urbanization, advanced technologies are the drivers of land-use/cover changes around the globe [5].
Knowledge of accurate and updated information about the changing patterns of land-use/cover and natural resources is important for future planning and management [6].
Land-use/cover changes are one of the main causes of climate change through the modification of carbon, water, and energy cycles [7]. As a result, agricultural lands, vegetative lands, forests, water bodies, as well as mineral resource lands are changing continuously throughout the globe [8]. At present, analysis for land-use/cover changes has become one of the major concerns for researchers, policy makers for sustaining and controlling environmental changes around the world [9]. Chatterjee and Razuiddin [10] reported that mapping the variations in ecological features is essential for proper planning, monitoring, and management of resources. Alphan [11], Muttitanon and Tripathy [12] also reported that information about changes in land-use/cover is necessary to update land cover maps for effective management and planning of the resources for sustainable development.
The advanced development of the geographical information system (GIS) and remote sensing (RS) technologies help in the accurate measurement of such ecological resources [8]. Some studies [8,13] have stated the usefulness of GIS and RS techniques in land-use/ cover change detection. Digital change detection techniques based on multi-temporal and multi-spectral remotely sensed data have demonstrated great potential for understanding landscape dynamics-detect, identify, map, and monitor differences in land-use/cover patterns over time, irrespective of the causal factors [14].
Due to climate change, arid and semi-arid areas of the world are under serious threat in the context of water scarcity and land degradation [15]. Pakistan is said to be one of the most vulnerable countries to climate change [16]. The literature reveals that during the current century, rainfall patterns, and river flows will be reduced, sea levels will rise all over the world [15]. The projected sea level rise of 180 to 590 mm will directly affect the Indus River Delta, which is the world's 5 th largest delta [17] with an active area of about 0.6 million hectares [18]. Due to dam construction and increasing water demand for irrigation and industry, the amount of freshwater flow into the Indus Delta significantly decreased [19][20][21]. As a result, saline water from the Arabian Sea is intruding into the delta at an alarming rate [20]. Many fertile lands have already degraded, fresh groundwater resources are converted into brackish water, and other adverse impacts are also vivid on vegetation, mangrove, water bodies and socioeconomic conditions of the people living in the Indus delta [19][20].
The mangrove forests are the backbone of the delta's ecosystem, providing a breeding ground and food for various species of fish and shrimps that are a key part of Pakistan's fisheries exports [20]. Besides this, mangrove performs as a defense line against storms, cyclones, and Tsunamis; contribute firewood and fodder for livestock; provide nourishing livelihood to local people living in the Indus Delta.
The area under the vegetation, mangroves forests, as well as water bodies, is continuously changing due to natural as well as anthropogenic activities. Studies on the changes in different land uses are important for forest monitoring and in overall environmental monitoring [22].
Thus, the present study was conducted to quantify the temporal dynamics of the vegetation, mangroves, and water bodies in the Indus Delta for the last 27 years (1990 to 2017). The outcomes of the study would be helpful for policymakers and planners to mitigate adverse impacts of seawater intrusion and climate change to save the ecosystem of the Indus Delta.

Description of the Indus River Delta
The Indus River Delta stretches in southern districts viz.
Thatta and Sujawal of the Sindh province of Pakistan

Data sources
For assessment of the area under vegetation, mangroves, and surface water bodies, a geo-referenced field survey was carried out in the entire delta during 2016-2017 using the Garmin GPS MAP 62s system [33].

Vegetation and water indices
Different vegetation and water indices viz., normalized difference vegetation, ratio vegetation, normalized difference water, normalized differential pond indices were used in the present study. These indices are briefly described in Table 2.

Extraction of Area of Interest (AOI) and Its Classification
ArcGIS 10.3.1 software was used to process the acquired Landsat imagery of the delta. The "Extraction by Mask" tool was used to extract the study area (AOI) from the entire scene [36]. The AOI classified using maximum likelihood supervised and unsupervised classification.
The area was thus trained for three classes' viz.
vegetation, mangrove cover, and water bodies. The area under each created class was calculated by converting the raster data into polygons and then adding the area of all polygons of the same class [36]. However, the variation trends were determined using Pearson correlation analysis. Normalized difference water index (NDWI)=Green-Near Infrared/Green+Near Infrared For this index, values >0 represent the water bodies, while the values <0 or equal to 0 are assumed to represent the non-water bodies [34].
Ratio Vegetative Index (RVI)=Red/Near Infrared The index values usually vary from zero to infinity, however, values <1.0 represent a vegetated area, whereas >1. 0 represent the non-vegetated area.
Normalized differential pond index (NDPI)= Shortwave Infrared-Green/Shortwave Infrared+Green It is also used to estimate the area under water bodies [35]. Changes in climate pattern have resulted in dramatic changes in the vegetation [17]. The change in vegetation is not only due to land degradation, but the increase in temperature is also affecting it significantly [23].
Overall the area of the delta under vegetation is slightly

TABLE 4. STATISTICAL SUMMARY OF THE TEMPORAL DYNAMICS OF MANGROVES IN THE INDUS DELTA (1990-2017)
Likewise, vegetative cover, no specific trend in the temporal variation of mangroves was observed as it decreased gradually from 1990 to 2005, but later it increased significantly.

Quantification of Temporal Variation in the Area under Surface Water Bodies
Various methods such as supervised, unsupervised classifications, water indices as described earlier were used to quantify the temporal variation in the area under water bodies during the last 27 years. Fig. 5 shows the water bodies masks of the Indus Delta for different years, i.e., from 1990 to 2017.
The statistical data matrix of temporal dynamics of water bodies is summarized in Table 5, while its graphical representation is portrayed in Fig. 6. Alamgir et al. [17] also reported that seawater is enhanced in the Indus delta due to the reduced flow in river Indus that is affecting the deltaic system.

Implications of the changes in vegetation and seawter intrusion
Due to seawater intrusion, and variation in the vegetation, mangroves, and surface water bodies in the Indus delta, shoreline erosion is increased, many fertile lands are degraded, the area under vegetative cover as well the production/yield is decreased significantly. That has ultimately affected the ecosystem and livelihood of the people living in the delta. Several families have migrated from their ancestral towns to the safe places of surrounded cities/towns in search of food and shelter. Mahar [38] also reported that due to seawater intrusion into the Indus delta, people are evacuating the populated areas, per acreage crop yield of agricultural lands is decreasing continuously, and socio-economic conditions of the people living in the deltaic areas are badly affected.

River flow below Kotri Barrage
The

CONCLUSIONS
In the present study, field and satellite data were used to United States Geological Survey is also highly acknowledged for providing multi-temporal satellite imagery for the study.