1 Introduction

Wetlands are the most valuable but vulnerable habitats on the Earth. Since from the beginning, man has utilized many direct and indirect benefits from wetland as it provides important function ranging from environmental, hydrological and socio-economic benefits to the local communities, such as, it is a direct source of water and other products like fishes and plant resources and also indirectly supports biodiversity, controls flood and purifies water etc. Therefore, the wetland habitat is considered as one of the most productive ecosystem on earth and miscellaneous habitat and even these are enriches continuously by nutrient affluent sediments by spilling of river [1, 2]. Although covering only 6% area of total earth surface, wetlands offer 40% of global ecosystem services [3]. The estimated value of the services provided by the flood plain wetlands is US$ 3920 × 109 yr−1 in the whole world [4]. Therefore, because of its extensive and valuable functions, wetlands are also viewed as a ‘biological supermarket’ [5].

Despite such important functions of wetland, uncontrolled and unsustainable human activities have made wetlands vulnerable [6]. From the historical periods of time, wetlands were viewed as a ‘waste land’ [7] due to its water logging condition and therefore, these are drained for agriculture, settlements and industries [8]. As a consequence about 50% of the world’s total wetland areas have been transformed [4].Agricultural extension, commercial developments, settlements construction [9,10,11,12,13], flow regulation [14, 15], groundwater lowering [16] are the established reason behind wetland loss and fragmentation. Because of the irreplaceable values of wetland, a vigilant monitoring and assessment of wetlands values are needed today for sustainable utilization and implementation of conservative measures. But, seasonal wetland with high irregularity in appearance is very difficult to delineate [9] and water presence frequency approach is very crucial in this regard [11] as it uses time series data layers.

India covers approximately 15.26 million ha (4.63%) of the total geographical area of the country [17] representing one of the most vital ecological and economic resources. Indo-Gangetic flood plain is the largest wetland system in India [18]. Like other countries, India is also facing the problems of continuous degradation and extinction of wetlands due to the reason mentioned above [18]. Diara, Barind and Tal region of India have also been experienced massive wetland transformation [1, 11, 15, 19].

Considering these issues, the present paper emphasized upon the identification of the wetlands, finding out the major reasons behind growing stress on wetland in Tangon river basin of Indo-Bangladesh within Barind tract.

2 Study area

Tangon river (267 km long), an Indo-Bangladesh trans-border river, is a major tributary of Mahananda river. Out of total length 125 km flows in Bangladesh and rest 142 km in India. Total basin area of this basin is 2049 km2 (to which 1243 km2 falls in India and rest part (806 km2) in Bangladesh) [19] (Fig. 1). Geologically the study region is come under four major geological units i.e. Shaugaon formation, Baikunthapur formation, Barind formation and Malda formation. Most of the wetlands are located within Baikunthapur and Barind formations. Neo-tectonic upliftment had started steady channel incision and it may cause disconnection of ox bow lakes, river scars located astride. Riparian flood corridor of this river is very distinct with good numbers of wetlands of different sizes. This semi-perennial river is mainly fed by the monsoonal rainfall (June–September). This region receives about 1500 mm rainfall annually and out of which 80% rainfall is received from monsoon months. Annual rainfall trend since 1901–2010 shows that there is insignificant change (y = − 0.1007x + 125.09, R2 = 0.0332) of annual total rainfall. It is important to mention that in upper reach of this river a barrage has been built up in 1989 (10 km west of Boda in Panchagarh district, Bangladesh), for irrigating about 5000 ha of land in dry season [20] (Fig. 1).

Fig. 1
figure 1

Location map of the study area a India, b West Bengal, c geological settings of Tangon river basin, d Barrage near Boda in Panchagarh district, Bangladesh, e confluence point of Tangon and Mahananda river at Aiho, Malda

Full size image

3 Materials

Total 32 Landsat images covering the study area are obtained from the US Geological Survey (USGS) Global Visualization Viewer. Out of the total images 18 are of pre monsoon season (March–April) and 14 are of post monsoon season (Oct–Nov). Path/row of the images is 139/42 and spatial resolution is 30 m. For analyzing river flow modification, day to day data since 1979–2016 are collected from Bamongola gauge station (under Irrigation & Waterways Dept. Govt. of WB). Rennell’s map (1776) is taken into consideration for ancient getting drainage pattern. Ground water level (GWL) data is collected from the Ground Water Year Book of West Bengal and Andaman and Nicobar Islands (2013–2014).

4 Methods

4.1 Wetland mapping and water presence frequency analysis

Wetland is delineated using Normalized Differences Water Index (NDWI) of McFeeters [21] from Landsat images. NDWI value above 0 indicates wetland and water body. For consistency mapping of the erratic wetlands in their appearance, wetland of different years are combined using water presence frequency (WPF) approach (Eq. 1) to get the nature of wetland consistency in appearance. This method is acceptable for such work [22] and subsequently applied by different scholars to analyzed wetland dynamics [11, 13, 19, 23].

$$WPF_{{s_{j} }} = \frac{{\sum\nolimits_{i = 1}^{n} {BPF_{ij} } }}{{N_{s} }}$$
(1)

where WPFsj is the water presence frequency value of the pixel j; BPFij is the sum of all the NDWI images counting pixel wise water presence frequency; Ns is the total number of NDWI images used for the study; WPFsj value ranges from 0 to 1; 1 means high water presence frequency and consistency.

Aggregated images have been reclassified into three equal frequency classes i.e. low frequency class (> 0.33 or 33%), moderate frequency class (0.33–0.66 or 33–66%) and high frequency class (> 0.66 or 66%). High frequency class indicates wetlands are consistent in their appearance vice versa in cases of low frequency class.

4.2 Methods for analyzing vectors of wetland change

For analyzing the nature of flow modification and its impact on wetland changes, monthly average maximum, minimum and average water level data have been plotted using simple line graph and percentage change is detected between pre and post barrage conditions. To see how Tangon river is detached from Tista river system, Rennell’s map (1776) and present map prepared from Google earth image is compared. Reclamation of wetland is assessed using Land use/land cover (LULC) change between pre and post barrage states. Supervised image classification technique is used for illustrating different LULCs. Kappa coefficient is calculated using 187 Google Earth based and 69 ground control points for both pre and post barrage periods to assess the accuracy level the classified images. Kappa coefficient > 0.8 indicates excellent agreement between map and ground reality.

5 Results and analysis

5.1 Wetland consistency in different seasons

Total identified wetland in pre and post monsoon seasons are respectively 112.74 and 168.85 km2 (Fig. 2a, b). From this information it is evident that 56.11 km2 wetland is seasonal in nature. Table 1 depicts the areal extent of wetland under different frequency classes. In post monsoon season high, moderate and low frequency wetland are respectively from 84.98, 44.02 and 39.85 km2. In pre monsoon season, due to scarcity of rainfall and external water supply from river and ground water a large part of the wetland is dried out and wetland area under high, moderate and low frequency wetland are respectively 98.69, 9.05 and 5 km2. Figure 2c displays that the spatial pattern of perennial and seasonal wetland.

Fig. 2
figure 2

Water presence frequency (WPF) zones of different seasons, a pre monsoon season and b post monsoon season, c wetland types based on hydro-duration

Full size image
Table 1 Season wise area of wetland under different frequency classes (1988–2016)
Full size table

5.2 Principal reasons behind wetland transformation

5.2.1 Impact of flow modification on wetland

Changes of the flow have significant impacts upon wetland modification as it is directly related with the dynamic nature of wetlands. Earlier it is mentioned that 1989, a barrage was constructed over the river and started water diversion. It causes attenuation of flow in barrage downstream river reaches. Figure 3a–c respectively shows the difference between pre and post barrage maximum, minimum and average water level. Seasonal flow reduction in all the cases range from 20 to 50% and it is highly disappointing. At present, river becomes completely dry during summer season due to complete arrest of water upstream. Attenuation of maximum water level indicates reduction of lateral flood limits of the river [15] and the wetland outside the present flood limit experiences water scarcity. Figure 3d, e depicts the yearly seasonal wetland area in between pre and post barrage periods. In post dam period, except few years with excessive rainfall, most of the cases, wetland area is reduced significantly. In post monsoon season, this change is very prominent. Figure 4a–d also shows the spatial consequences of water level change on wetland distribution and consistency. In post dam period, wetland area is reduced by 16.0% (77.78–67.05 km2) in post monsoon season and high frequency wetland area is attenuated by 21.57%.

Fig. 3
figure 3

Flow change in pre and post dam periods in connection with a maximum, b minimum and c average water level and year wise wetland area in pre and post dam periods in d pre monsoon and e post monsoon seasons

Full size image
Fig. 4
figure 4

Water presence frequency zones in different seasons in a pre dam, pre monsoon, b post dam, pre monsoon c pre dam, post monsoon and d post dam, post monsoon

Full size image

5.2.2 Impact of river disconnection

During 18th century, Tangon river was connected with Teesta river system through a distributary of Karatoa river (Fig. 5a). Huge water expectedly supplied from Teesta river system to Tangon river but delinking of river has reduced the flow availability of the later river (Fig. 5b). Geological history of this region provides some evidences. Different proofs on the neo-tectonic activities has already found in lower Gangetic plain region like incised channel, incised meander, sag ponds etc. [24,25,26]. From Due to upliftment of Barind tract and tilting of Teesta system toward south east some of the tributary channels of the Barind tract have lost their head knot with the master stream [26] and in consequence of this water supply to the major channels like Atreyee, Punarbhaba, Karatoa and Tangon have attenuated.

Fig. 5
figure 5

Beheading incidence of Tangon river from river Karatoa, one of the major distributaries of river Teesta a linkage exists between Tangon and Karatoa, b beheading of Tangon river

Full size image

5.2.3 Impact of changing land use/land cover pattern on wetland vulnerability

Figure 6a, b shows the spatial pattern of wetland land use/land cover in 1989 and 2010 representing pre and post dam periods. Water body is declined from 76.69 to 17.09 km2, agriculture land increased from 589.75 to 1402.96 km2 and built up area is inflated from 57.27 to 153.31 km2 (Table 2). These information highlight that wetland area is replaced by built up land, agriculture land etc. To show the reason of wetland reclamation, change matrix is calculated between two periods and displayed in Fig. 6c. It’s found that about 831.62 km2 area converted from agricultural land with seasonal fellow to agricultural land, about 502.84 km2 land transformed into agricultural land from scatter tree and shrub area, about 204.30 km2 water body and forest area converted into agriculture land with seasonal fellow and about 537.76 km2 area remain unchanged in between 1989 and 2010.

Fig. 6
figure 6

LULC pattern and change matrix of Tangon river basin in a 1989 and b 2010 of Post monsoon season, c land use/land cover change matrix between 1989 and 2010

Full size image
Table 2 Area under different land use/land cover in 1989 and 2010
Full size table

5.2.4 Impact of ground water level on wetland vulnerability

Ground water table above or very nearer to the bed level of the wetland helps a lot to make a wetland perennial and ecologically stable [12, 27]. But, lowering of ground water table disconnects such valuable vertical linkage. In the study area, only 8.05% wetlands falls under the water table depth < 3 and 71.03 and 20.43% of wetland areas are found within 3–6 and 6–9 m ground water depth zones respectively (Fig. 7, Table 3). Such statistics clearly displays that seasonal water level rise and seepage water are the major sources of water to the wetland. During monsoon season, saturation of soil profile supports areal extension of wetland area in this time.

Fig. 7
figure 7

Depth of ground water level (GWL) below to ground surface of Tangon river basin and wetland overlying the basin

Full size image
Table 3 Area and percentage of area under different depth of GWL below to ground surface
Full size table

6 Conclusion

The study shows the wetland state of Tangon river basin and the major reasons behind the growing vulnerability to such valuable natural capital. Total identified wetland in pre and post monsoon seasons are respectively 112.74 and 168.85 km2 in the study area. Wetland area is declined by 27.39% in between 1988 and 2010 in post monsoon season. Flow modification in river is emerged as the major reason behind identity crisis of the wetland habitat. After construction of dam over the river in 1989, flow level in river has attenuated from 22 to 17.5 m from mean sea level. Reduction of maximum flow level has decreased the lateral expansion of flood limit and as a consequence a good proportion of wetland is now out of flood water service. In post dam period, wetland area is reduced by 16.0% (77.78–67.05 km2) in post monsoon season and high frequency wetland area is attenuated by 21.57%. Historically, the river and riverine wetland have undergone into hydro-ecological modification. Disconnection of river Tangon from Teesta river system has attenuated flow availability in the river since long dates back. Agricultural extension and expansion of built up areas replacing wetland is one of the strong trend in the region and it causes death and deterioration of wetlands. Shallowing of wetland, conversion of high frequency water presence wetland to low frequency wetland has provided scope to the peasants to enter into the wetland and reclaim it for cultivation. People are mainly thinking about tangible direct return from the land. Therefore, in spite of having a good number of environmental benefits, wetlands are treated as waste land. If minimum effort is paid scientifically, not only ecological economic turnover will be far greater than same unit of wetland. Millennium Ecosystem Assessment report [28] documented that wetland has 2.5 times greater economic return than agriculture land.

Moreover the live existence of the flood plain is also largely influenced by the consistent presence of the wetland of different kinds. These are some good habitat of a good number of flood plain species. So, adoption of suitable strategy toward minimizing vulnerability is vital. For preserving and restoring the wetlands, ecological flow of the river should be computed and that amount of flow should be allowed to the river which may help to sustain the natural regime of the river and riparian zones. Gradual squeezing of flood plain extension has started conversion of para fluvial wetlands into ortho fluvial and ortho fluvial wetlands into only sub saturated moist land. Reclamation of wetland for any other activities should be stopped strictly. Effort in this regard should be deployed both from the government’s end and from the end of the community level. Justified land use extension toward wetland is highly required. For doing all these thing in effect ecological livelihood opportunity should be generated in few of the wetlands and people there form learn the approach of alternative uses and economic benefits from there.