Climate change in Tram Chim National Park
The climate of TCNP exhibits a significant increase in temperature and this change has occurred markedly since 20 years ago. During 42 years from 1978 to 2019, regarding temperature, there is a significant increase in the average temperature of year and in January (p value <0.05) (table 2), meanwhile in April, the temperature increases but not significant (p value >0.05). The changing point is in 1999 for yearly temperature, in 2002 for April temperature and in 1999 for January temperature.
Table 2. Results calculated from Mann-Kendall method for temperature characteristics at Caolanh station (Dong Thap) from 1978-2019
|
By year
|
April (the hottest month)
|
Jan (the coldest month)
|
S †
|
Z ‡
|
Yt §
|
P.T |
|
S †
|
Z ‡
|
Yt §
|
P.T |
|
S †
|
Z ‡
|
Yt §
|
P.T |
|
Tave
|
4.75
|
1.96E-06
|
0.021
|
23
|
1.38
|
0.16
|
0.009
|
24
|
9.32
|
1.16E-20
|
1
|
21
|
Tmax
|
3.69
|
0.0002
|
0.03
|
12
|
2.73
|
0.006
|
0.026
|
12
|
2.73
|
0.006
|
0.026
|
12
|
Tmin
|
-3.89
|
9.84E-05
|
-0.08
|
14
|
-2.02
|
0.04
|
-0.012
|
12
|
-2.81
|
0.0048
|
-0.072
|
14
|
†S - Mann-Kendall S - positive S value means an uptrend; a negative S value means a downtrend
‡ Z - standard normal deviate – p value
§ Yt - trend slope
| P.T – changing point
In the whole year, temperature increase (with S>0 in table 2) in all month with significant increase (p value <0.05) occurring in January, March, June, July, August, September, October, November and December.
Regarding rainfall, the rainfall slightly increases by year, in October, February and in dry season (S>0) but not significant (p value all >0.05). Only in rainy season, the rainfall has slight decrease trend (S<0) but not significant (p value = 0.9) (table 3).
Table 3. Results calculated from Mann-Kendall method for precipitation characteristics at Caolanh station from 1978-2019
|
S
|
Z
|
Yt
|
P.T
|
Annual amount
|
0.44
|
0.65
|
1.50
|
17
|
Max month (Oct)
|
1.34
|
0.17
|
2.53
|
17
|
Min month (Feb)
|
1.09
|
0.27
|
0
|
19
|
Rainy season
|
-0.10
|
0.91
|
-0.44
|
36
|
Dry season
|
1.80
|
0.07
|
2.61
|
21
|
Thus, the change of climate is manifested in the increase of the mean annual temperature since 2001 and the annual temperature range since the early 1990s with the increase of maximum temperature and decline of minimum temperature. Meanwhile, the precipitation shows an insignificant increase. That leads to the risk of a large increase in evapotranspiration and potentially reduced surface water, increasing the risk of hydrological drought and forest fires for the region. The facts have shown that natural disasters also change over time according to the change of climate parameters. In the period after its establishment from 1998 to 2001, flooding was considered a risk for this area. However, after 2003 to present, droughts and forest fires in the dry season are considered to be major risks of the region (figure 4). In addition, drought years usually accompany with forest fire such as in 1998, 2010, 2016.
Interactions between external ecosystems and resource system of WPA
Positive response of vegetation to climate change
The interactions of vegetation cover and climate change are analyzed by NDVI dynamics by seasons and years. First of all, we find that there are many variations of vegetation corresponding to changes in rainfall between the dry and rainy seasons. The result of Landsat-based detection showed that NDVI has a difference between dry and rainy seasons. Specifically, during the rainy season, in all surveyed years, we found that in the wet season average NDVI (0. 152) is higher than that of the dry season (0.14). In the wet season, the landscape looks more diverse and vivid in all zones (figure 5). Specifically, Melaleuca is the vegetation that thrives in the rainy season, with the NDVI being mainly at medium and high, while in the dry season being mainly at low NDVI (table 4). Some marsh grasses like Is. Indicum, Ischeamum, Leersia sp - O. rufipogon, Panicum repens, Oryza rufipogon also thrive in the rainy season with medium and high NDVI value. Polygonum hvdropiper L in the dry season only concentrated in A1 zone, but in the rainy season, it developed in both A1 and A2 zones. Aquatic plants like Nelumbium nelumbo grow more widely in A1 in the rainy season. Figure 5 shows that the area with NDVI value < 0 shrinking in the rainy season, which results from the rapid growth of aquatic plants (floating and emergent plants) on open water bodies. Whereas, in dry season, low NDVI dominates all zones of TCNP. The seasonal change of vegetation reflects the essential role of rainfall in the development and transformation of seasonal vegetation. In the wet season with major proportion of rainfall which provide abundant water for vegetation grow lush and healthy and even cover water surface of canals, ponds, etc.
Table 4. Spatial distribution and NDVI classification of vegetation cover at TCNP
NDVI classification
|
Tram Chim National Park
|
|
Dry season (Feb/2014)
|
Wet season (Sept/2014)
|
|
Class 1 (<0)
|
Open water bodies (ponds) and water canals without aquatic plants
|
|
Class 2 (Low_NDVI vegetation)
(0 to 0.2)
|
Name of vegetation
|
Zone
|
Name of vegetation
|
Zone
|
Melaleuca forest
|
A1, A2, A3, A5
|
Eleocharis dulcis
|
A1, A3, A4, A5
|
Oryza rufipogon
|
A1, A2
|
E. atropurpurea
|
A1, A5
|
Leersia sp - O. rufipogon
|
A1
|
Xyris indica L
|
A4, A5
|
Panicum repens
|
A1, A2, A3, A4, A5
|
Oryza rufipogon
|
A1
|
E. atropurpurea
|
A1, A5
|
Panicum repens
|
A1
|
Eleocharis dulcis
|
A1, A2, A3, A4, A5
|
|
Is. Indicum
|
A4
|
Ischeamum
|
A1
|
Xyris indica L
|
A4, A5
|
|
Class 3 (Medium_NDVI vegetation)
(0.2 to 0.4)
|
Melaleuca forest
|
A1, A4, A5
|
Melaleuca forest
|
A1, A2, A3, A4, A5
|
Eleocharis dulcis
|
A1
|
Oryza rufipogon
|
A1, A5
|
Polygonum hvdropiper L
|
A1
|
Eleocharis dulcis
|
A1, A5
|
Nelumbium nelumbo
|
A1
|
Panicum repens
|
A1, A2, A3, A5
|
|
Is. Indicum
|
A4
|
Ischeamum
|
A1
|
Polygonum hvdropiper L
|
A1, A2
|
Nelumbium nelumbo
|
A1
|
|
Class 4 (High_NDVI vegetation) (>0.4)
|
Polygonum hvdropiper L
|
A1
|
Melaleuca
|
A1, A2, A5
|
Nelumbium nelumbo
|
A1
|
Nelumbium nelumbo
|
A1
|
Oryza rufipogon
|
A1
|
|
Leersia sp - O. rufipogon
|
A1
|
Eleocharis dulcis
|
A1
|
Panicum repens
|
A1, A3
|
Polygonum hvdropiper L
|
A1, A2
|
Secondly, over years, recovery of vegetation was found mainly in rainy season with an increase of medium_NDVI vegetation areas (figure 6). In addition, a transformation from low_NDVI to medium_NDVI vegetations was found scatteredly in some marginal areas of the Melaleuca forest and from low_NDVI and medium_NDVI to high_NDVI vegetation in the boundary of zone A1 and A5; and A2 and A4. Most of open water spaces in zone A5 and southwest of Zone A4 was transformed to medium_NDVI vegetation (dark blue) and low_NDVI vegetation (light purple) (figure 6b). This is also a sign of the recovery of Xyris indica L and Eleocharis Dulcis in these areas. Water body was shrunk significantly in 2014. However, water gradually increased in 2018 especially in the west of zone A1 which is also area of aquatic plants like Nelumbium nelumbo and Polygonum hvdropiper L.
The transformation of the dominant vegetation communities of TCNP also changes over time, from 2009 to 2016, the statistical results show that, in 2009 and 2013 (Ni & Tuan, 2015), the six dominant biomes were Melaleuca, Panicum, Eleocharis, Ischeamum, Nelumbium nelumbo, O. rufipogon. Whereas, in 2016, Ischeamum vegetation had gradually shrunk and was replaced by Polygonum tomentosum occupying with a significant area (30,509) (table 5). The biological succession of this area dominates with a transition from upland community like grass to lowland community like emergent and aquatic plants. It can be speculated to be due to the increase of the water level of the TCNP especially in the dry season in recent years.
Table 5. Area changes of the major plant community at TCNP in 2009, 2013 and 2016
|
|
Area of plant communities by years
|
Trend of change from 2009 to 2016
(+ increase; - decrease)
|
No.
|
Plant communities
|
2009
|
2013
|
2016
|
1
|
Melaleuca
|
1901
|
2211
|
2435.7603
|
+534.7603
|
2
|
Panicum
|
451
|
269
|
617.1078
|
+166.1078
|
3
|
Eleocharis
|
1109
|
651
|
820.8972
|
-288.1028
|
4
|
Ischeamum
|
26
|
6
|
Polygonum tomentosum (30.509 ha)
|
|
5
|
Nelumbium nelumbo
|
221
|
85
|
327.29
|
+106.29
|
6
|
O. rufipogon
|
27
|
37
|
38.154
|
+11.154
|
7
|
Mix-up
|
3853
|
4329
|
3073.8
|
-779.2
|
Thirdly, the increase of temperature in the whole year especially in dry season can be a culprit of extreme events like droughts, forest fires in recent years. Based on the disaster map which is made by local community and park's officers (figure 7), the zone A1 suffered most natural disaster events such as floods in 1996, 2000, 2004, 2011; forest fires in 1996, 2006, 2010, 2014, 2016 as well as droughts in 2008, 2010, 2016. This partly explains the drastic change of NDVI dynamics especially in grassland areas such as Eleocharis dulcis and Ischaemum which have significantly decreased.
Negative response of vegetation to climate change
Although the water amount in dry season from rainfall has significant increase, degradation trend still dominates in vegetation cover in dry season especially after 2008. The change is more obvious with a decrease of medium_NDVI vegetation including part of Melaleuca forest; Eleocharis dulcis; mainly distributing in the A1 zone. Instead, after 2014, there is an increase of low_NDVI vegetation in all zones A2, A3, A4, A5, the low_NDVI vegetation increase in the area with diverse vegetation like Melaleuca forest; Oryza rufipogon; Leersia sp - O. rufipogon; Panicum; E. Atropurpurea; Eleocharis dulcis; Is. Indicum; Ischeamum; Xyris indica L (figure 6). In the whole period from 2002 to 2020, a large area of medium_NDVI vegetation has transformed to low_NDVI vegetation (light green color) is dominated in all zones (figure 6a).
In brief, through the large-scale decline of NDVI in TCNP, we argue that climate change has manifested a small impact on a few areas by increasing the risk of forest fires which has been recorded to occur on a small scale in TCNP. Other causes that have a greater impact on the large-scale decline of vegetation in the TCNP may stem from human management and other external ecosystems such as water level, the amount of sediment, etc.
Regarding factors influencing to landscape changes of VMR, beside on climate change, hydrological regime changes of the Mekong river as well as intensive rice production are mentioned as main factors (ICEM 2010; Nguyen et al. 2017; Le et al. 2015). The hydrological regime of the Mekong River in general and its two tributaries in Southern Vietnam, Tien and Hau rivers, have two distinct dry seasons and flood seasons. The dry season usually lasts from December to May of the following year, and the flood season is usually from June to November. However, in the past two decades, the change of water resources from the Mekong River into the Vietnamese Mekong Delta is unstable and unpredictable in each part of Mekong River Basin. Research of Lu and Chua 2021 stated the water flows from upstream Mekong river to Vietnamese Mekong Delta reduced significantly after 2010 since hydropower dams have significant impacts on increasing droughts not only in dry season but also droughts in wet season. The 2019-2020 drought is thus both a natural hazard compounded with manmade vulnerabilities. However, our data during the period 2001-2016 shows that the water flow from the Mekong River through Tan Chau station which is located at the exit of water flows from Mekong river to Southern Vietnam tended to increase significantly in the dry season (from December to May (next year)) with p<0.05, while the flood water level from 1978- 2020 tends to decrease with p<0.05 (table 6).
Table 6. Results calculated from Mann-Kendall method for water flow characteristics at Tan Chau station (An Giang)
|
S
|
Z
|
Yt
|
P.T
|
Q_value in dry season (from 2001-2016)
|
4.11
|
4.008352e-05
|
131.95
|
1 (2002)
|
Flood water level (from 1978-2020)
|
4.11
|
4.008352e-05
|
-0.87
|
1 (1979)
|
The increasing trend in water discharge at Tan Chau station is similar to the change in the rainfall regime in dry season. In previous studies, that trend can have a positive impact on vegetation especially in the context of prolonged drought and water shortage is serious (Salimi et al. 2021). However, our above analysis has shown that NDVI of TCNP has even declined by years. Therefore, we argued that the decline of NDVI and the transition from upland community to lowland community may be the result of human factors.
Relationships between resource system, external ecosystem and governance system
Human intervention in the natural system surrounding the TCNP has long history since after 1960-1970s a large network of canals was excavated across the Plain of Reeds to lower wetland water tables and also impacted to average flooding in the depressions decreased from 12 months to 4-6 months (Hanhart and Ni. 1993). From 1990-1996, under pressure of population increase and rice production, a large reclamation on the Plain of Reeds required digging canals and drainage to remove acidity from soils. However, it caused several environmental issues like soil acidification and loss of functioning wetland ecosystem (Ni et al. 2006; Ni and Tuan. 2015, Ni 2003). In 1994, Tram Chim Wetland Protected Area was established and then was upgraded to Tram Chim National Park in 1998 with initial purpose to protect Sarus Crane. Since then to now, there are several regulations and operations implemented, which modified TCNP environment (table 7).
Table 7. Timeline of remarked regulations/actions implemented and its impact zones in TCNP from 1994-present
Time
|
Actions/regulations implemented
|
Purpose
|
Impacted zones
|
1994-1996
|
Infrastructure development
|
Create the boundary for the national park
|
A1, A3, A2
|
Water level regulation (through gates and boundary canals)
|
Regulate water level
Forest fire control
|
Buffer zone socio-economic development and research
|
Reduce penetration of people into core zones
|
Buffer zone
|
1995–2004
|
Keeping high water levels inside Tram Chim
|
Prevent forest fires
|
A1, A2, A3, A4, A5
|
2000
|
Complete 60 km of dike around TCNP
|
Store water
Forest fire control
Recover Eleocharis Dulcis to attract Sarus crane
|
A1, A2, A3, A4, A5
|
2001
|
Prescribed fire in some grass areas to decrease the risk of uncontrolled fire by reducing fuel loading
|
Water level has been regulated according to plan: dikes, water gates, protection stations, digging ponds, boundary poles, fire controls
|
2005-2008
|
A comprehensive fire and water management plan was developed and tested at Tram Chim by a team of international experts
|
|
A1, A2, A3, A4, A5
|
2008
|
A sharing benefits of fuelwoods towards wetland resource management has been applied in Tram Chim
|
Reduce illegal penetration of people into core zones
Local people can harvest grass and wooden fuel at some time to reduce accumulated fire fuels
|
A1, A2, A3, A4, A5
|
2011 to present
|
The plan was approved by Vietnamese
authorities to be applied permanently at Tram Chim
|
|
Buffer zone and core zones
|
(Source: Referred from Meynell et al. 2012; Torell et al. 2003)
The relationship of the related ecosystems (climate change) and resource systems (vegetation) and governance system (top-down law system and practical implementation) is clearly expressed through the process of achieving Ecological performance (a part of Outcome in SES) (figure 1). In the case of TCNP, the Ecological performance is most evident in the three management goals of TCNP: preventing forest fires, increasing the area of the native species Eleocharis Dulcis, and restoring the ecosystem to attract the iconic bird of TCNP (Sarus crane) (Ramsar Information Sheet (RIS), TCNP, 2012). Based on the synthesis of disaster prevention reports of Dong Thap province from 2010 to 2020 (PCDTP from 2010 to 2020) climate change is perceived as a risk which often used to explain for failures in water management of TCNP especially after 2014, but the decrease of water flow from the Mekong river is perceived as the main drivers for the shortage of water in NP in the dry season and also for strategy of water retention. This makes water and forest fire management is identified as the key adaptation activity of governance systems (table 7). Specifically, the disaster map showed that among all extreme events were mentioned by local people and officers, the historical flood in 2000 was mostly mentioned. That can be a focusing event (Birkland 1998) for boosting the implementation of a dyke system around TCNP after 2000, which also plays a key in vegetation transformation of TCNP. A dike system surrounding national parks with a length of up to 60km was constructed in 2000 to regulate water through a system of sluices and spillways located at the surrounding dike (table 7). Since 2000 to present, to reduce the risk of fire in the dry season, the water level inside the national park is always kept higher than the conditions in the past. In addition, to store water for firefighting, old ponds inside the park were dug and extended (table 7). Thus, almost all areas are affected by human factors through water level adjustments, especially zone A1, A2, A3. That adjustment has narrowed the water tables in flood season and dry season (figure 11) and zones of A1, A2, A3 have become too wet for many vegetation species compared to the optimum conditions (Ni et al. 2006). That explains the transformation of upland communities to lowland communities in these zones (figure 8).
However, the decline of NDVI in all zones (figure 6) and native species in TCNP present a decrease in the health of the TCNP ecosystem and a failure of TCNP in ecological performance. Ecological succession with the expansion of lowland vegetation in contrast to the changing climate patterns reveals a significant intervention of human actors into natural laws. A tendency to cope rather than adapt the climate change and a human-based solution instead of a nature-based solution can be noticed in WPA management of the Vietnamese Mekong Delta.