Spatial Mapping of Carbon Stock in Riverine Mangroves Along Amanzule River in the Ellembelle District of Ghana

Compared to other wetland ecosystems mangroves are well known for their numerous ecosystem services, especially carbon pool. In Ghana, there is limited information on the sequestered carbon in mangroves. There is increasing interest on national climate change mitigation and adaptation plans in mangroves in developing nations, and Ellembelle in the Western Region of Ghana is of no exception. Ellembelle is one of the areas with little information on the size and variation of mangrove carbon stock which needs to be addressed. This research is aimed at determining the carbon stock from the carbon sequestered in mangrove and the areal extent in mangrove forest using remote sensing and allometric equation. The ecosystem carbon density estimate for the mangrove forest was weighted based on their spatial distribution across the landscape to yield a total carbon stock of for the Ellembelle mangrove forest. The error obtained from the 95% Confidence Interval was + 1.53%, which is within the acceptable levels of uncertainty based on the Monte Carlo Analysis. The overall carbon estimated for 2015 based on the area for mangrove (374.49 ha) was 1.550Mt with an uncertainty of +57.125Kt indicating a high amount of carbon sequestered in mangroves.


Introduction
Mangrove forests are considered as one of the world's most productive ecosystem (Kathiresan & Qasim, 2005).Mangrove tree have unique adaptation to the severe conditions of coastal environments.Research shows that mangrove forests are rated as one of the carbon richest forest in the forest.Mangroves swamp with its wealth in stored carbon provides a potential sink for atmospheric carbon.If mangroves are not well catered for, they may become the sources of Green House Gases (GHG) in the likes of carbon and methane.It is therefore necessary to know the amount of carbon sequestered in mangrove.
According to Donato et al. (2011), mangrove forest in the tropics contains an average of 1023 tons carbon per ha.According to Donato et al. mangroves forests in the tropics, contains an average of 1,023 tons carbon per ha.In the case of deforestation mangroves are recorded to give out about 0.02 -0.12 Pg Carbon per year, while storing up to about 20 Pg C every year.Mukherjee, 2007 indicates that mangrove forest are practically highest sequesters of carbon and their ability to sequester carbon reduces as they reach maturity.Compared to other forest wetland ecosystems, salt marshes, mangroves and sea grass beds, can store large amount of carbon.This is possible for two main reasons: (1) Plants usually grow a lot each year, and for that reason a large amount of Carbon Dioxide (CO 2 ) is sequestered; and (2) the soils are without oxygen so carbon that gets incorporated into the soils decomposes quite slowly and can persist for hundreds or even thousands of years (NOAA, 2016).
Mangroves absorbs more carbon than they emit.Knowing the amount of carbon sequestered is usually done by measuring directly on the field, the biomass of dried plant species.There are many conventional methods that are used for quantification of stored carbon.Most of these methods are labour and cost intensive in terms of the coverage.These limitations hinder comprehensive calculation and monitoring of carbon.Appropriate and cost effective methods are required to reduce the laborious method of manually calculating for the amount of carbon sequestered.There should be a good but cost effective means of determining the amount of carbon sequestered.Remote Sensing (RS) is noted for giving a good classification of mangroves.Therefore, integrating RS and Geographic Information Systems (GIS) will be an option in this regard.According to Sellers et al. (1995) and Bastiannssen et al. (1998), RS have been used for the estimation of plant biomass.RS approach can be used for

(Source En
In measuri 1.3m was single tree slightly ab

Equati
The AGB specie Rhi Forestry C Table 1.Eq Living Bio    Total carbon stock of the study area was estimated by finding the product of the carbon stock and the area of the study area.This gave a total of 1,550,295 tons with and uncertainty of +57,125 tons.The results are shown in the Table 2. Table 3 show the output for the parameters that gave the total carbon stock estimated for the mangrove area of in 2015.The area covered by mangrove was obtained from the classified image of 2015.The study revealed that the carbon sequestered in each plot is relatively high, depending on the biomass sequestered at the plot.The carbon calculated for each plot is based on the AGB and the BGB, the dead downed wood and the dead standing.These do not give a comprehensive sum total of the carbon pool as the soil and litter were not considered a part of the carbon pool in this study.Most of the sampled plot had no juvenile plant recorded below 1cm DBH except for plots A and Plot H.The presence of dead standing is recorded only in Plot K and dead downed wood were recorded in Plots B, G, H and I.The absence of the dead downed wood and the dead standing wood can be attributed to the sensitization given to the people to make use of the dead downed and the dead standing biomass for firewood.It is therefore evident that with time there will be no dead downed woods and as such the living biomass will be the next option.

Standing
Plot G recorded the least AGC.This is due to less mangrove tree within the plot.The population of mangrove in plot K is more, giving an indication of high carbon in plot K. Plot K has rich mangrove stand with dense canopy.
Mangrove in this section is bigger with average of 10cm DBH.Comparing Plot K to the other plots, it observed less disturbance as there were presence of dead standing wood.This is because it is farther from the river.These contributed to the high Carbon obtained in Plot K. Again, it was observed that mangrove with DBH greater than 10cm sequesters more carbon.
The juvenile plants dead downed and the dead standing plants resulted in a higher uncertainty since the data collected per the plot size were few.Due to the large uncertainty, it was inappropriate to use a simple propagation of error to determine uncertainty of the carbon hence the use of the Monte Carlo simulation to normalize the uncertainty (Goslee et al., 2010).The summation of all the uncertainty of the carbon pool was used to simulate and determine the 95%CI of the data which was +1.53%.The common choice for confidence level was 95% and the level corresponded to percentage of area of the normal curve and the probability of observing a value outside the area is less than 0.05 (Goetz et al., 2009).This was because the normal curve obtained was symmetrical and half of the area was in the right side of the curve and the other half was in the left side of the curve.The CI gave an estimated range of values showing that the probability that the CI will contain the true parameter value for carbon, falls within the 95% CI.

Conclusion
The total area occupied by mangrove in 2015 was 374.49ha therefore gave an estimate of a total carbon stock of 1,550,294.566tons (1.55 Mt) and the equivalent carbon emission was 5,689,581.057tons CO2e (5.690Mt CO2e).
The uncertainty of the estimated carbon stock falls within +57,125.4tons(57Kt).From the results obtained, more sample plots and a complete assessment of the remaining carbon pools such as the litter and the soil will improve upon the estimated sequestered carbon.Nonetheless the research shows that the mangrove forest in the Ellembelle District sequesters a large amount of carbon and the availability of mangrove biomass carbon is helpful for the supervision of activities and also for the resilience of mangrove to changing environment.Deforestation of mangrove will however lead to loss in mangrove and loss of a good carbon sink.

Recommendation
Further research should be carried out on the BGB of the root, soil carbon and litter to understand the carbon sequestration over the entire mangrove ecosystem so that a full application for calculating the carbon stock can be utilized.BGB used in this research was based on the AGB calculated.Any error in the AGB could affect the BGB.
Future carbon stock map could be refined with a well distributed plot evenly across the mangrove area.

F
Figure

Table 2 .
Sequestered carbon in Mangrove from AGC and BGC.

Table 3 .
Table showing estimated carbon stock for the whole mangrove site