USGS - science for a changing world

South Florida Information Access (SOFIA)


publications > paper > controls on mangrove forest-atmosphere carbon dioxide exchanges in western Everglades National Park > results

Controls on mangrove forest-atmosphere carbon dioxide exchanges in western Everglades National Park

Abstract
Introduction
Methods
> Results
Discussion
Summary & Conclusions
References
Figures & Tables

3. Results

3.1. Seasonal Factors Controlling NEE

[19] Throughout 2004-2005, NEE exhibited variable patterns primarily in response to differences in TA (Figure 3b) and PAR (Figure 4). Midday, dry season NEE in 2004-2005 ranged from -15 to -25 µmol (CO2) m-2s-1, while nighttime Rd during this period was generally <5 µmol (CO2) m-2s-1 and seldom exceeded 8 µmol (CO2) m-2s-1. NEE generally decreased with TA in February and March while salinity values remained low (18-28 ppt). With the onset of the wet season and higher TA (>25°C), nighttime Rd increased up to 10 µmol (CO2) m-2s-1, and daytime NEE increased to -14 to -23 µmol (CO2) m-2s-1. Minimum daily NEE was as high as -5 µmol (CO2) m-2s-1 when PAR dropped below ~800 µmol (photons) m-2s-1 during this period as the result of afternoon convective cloud formation. During cloudless days, the lowest NEE values (-15 to -22 µmol (CO2) m-2s-1) occurred during the late morning, 1-2 h and 3-4 h before the daily maximum in PAR and TA, respectively. In general, NEE increased throughout the middle of the afternoon in the summer-wet season as TA increased above 30°C. These patterns in Rd and midday NEE persisted until October when the frequency of thunderstorms decreased, daily solar irradiance (Figure 3a) became less variable, and the daily maximum TA was <30°C.

graphs of diurnal carbon dioxide flux patterns, photosynthetic active irradiance levels, surface water salinity, and water level
Figure 4. Representative diurnal CO2 flux patterns during the (a) mid-dry season and (b) mid-wet season. (c and d) Photosynthetic active irradiance levels, (e and f) surface water salinity, and (g and h) water level are included for the same dry season (Figures 4c, 4e, and 4g) and wet season (Figures 4d , 4f, and 4h) periods as the CO2 fluxes. [larger image]

3.2. NEE Responses to Light and Temperature

graphs of half-hourly carbon dioxide flux response to photosynthetic active radiation for the highest and lowest air temperatures during January 2004 through August 2005
Figure 5. Half-hourly CO2 flux response to PAR for the (top) highest (≥28°C) air temperatures and (bottom) lowest (≤21°C) air temperatures during January 2004 through August 2005 period when air temperature was measured at 27 m above ground. The CO2 fluxes in each clearness index (Kt) range (Table 2) were bin averaged by PAR into 30 bins. The Michaelis-Menten function (3) was best fit to each set of binned values. [larger image]

[20] NEE response to the proportion of diffuse irradiance (Kt) depended on air temperature (Figure 5). Temperature also affected the initial canopy quantum yield (a') and daytime Rd, both of which were higher at TA ≥ 28°C compared to the values at TA ≤ 21°C (Table 2). This effect was apparent under both clear and cloudy sky conditions. At TA ≤ 21°C diffuse PAR conditions (Kt ≤ 0.69) lead to a significant increase in GEP2000 (p < 0.02, one-tailed t test) and a decrease in Rd (p < 0.01, one-tailed t test), resulting in an average decrease of ~3 to 5 µmol (CO2) m-2s-1 in NEE.

[21] Minimum NEE values (-15 to -19 µmol (CO2) m-2s-1) occurred when PAR varied between 1400 to 2100 µmol (photons) m-2s-1 and TA ranged from 24 to 28°C. NEE was higher at TA < 21°C compared to rates at higher (>21°C) TA and equivalent PAR. NEE was also generally 1-3 µmol (CO2) m-2s-1 higher when salinity values exceeded 29 ppt, PAR > 600 µmol m-2s-1, and 18°C < TA < 33°C (Figure 6).

[22] There were small but significant linear decreases in LUE with increasing salinity (Figure 7). The slope of this relationship with 95% confidence intervals is -0.00042 ± -0.00008 µmol (CO2) µmol (photons)-1 ppt (salt)-1 and is significantly different from zero (p < 0.05, one-tailed t test). A 48% decrease in LUE occurred when salinity levels increased from 16.7 to 34.7 ppt during the study period.

3.3. Ecosystem Respiration and Tidal Effects

[23] Daytime and nighttime Rd increased during low tides (Figure 8). The daytime reference respiration rate in the Arrhenius model, Rd, 20, increased by 1.9 µmol (CO2) m-2s-1 during low-tide conditions (p < 0.10, one-tailed t test; Table 3). Daytime activation energies (Ea) in the model were also different (p < 0.10, one-tailed t test) between tidal cycles. Differences in Rd due to tides were greater during the daytime (0.9 µmol m-2s-1) compared to nighttime (0.5 µmol m-2s-1). High- and low-tide Rd converged at air temperatures above 29.5°C and 25.4°C during daytime and nighttime periods, respectively. Annual maxima nighttime Rd occurred during June through October reaching 4-7 µmol (CO2) m-2s-1. Annual minimum nighttime Rd (1-3 µmol (CO2) m-2s-1) occurred during December through February.

Table 2. Michaelis-Menton Parameters of Light Response During January 2004 though August 2005 for the Highest and Lowest Daytime Air Temperaturesa
Panel TA,min TA,max Kt,min Kt,max a' GEP2000 Rd n
Highest temperatures 28.0 33.15 0.04 0.69 0.0376 ± 0.0126 20.93 ± 2.09 5.29 ± 2.27 1462
28.0 33.15 0.69 1.00 0.0590 ± 0.0503 24.67 ± 9.41 11.11 ± 9.40 1462
p valueb         0.503 0.523 0.327
Lowest temperatures 5.5 21.0 0.02 0.65 0.0221 ± 0.0054 22.07 ± 2.41 1.74 ± 0.90 895
5.5 21.0 0.65 1.00 0.0207 ± 0.0053 17.58 ± 1.34 4.78 ± 1.37 895
p valueb         0.801 0.019 0.007
aHighest daytime air temperatures are ≥28°C, and lowest daytime air temperatures are ≤21°C. For each temperature range, data were binned to represent high or low Kt. Higher and lower values of Kt represent clear and cloudy sky conditions, respectively. Significance tests were determined at the 95% confidence level. In the following form of the Michaelis-Menten equation, NEE = a' PAR/(1-(PAR/2000) + (a' PAR/GEP2000))-Rd, NEE is the net ecosystem exchange of CO2, a' is the ecosystem quantum yield (µmol CO2 (µmol PAR)-1), GEP2000 is the gross ecosystem productivity (µmol CO2 m-2 s-1) at PAR = 2000 µmol m-2 s-1, and Rd is the ecosystem respiration.

bThe p values refer to the differences in Michaelis-Menten model parameters at high versus low Kt within a temperature range.

contours of daytime average carbon dioxide flux as a function of both incident photosynthetic active radiation and air temperature measured at 27 meters above ground during high-salinity and low-salinity conditions
Figure 6. Contours of daytime average CO2 flux as a function of both incident PAR and air temperature measured at 27 m above ground during both (top) high-salinity (≥29 ppt) and (bottom) low-salinity (<29 ppt) conditions. [larger image]

plot of daily photosynthetic active radiation use efficiency when photosynthetic active radiation exceeded 600 micromoles as a function of daily average salinty
Figure 7. Daily PAR use efficiency (Σ GEPPAR) when PAR exceeded 600 µmol (photons) m-2 s-1 as a function of daily average salinity. The regression line and 95% confidence intervals are included. [larger image]

plots of control of air temperature, measured at 27 meters above ground during daytime and nighttime, on ecosystem respiration rates determined separately for both low-tide and high-tide periods
Figure 8. Control of air temperature, measured at 27 m above ground during (top) daytime and (bottom) nighttime, on ecosystem respiration rates determined separately for both low-tide and high-tide periods. Daytime respiration rates represent daily averages, and nighttime respiration rates represent averages during continuous low- or high-tide periods. An Arrhenius-type exponential function was best fit to ecosystem respiration rates during the daytime and nighttime high- and low-tide periods. [larger image]

Table 3. Mean Daytime and Nighttime Ecosystem Respiration Rates and Arrhenius Model Parameters Under exposed Soil and Inundated Conditions January 2004 Through August 2005a
Summary Characteristic Daytime Nighttime
Exposed Soil at Low Tide Inundated Soil at High Tide Exposed Soil at Low Tide Inundated Soil at High Tide
Number of contiguous periods 398 113 371 191
Mean Rd (µmol m-2 s-1) 3.48 2.57 2.68 2.20
Mean air temperature (°C) 24.4 25.7 19.6 20.7
Rd20 b (µmol m-2 s-1) 2.546 ± 0.288 0.611 ± 0.298 2.579 ± 0.141 2.101 ± 0.189
Ea/Rb (K) 5702 ± 1414 18980 ± 4767 899 ± 935 4237 ± 1697
R2b 0.115 0.419 0.011 0.131
RMSE 2.42 2.01 1.33 1.18
aShown with 95% confidence intervals. Exposed soil conditions are at low tide, and inundated conditions are at high tide.

bThe Arrhenius-type model of Lloyd and Taylor [1994] was used to relate ecosystem respiration (Rd) to air temperature T (deg K).

Rd. = Rd20 exp[(Ea/R)(1/293K-1/TK)].

3.4. Seasonal and Annual NEP

[24] There were distinctive seasonal patterns in total daily and nighttime NEP (Figure 9). Daily total NEP was greatest between March and May and lower in July through October (Figure 10). During this period, NEP was reduced (NEE of -10 to 0 µmol (CO2) m-2s-1) during afternoon thunderstorms that reduced solar irradiance (<500 W m-2). However, reductions in NEP values were also observed during cloudless conditions in July to October when the combination of high PAR (>1400 µmol (photons) m-2s-1) and high TA (>28°C) contributed to increased daytime and nighttime Rd (4-7 µmol (CO2) m-2s-1). Low NEP (1 to 4 g C m-2 d-1) during December to February was attributed to low-temperature inhibition of photosynthesis and shorter day lengths. During December to February reductions in total daily NEP occurred even though this period was also marked by annual minimum respiration rates (1-3 µmol (CO2) m-2s-1). Low-temperature (TA < 21°C) effects on carbon assimilation, rather than respiration, caused reduction in NEP during this period.

plots of total daily carbon net ecosystem production and daytime and nighttime contributions to total daily carbon net ecosystem production during January 2004 through August 2005
Figure 9. (top) Total daily C NEP and (middle) daytime and (bottom) nighttime contributions to total daily C NEP during January 2004 through August 2005. Centered moving averages (15 day) are included. [larger image]
graph of monthly sums and errors of carbon net ecosystem production during January 2004 through August 2005
Figure 10. Monthly sums and errors of C NEP during January 2004 through August 2005. [larger image]

[25] During 2004, monthly NEP (Figure 10) ranged from 126 ± 9 to 132 ± 15 g C m-2 between March and May and varied from 74 ± 10 to 86 ± 8 g C m-2 between July and October. NEP increased to 101 ± 10 g C m-2 during November as a result of decreasing nighttime temperatures (lower Rd), low salinity stress, and fewer afternoon thunderstorms compared to the warmer summer months. Salinity values remained low (<29 ppt) during December 2004 to January 2005, but annual minimum NEP values (75 ± 6 to 76 ± 7 g C m-2) during this period were the result of reduced daytime carbon assimilation. During 2004, the mangrove forest assimilated 1170 ± 127 g C m-2 (Table 1). During the 8 months of measurements in 2005, the forest assimilated 832 ± 97 g C m-2, which is equivalent to an annual rate of 1175 ± 145 g C m-2 yr-1. We found pronounced interannual differences in monthly NEP during 2004 and 2005. For example, NEP in March 2005 was ~20% lower than in March 2004. This resulted from a combination of greater cloud cover, lower solar irradiance, higher nighttime TA, and higher salinity values during March 2005. In contrast, NEP during July-August 2005 was 35% higher compared to the same period in 2004. The climatic conditions during these months in 2004-2005 were similar, with the exception of local rainfall. At a monitoring station near the tower site, rainfall was 261 mm and 590 mm during June-July 2004 and 2005, respectively. The increased rainfall in 2005 resulted in increased freshwater discharge, which lowered salinity levels, increased the duration of flooding, and lowered soil respiratory fluxes compared to 2004. During June-July 2005, average Rd reached 3.08 ± 1.12 g C m-2s-1 whereas in 2004 Rd attained 3.83 ± 1.93 g C m-2s-1.

< Methods | Discussion >



| Disclaimer | Privacy Statement | Accessibility |

U.S. Department of the Interior, U.S. Geological Survey
This page is: http://sofia.usgs.gov/publications/papers/mang_exchange/results.html
Comments and suggestions? Contact: Heather Henkel - Webmaster
Last updated: 04 September, 2013 @ 02:04 PM(TJE)