Impact of forest conversion to oil palm and rubber plantations on microclimate and the role of the 2015 ENSO event

Agricultural and Forest Meteorology 252 (2018) 208–219

Contents lists available at ScienceDirect

Agricultural and Forest Meteorology journal homepage: www.elsevier.com/locate/agrformet

Impact of forest conversion to oil palm and rubber plantations on microclimate and the role of the 2015 ENSO event

T

⁎

Ana Meijidea,b, , Chandra Shekhar Badua, Fernando Moyanoa, Nina Tirallaa, Dodo Gunawanc, Alexander Knohla,d a

University of Goettingen, Bioclimatology, Büsgenweg 2, 37077 Göttingen, Germany University of Granada, Department of Ecology, Avda. Fuente Nueva s/n 18071, Granada, Spain c Center for Climate Change Information, Agency for Meteorology Climatology and Geophysics Jl. Angkasa I No. 2, 10720, Jakarta, Indonesia d University of Goettingen, Centre of Biodiversity and Sustainable Land Use (CBL), 37073 Göttingen, Germany b

A R T I C L E I N F O

A B S T R A C T

Keywords: Microclimate Tropical forest Land-use change Canopy openness ENSO

Oil palm and rubber expansion is a main driver of the widespread deforestation of tropical rainforests taking place in South-East Asia, particularly in Indonesia. The replacement of forests with monoculture plantations of rubber and oil palm reduces biodiversity and carbon pools but also modifies canopy structure, which is an important determinant of microclimate. There is, however, a lack of quantitative information characterizing the effect of such land transformation on microclimate. We report the first medium-term observations of belowcanopy microclimatic conditions (air temperature, relative humidity, vapour pressure deficit and soil temperature) across forest, jungle rubber agroforest, oil palm and rubber monoculture plantations in Sumatra/ Indonesia. The data set covers a period of approximately three years (2013–2016) and includes one of the strongest El Niño-Southern Oscillations (ENSO) of the last decades. Forests were up to 2.3 and 2.2 °C cooler than oil palm and rubber monocultures respectively. The monocultures were also drier (11.9% and 12.8% less in oil palm and rubber respectively) and had higher vapour pressure deficit (632 Pa and 665 Pa higher in oil palm and rubber respectively) than the forest, while differences in soil temperature were less pronounced. Conversion from forest to other land uses, especially to monocultures, also amplified the diurnal range of all microclimatic variables studied. Jungle rubber stands out as the transformed land-use system that maintains more stable microclimatic conditions. Our results indicate that canopy openness is a key driver of below-canopy microclimate, and hence could be used in climate models to better evaluate climatic feedbacks of land-use change to rubber and oil palm. The ENSO event of 2015 led to warmer and drier conditions than in the previous two years in all four land-use systems, especially in the forest (up to 2.3 °C warmer, 8.9% drier and up to 351 Pa more during ENSO). The relative effect of ENSO was lower in the monoculture plantations, where below-canopy microclimate is generally more similar to open areas. Forests exhibited the largest differences with the pre-ENSO years, but still maintained more stable microclimatic conditions (lower temperatures and vapour pressure deficit and higher relative humidity) due to their higher climate regulation capacity. During ENSO, microclimatic conditions in jungle rubber were comparable to those in the monocultures, suggesting that while forests buffered the increase of temperature, jungle rubber might have surpassed its buffering capacity to extreme events. This capacity of buffering extreme climatic events should be considered when assessing the effects of land-use change.

1. Introduction Global demand for agricultural products such as food, feed and fiber is a mayor driver of land-use change in the tropics, which occurs mainly at the expense of forests (Gibbs et al., 2010). In Southeast Asia, rainforests have been logged since the mid-20th century, usually followed by tree cash crops such as oil palm (Elaeis guineensis) and rubber

⁎

(Hevea brasiliensis) monocultures (Abood et al., 2015; Wilcove and Koh, 2010; Ziegler et al., 2009). Indonesia has the highest annual loss of rainforest worldwide (Margono et al., 2014), while being the largest palm oil producer and the second largest rubber producer worldwide (FAO, 2017). Oil palm and rubber monocultures are more profitable than forests and agroforests, but are not able to maintain most ecological functions (Clough et al., 2011), as a result of decreases in biomass

Corresponding author at: Department of Ecology, University of Granada, Avenida de Fuentenueva s/n 18071 Granada, Spain. E-mail addresses: [email protected], [email protected] (A. Meijide).

https://doi.org/10.1016/j.agrformet.2018.01.013 Received 10 October 2017; Received in revised form 5 January 2018; Accepted 6 January 2018 0168-1923/ © 2018 Elsevier B.V. All rights reserved.

Agricultural and Forest Meteorology 252 (2018) 208–219

A. Meijide et al.

The El Niño Southern Oscillation (ENSO) is a major mode of variability of global precipitation and temperature, comprising alternating warming (El Niño) and cooling (La Niña) phases. The El Niño event in 2015 could be the second strongest ENSO event reported so far after 1997 (Varotsos et al., 2016). In Indonesia, ENSO is expected to create hotter and drier conditions (Allan, 2000; Harger, 1995; Susilo et al., 2013). To our knowledge, no studies are available showing the effect of the ENSO event on microclimatic conditions in these land-use systems, but results from a modeling study suggest that land clearing can amplify the effect of ENSO in Southeast Asia (Tölle et al., 2017). Natural ecosystems have greater climate regulation capacities than agroecosystems (Anderson-Teixeira et al., 2012), and thus, responses to climate warming are attenuated in forests with denser canopies (De Frenne et al., 2013). Therefore, we hypothesized that ENSO will have a stronger effect on the less natural land-use systems. Buffering from extreme conditions is beneficial for biodiversity and ecosystem functioning (Chen et al., 1999; Clough et al., 2016), and on this basis, the stability in microclimatic conditions is assessed. We performed the first systematic study evaluating the microclimatic conditions below the canopy in different land-use systems in the lowlands of Sumatra, namely forests, jungle rubber, rubber and oil palm monocultures. We measured air temperature (Ta, °C), relative humidity (RH, vol%), vapor pressure deficit (Vpd, Pa) and soil temperature (Ts, °C), and assessed the microclimate in these land-use systems during three consecutive years including the ENSO event in 2015. The aims of this study were i) to quantify microclimatic conditions in forests, rubber agroforests, oil palm and rubber monocultures and evaluate their differences; ii) to investigate the drivers of these climatic differences and iii) to assess the effect of the ENSO event in 2015 on the microclimatic conditions in the different land-uses evaluated.

(Kotowska et al., 2015), soil carbon (Guillaume et al., 2015) and biodiversity (Barnes et al., 2014; Rembold et al., 2017). The climatic impact of carbon emissions associated to deforestation and land-use change have long been studied, including some recent work on carbon dioxide emissions following forest conversion to oil palm (Carlson et al., 2012a, b; Ramdani and Hino, 2013). Biophysical effects of land cover on microclimate remain poorly understood and are therefore not considered in most models and treaties (Alkama and Cescatti, 2016), despite the fact that microclimate differences between different land covers can be of similar scale or even larger than those projected to happen under climate change (Sabajo et al., 2017; Suggitt et al., 2011). Changes in land cover type affect the local climate by modulating the land-atmosphere fluxes of energy and water (Alkama and Cescatti, 2016; Bright et al., 2017; Ellison et al., 2017). Forest conversion to other land uses typically leads to an amplification of the diurnal temperature variation and increases the mean and maximum air temperature (Alkama and Cescatti, 2016). Accordingly, studies in temperate and tropical ecosystems show that forests are usually cooler than clear cut areas or the agroforestry systems that substitute them (Chen et al., 1999, 1993; Porté et al., 2004). The lower temperatures in forests are related to their cooling effect, mainly a result of higher evapotranspiration, which is especially strong in the tropics (Li et al., 2015). Additionally, the top of the canopy reflects or intercepts the sun light, and therefore, denser canopies will result in lower light penetration and thus lower below-canopy temperatures (Foley et al., 2003). For oil palm plantations, the limited studies available suggest that their microclimate is different from that of forests in the same regions (Drescher et al., 2016; Hardwick et al., 2015; Luskin and Potts, 2011; Sabajo et al., 2017), experiencing higher temperatures and lower humidity. Hardwick et al. (2015) observed that maximum temperature in logged forests and oil palm plantations was 2.5 °C and 6.5 °C greater than in primary forests, respectively. Similarly, Luskin and Potts, (2011) measured that during daytime hours, oil palm plantations were 2.8 °C warmer and drier than natural vegetation. These differences seem to be related to decreased leaf area index (LAI) (Hardwick et al., 2015). However, these studies are based on short-term data series or modeled results and fail at providing information on the inter-annual variability. Despite of the relevance of oil palm expansion, in the Indonesian island of Sumatra, where oil palm production is concentrated (OECDFAO, 2012) and the highest forest cover loss in Indonesia is found (Laumonier et al., 2010; Margono et al., 2014; Miettinen et al., 2011), rubber plantations cover more than 25% of the territory (Clough et al., 2016). Rubber is grown mainly as a monoculture, but also as an agroforestry system (i.e. jungle rubber) where secondary forests are enriched with rubber (Gouyon et al., 1993). Therefore, in order to assess microclimatic effects due to land-use change in the region, rubber ecosystems should also be studied. Studies so far (Böhnert et al., 2016; Clough et al., 2016; Drescher et al., 2016, Sabajo et al., 2017) have been based on measurements over short time periods. Jiang and Wang, (2003) assessed the climatic effects due to the conversion of natural forest to rubber plantations and found that replacement of natural forest with rubber had no effect on the local rainfall. However, to our knowledge there are no detailed studies addressing microclimatic effects (i.e. temperature or relative humidity) of rubber, neither as monoculture nor as agroforestry system. Thus, there is a need to understand how microclimate varies with the change in land-use systems in a tropical landscape, where forests are being replaced by oil palm and rubber plantations. Additionally, climatic impacts due to land-use change are expected to be stronger under maritime conditions, as found in Indonesia, than under continental conditions, as 40% of the global tropical latent heating of the upper troposphere takes place over the Maritime Continent (van der Molen et al., 2006). Therefore, the climatic effects of extensive land-use change due to oil palm and rubber expansion in Indonesia could have global climatic implications, which need to be properly evaluated.

2. Materials and methods 2.1. Measurements sites Our study was carried out within the frame of the EFForTs project (Drescher et al., 2016) in the Jambi province in Sumatra, Indonesia. Average annual temperature and rainfall in the area of study (mean ± sd between 1991 and 2011, data from the Airport Sultan Thaha in Jambi) were 26.7 ± 0.2 °C and 2235 ± 381 mm. We evaluated microclimatic conditions below the canopy in four land-use systems: primary degraded forest (see Drescher et al., (2016) for definition), from now on called forest (F), jungle rubber (JR), rubber monoculture (R) and oil palm monoculture (O). We established eight 50 × 50 m core plots from each land use system in two different landscapes (4 each) in the lowlands of the Jambi province (Fig. 1). The two landscapes, i.e., the ‘Bukit Duabelas landscape’ and the ‘Harapan landscape’ had a clay loam and sandy loam texture respectively (Allen et al., 2015; Drescher et al., 2016). There was a total of 32 plots in our study, 16 in each landscape. We additionally installed four “reference meteorological stations” in open areas, two in each landscape, one closer to the forests plots and the other one closer to the transformed land uses (rubber and oil palm plots; Fig. 1). Additional details on the experimental design can be found in Drescher et al., (2016). 2.2. Micrometeorological measurements A meteorological station was installed in the center of each core plot. Each station was equipped with a thermohygrometer (Galltec Mela, Bondorf, Germany) placed at 2 m height to record Ta (°C) and RH (%) below the canopy, and a Trime-Pico 32 (IMKO, Ettlingen, Germany) at 0.3 m depth in the soil to monitor Ts (°C). Both sensors measured once every hour, and data were recorded in a UIT LogTrans 16-GPRS data logger (UIT, Dresden, Germany). Data were collected from April 2013 to March 2016 in the 32 core plots. 209

Agricultural and Forest Meteorology 252 (2018) 208–219

A. Meijide et al.

Fig. 1. Location of the study site representing Indonesia and the Jambi province in grey (a), the Jambi province (b), core plot design and reference meteorological stations in the Bukit Duabelas region (c) and core plot design and reference meteorological stations in the Harapan region (d).

The lack of repetitive measurements of canopy openness did not allow us to assess the effect of rubber leaf shedding on the microclimatic conditions. However, litter in rubber plots was higher during the dry months (Kotowska et al., 2016), suggesting that indeed, leaf shedding is larger during this period. In order to assess the possible effects of the seasonality of leaf shedding on the microclimatic conditions, we compared our measurements below the canopy in R and JR with data from other land uses and the reference meteorological stations (Fig. S1). We did not observe any pattern in R and JR plots (especially during the drier months, when leaf shedding should have been the highest), that was not observed in the reference meteorological stations and other plots, and therefore could be related to the partial leaf shedding of rubber trees.

Reference meteorological stations were equipped with 2 thermohygrometers (type 1.1025.55.000, Thies Clima, Göttingen, Germany) at 0.5 and 2 m height, a global radiation sensor (CMP3 Pyranometer, Kipp & Zonen, Delf, The Netherlands) installed at a height of 3 m, a net radiometer (NR Lite2, Kipp & Zonen), two precipitation transmitters (Thies Clima,) at 1.5 m height and separated by about 6 m, a 3-cup anemometer and a wind direction sensor (both from Thies Clima). Measurements were taken every 15 s and averaged and stored on a DL16 Pro data logger (Thies Clima) every 10 min. These stations measured during the same period as the measurements were taken in the core plots.

2.3. Canopy openness

2.4. Data analysis

Canopy openness was derived by Drescher et al. (2016) from hemispherical photographs taken at 1.2 m above the ground from 32 positions within each plot using a Canon EOS 700D SLR camera (Canon Inc., Ōta City, Japan) and SIGMA 4.5 mm F2.8 EX DC circular fisheye lens (Sigma Corp., Ronkonkoma, USA). The camera was vertically levelled and aligned to the magnetic north using a bubble level slotted into the flash socket. Photographs were processed using ‘ImageJ’ (Rasband et al., 1997). For additional information see Drescher et al., (2016). Canopy openness was measured once in each plot between March and April 2014. Variations in canopy openness are expected to be small in all land-uses except for the rubber plantations, as rubber shed their leaves at certain periods during the dry season. We thus did not account for seasonal variation of canopy openness, similarly to Hardwick et al., (2015). Measurements were carried out in each plot when some rubber trees had partially shed their leaves, but some others were still fully leaved. Therefore, our measurements represent an intermediate level of canopy openness in rubber plots during the year.

Data were filtered to remove outliers produced due to instrument failure, setting upper and lower limits for all variables measured in the plots (ie. Ta: 10–40 °C; RH: 10–100%; Ts: 10–40 °C). Vapour pressure deficit (Vpd, Pa) was calculated as the difference between saturated water vapour pressure (es) and actual water vapour pressure (e) following Eq. (1).

Vpd = (es − e ) × 100

(1)

where es was calculated according to Bolton (1980) from Ta (Eq. 2).

17.67 × T a ⎤ es = (0.6112) exp ⎡ ⎣ Ta + 243.5 ⎦

(2)

And e was estimated using Eq. (3).

e =

210

RH × es 100

(3)

Agricultural and Forest Meteorology 252 (2018) 208–219

A. Meijide et al.

Fig. 2. Weekly sums of precipitation (a) and means for air temperature (b), relative humidity (c) and vapor pressure deficit (d) from measurements at the four reference meteorological stations (open areas) from September 2013 to March 2016. Shaded area indicates the period when the effects of the ENSO in 2015 were more pronounced.

3. Results

The rainfall and temperature data collected from reference meteorological stations was used to identify the months where the effects of ENSO in 2015 were more pronounced (Fig. 2), which were August, September and October. The statistical significance of differences in means of measured climate variables across four different land-use systems was tested using the one-way analysis of variance (ANOVA, p < 0.05) with F-test (F > 1 rejects the null hypothesis, which assumes that the means are equal). Further, the Tukey’s Honest Significant Difference (HSD) test was used to test for significant differences between land-use systems or differences in means before and after the ENSO period. Different models were tested to assess the possible relationships between mean canopy openness and climate variables. Data processing, statistical analysis and plotting were performed using R version 3.2.5 (R Core Team, 2016-04-14).

3.1. Microclimatic differences between land-uses The overall daily means, maximum and minimum of the climate variables across the four land-use systems are presented in Table 1. The mean Ta during the study period was highest in O and R (25.5 ± 0.2 and 25.5 ± 0.1 °C respectively), followed by JR and with F exhibiting the lowest temperatures (24.8 ± 0.1 °C). An inverse trend was observed in RH, with lower values in O and highest in F. Consequently, Vpd was highest in R (both on the mean, i.e. 391 ± 44 Pa, minimum and maximum daily means) and similarly high in O, medium in JR and the lowest in F (169 ± 39 Pa). Soil temperature followed a similar trend as Ta, even if the differences among the different land uses were smaller. Mean daily maximum and minimum for all land-uses are 211

Agricultural and Forest Meteorology 252 (2018) 208–219

25.2 ± 25.7 ± 26.4 ± 26.3 ± 0.009 4.68 24.8 ± 25.4 ± 25.8 ± 25.9 ± 0.029 3.47 25.0 ± 25.6 ± 26.1 ± 26.1 ± 0.017 4.01 719 ± 101a 1104 ± 86b 1351 ± 96c 1384 ± 72c < 0.001 45.05 3 ± 2a 2 ± 2a 6 ± 6a 15 ± 20a 0.306 1.26

Max.

99.9 ± 99.9 ± 99.8 ± 99.6 ± 0.325 1.21

Min.

82.1 ± 2.5a 74.6 ± 1.7b 70.2 ± 1.5c 69.3 ± 1.3c < 0.001 40.91

0.1a 0.0a 0.2a 0.5a

169 ± 39a 272 ± 33b 359 ± 38c 391 ± 44c < 0.001 25.45

Mean Mean

Vpd (Pa)

Min.

Max.

Ts (°C)

0.7a 0.2ab 0.7b 0.2b

Min.

0.7a 0.2ab 0.7b 0.2b

Max.

0.7a 0.2ab 0.7b 0.3b

reported in Table 1. We observed a temperature difference of up to 2.3 and 2.2 °C between forests and oil palm and rubber monocultures respectively. Differences with the F in minimum RH and maximum Vpd were 11.9 and 12.8% and 632 and 665 Pa for O and R respectively. No significant differences were found in the daily means between the “Bukit Duabelas” and “Harapan” landscapes for any of the studied microclimatic variables. The daily amplitude of the different microclimatic variables, estimated as the difference between the maximum and minimum daily means presented in Table 1, is shown in Fig. 3. Microclimatic conditions were quite constant during the day in F, showing the smallest daily amplitude of Ta, RH and Vpd. For the same variables, daily amplitude was intermediate in JR, and the largest in O and R. Daily amplitude in soil moisture was smaller in F and JR and larger in O and R. The mean diurnal cycles of Ta, RH, Vpd and Ts followed similar trends across the four land-use systems (Fig. 4). During the night, from 20:00 h to 7:00 h, Ta was similar in all land-uses, and its daily minimum was observed at 6:00 h (around 23 °C; Fig. 4a). Ta increased following sunrise, reaching its daily maximum at 14:00 h. Maximum temperature values were higher (around 30 °C) in monoculture plantations than in the forest (28 °C). Ta decreased in the afternoon and was then again similar in all four land-use systems after 19:00 h. The mean RH (Fig. 4b) was as its maximum in all land-use systems from 21:00 to 6:00 h. Relative humidity decreased in parallel with the increases of Ta, and reached its minimum at 14:00 h. The minimum was < 75% in rubber and oil palm plantations while it remained > 85% in the forest. Decreases of RH were observed after 16:00 h. Jungle rubber exhibited an intermediate behavior between forest and monoculture plantations, with more moderate temperatures and higher RH than O and R. We observed greater differences between land-uses in the diurnal cycles of Vpd than in diurnal cycles of the other microclimatic variables (Fig. 4c). Mean maximum daily Vpd was around 1200 Pa in O and R, 900 Pa in JR and 600 Pa in F. Additionally, in all land-uses, the increases of Ta and Vpd or decreases of RH in the morning, were delayed in the F up to 2 h in relation to the monocultures, and started 1–2 h earlier in the evening. Soil temperatures did not show such a clear diurnal cycle, and even if similar trends as in Ts were observed, with lower values in F, intermediate in J and lower in O and R, the differences were not significant (Fig. 4d).

95.6 ± 1.0a 93.3 ± 0.8b 91.3 ± 0.8c 90.7 ± 0.8c < 0.001 24.94 28.3 ± 0.2a 29.8 ± 0.3b 30.6 ± 0.3c 30.5 ± 0.1c < 0.001 72.63

The mean average canopy openness in the four land-use systems is presented in Table 2. Canopy openness varied among the studied landuses: lower values were observed in F (2.54%) and highest in O (15.64%). The microclimatic variables in the different land-uses showed a non-linear logarithmic relationship with the mean canopy openness. The model which best fitted the relationship between these climatic variables and canopy openness is presented in Eq. (4). y = 3 × log (x)

22.4 ± 22.2 ± 22.2 ± 22.4 ± 0.035 3.28 24.8 ± 0.1a 25.2 ± 0.1b 25.5 ± 0.2c 25.5 ± 0.1c < 0.001 24.78 F JR O R p-val. F- stat.

Min. Mean

Ta (°C)

(4)

where y represents the different climatic variables (Ta, RH, Vpd or Ts) and x is the canopy openness. The mean daily Ta (Fig. 5a; R2 = 0.75, F1, 29 = 84.85, p < 0.001) and mean daily Vpd (Fig. 5c; R2 = 0.71, F1, 29 = 69.49, p < 0.001) had a strong positive relationship with the mean canopy openness. Similarly, the mean daily soil temperature also had a positive relationship with canopy openness, even if this relationship was not as strong (Fig. 5d; R2 = 0.24, F1, 29 = 9.40, p < 0.001). The mean daily RH had a negative relationship with canopy openness (Fig. 5b; R2 = 0.68, F1, 29 = 61.5, p < 0.001). When the relationships with canopy openness are assessed with the daily minimum and maximum values, we observe that for Ta, Vpd and Ts, a more significant relationship is found for the daily maximum values (Fig. 5e, g and h), or for the daily minimum values in the case of RH (Fig. 5f).

0.1a 0.1a 0.1a 0.2a

Mean Max.

RH (vol. %)

3.2. Relationship with canopy openness

L-U

Table 1 Daily mean, maximum and minimum values (and their margin of error for a 95% confidence interval) of the measured climatic variables across four different land-use (L-U) systems, i.e. forest (F), jungle rubber (JR), oil palm (O) and rubber (R) from April 2013 to March 2016. Results from one-way ANOVA test, carried out to identify significant differences between land-uses, are presented (p-value and F-statistic). Different letters within columns indicate significant differences applying Tukey’s Honestly Significant Difference (HSD) test.

A. Meijide et al.

212

Agricultural and Forest Meteorology 252 (2018) 208–219

A. Meijide et al.

Fig. 3. Diurnal amplitude (difference between daily maximum and daily minimum) for air temperature (a), relative humidity (b), vapour pressure deficit (c) and soil temperature (d) across the different land-use systems i.e. forest (F), jungle rubber (JR), oil palm (O) and rubber (R). Error bars represent the margin of error for a 95% confidence interval.

evaluating the effects on the daily maximum values for Ta and Vpd or on the daily minimum values for RH (Table 3). ENSO also had a significant effect on mean diurnal cycles (see Fig. 6), which was more pronounced in F and JR.

3.3. Effect of ENSO Weekly means of Ta, RH and Vpd and weekly rainfall sums from our 4 reference meteorological stations in the area of study show an anomalous dry period in 2015, during the months of August, September and October (Fig. 2), coinciding with the ENSO event. In this period, rainfall was significantly lower than in the previous 2 years (Fig. 2a), and we observe a decrease in RH (Fig. 2c and Fig. S1) and an increase of Vpd (Fig. 2d and Fig S1). Regarding Ta, in 2015 it was more constant throughout the year than in 2013 and 2014, when it usually decreased for some months during the rainy season. To evaluate the effect of the ENSO event in 2015, we compared 3 years of study, i.e. the ENSO year (2015) and the previous two years, focusing only on the months of August, September and October, which correspond to the dry season. ENSO resulted in an increase in overall mean Ta, Vpd and Ts in 2015, and a decrease in RH in all four land-use systems (Fig. S2). The mean Ta in 2015 in F and JR increased by 1 °C in 2015 when compared to the previous two years, while it only increased 0.7 and 0.6 °C in oil palm and rubber plantations respectively (Table 3). The mean RH decreased by 8.9% in F and by 9.1% in JR, and by 6.8 and 7.6% in O and R plantations respectively. Accordingly, the mean Vpd increased by 351 Pa in F, 301 Pa in JR and around 275 and 262 Pa in O and R respectively (Table 3). The relative increase of Vpd represented an increase of 234% in F, while it was much lower in the monoculture plantations (70 and 61% in O and R respectively). Differences in the daily means between 2015 and the previous 2 years were significant (p < 0.001) for all land-uses for Ta, RH and Vpd (see also SFig. 2). Differences in Ts between 2015 and the previous years were not significant in any land use. Changes were even more significant when

4. Discussion 4.1. Land-use change effects in local microclimatic conditions Our results indicate that the conversion of forest to jungle rubber, rubber and oil palm plantations modifies the microclimate (i.e. Ta, RH, Vpd and to a lower extent, Ts). Monoculture plantations (R and O) were warmer and drier (in terms of relative humidity) than F and JR. Jungle rubber, even if it also exhibited warmer and drier conditions than F, led to more limited increases of Ta and Vpd, and lower decreases in RH, with no significant differences in Ts. JR had an intermediate behavior between the monoculture plantations and F, and as observed by Clough et al., (2016) maintains the ecosystem function of climate regulation. Large microclimatic differences have already been observed between F and O (Hardwick et al., 2015; Luskin and Potts, 2011; Sabajo et al., 2017), pointing to large climatic effects due to oil palm expansion. However, in tropical areas where land use is largely changing, forest fragments very often coexist with monoculture plantations and agroforestry systems creating mosaic landscapes (Clough et al., 2016; Laurance et al., 2014). The results presented in this study are relevant for understanding the microclimatic conditions of each of the land uses assessed, but can also provide information to evaluate the effects at the landscape level, in areas where different agroforestry systems coexist. In previous studies comparing forests and oil palm plantations, the 213

Agricultural and Forest Meteorology 252 (2018) 208–219

A. Meijide et al.

Fig. 4. Mean diurnal cycles of the air temperature (a), relative humidity (b), vapour pressure deficit (c) and soil temperature (d) expressed as mean (line) and the margin of error for a 95% confidence interval (shaded area) in the different land-uses, i.e. forest (F), jungle rubber (JR), oil palm (O) and rubber (R).

changes in their composition and structure (Margono et al., 2012). These changes in the canopy and structure might justify the lower differences observed in our study. Information on microclimatic conditions in rubber plantations is still very limited (Sabajo et al., 2017; Böhnert et al., 2016; Clough et al., 2016; Drescher et al., 2016), and to our knowledge, mainly coming from our study plots. We observed a very similar behavior when comparing O and R, both with higher temperatures (mainly in the air but also in the soil) and Vpd and lower RH than in the F. This seems to be related to their similar high canopy openness and a similar simple vegetation structure (Clough et al., 2016). Additionally, they are both a monoculture, with very limited or no underground vegetation. Even if rubber shed their leaves while oil palms are perennial, in our study we did not find significant microclimatic differences between both land uses. In our plots, the mean canopy openness was slightly higher in O (Table 2), but when evaluating individual canopy openness in O and R plots, there was a similar and large variability in both land uses (Fig. 4), which might be related to their different ages, plantation patterns or possible leaf shedding in rubber. As canopy openness was not measured repetitively during the year, we cannot provide information on the possible effects that leaf shedding might have in the microclimatic conditions in R plots. Mean diurnal temperature in the transformed land uses was higher than in the forest, but the effect was even more significant in the daily maximum temperatures: O and R were up to 2.3 and 2.2 °C warmer than the forest, respectively. Consequently, our results are in agreement with Alkama and Cescatti (2016), and we can conclude that forest losses largely amplify the diurnal temperature variation, which was below 6 °C in the forests and more than 8 °C in the monocultures

Table 2 Mean canopy openness (and the margin of error at 95% confidence interval) of four land-use (L-U) systems, i.e. forest (F), jungle rubber (JR), oil palm (O) and rubber (R). L-U

Canopy openness (%)

F JR Ol R

2.54 ± 0.33 7.10 ± 1.42 15.64 ± 2.94 13.46 ± 2.38

latter were up to 6.5 °C warmer (Hardwick et al., 2015). In our study region, oil palm plantations were also warmer. The differences we observe are much smaller (2.3 °C), but comparable to other studies where forest and mature plantations were compared (Luskin and Potts, 2011). The 6.5 °C difference found by (Hardwick et al., 2015) resulted from comparing oil palm plantations with primary forests. Large temperature increases have been reported in young plantations (Luskin and Potts, 2011, Sabajo et al. 2017) which could be related to the decreased evapotranspiration and increase of sensible heat fluxes (Meijide et al., 2017). The oil palm plantations evaluated in Hardwick et al., (2015) were relatively young, i.e. from 6 to 12 years old (Ewers et al., 2011), which could partly explain the larger differences they observed. However, they found a smaller (i.e. 4 °C) difference of temperature when oil palm was compared with logged forests. In our study, the forest plots represent primary degraded forests (Drescher et al., 2016) following the classification of Margono et al., (2014). These degraded forests are either fragmented forests or they have been affected by selective logging or human disturbances, resulting in partial losses of their canopy or 214

Agricultural and Forest Meteorology 252 (2018) 208–219

A. Meijide et al.

Fig. 5. Relationships between canopy openness of each land-use type, i.e. forest (F), jungle rubber (JR), rubber (R) and oil palm (O), and the daily mean (a, b, c and d) as well as the mean daily maximum and minimum (e, f, g and h) of microclimatic variables. Solid grey lines indicate exponential fitting for mean and maximum diurnal values and dashed grey lines exponential fitting for minimum diurnal values.

4.2. Effect of canopy openness on microclimatic conditions

(Fig. 3). Increases in the daily amplitude were even more striking in the RH, with an amplitude 1.7 times larger in the monocultures than in the forest and in Vpd, where daily amplitude was doubled in the monocultures when compared to the forests (see Fig. 3). Other studies in the area have also found that while forest and jungle rubber show a variety of microclimatic conditions from the ground to the upper parts of the canopy, conditions are relatively constant with height in the monocultures (Böhnert et al., 2016). All these changes in microclimatic conditions, both between land uses and at different heights of the canopy, may potentially result in decreases of species biodiversity caused by the modification of the habitat (Böhnert et al., 2016; Fitzherbert et al., 2008).

It is well known that canopies influence forest microclimates (Aussenac, 2000; Carlson and Groot, 1997; Chen et al., 1993; Kovács et al., 2017). Modification of microclimatic conditions has been linked to changes in vegetation cover and seasonality (Beltrán-Przekurat et al. 2008; Betts 2001). In this study, we evaluated the possibility of using canopy openness as an explanatory variable for understanding belowcanopy microclimatic conditions. For forests and oil palm, Hardwick et al. (2015) showed that leaf area index was negatively correlated to Ta and positively related to RH. Tree canopies limit the amount of sunlight reaching the soil surface and thus reduce air and soil temperature below the forest canopy (Bhatti et al., 2016), explaining the higher temperatures in the land uses with lower canopy cover. We also identified a strong positive relationship

Table 3 Mean ± standard deviation of the increase/decrease in the daily means, percentage or increase/decrease during ENSO in relation to the previous two years and mean daily maximum or minimums in 2013 and 2014 compared to 2015, of the respective variables, i.e. air temperature (Ta), relative humidity (RH),vapour pressure deficit (Vpd) and soil temperature (Ts) for each land-use (L-U) system, i.e. forest (F), jungle rubber (JR), oil palm (O) and rubber (R). Increases during ENSO are indicated with a positive symbol (+) and decreases with a negative symbol (-). L-U

F JR O R

Ta (°C)

RH (vol. %)

Vpd (Pa)

Ts (°C)

Daily mean

% of daily mean

Daily max.

Daily mean

% of daily mean

Daily min.

Daily mean

% of daily mean

Daily max.

Daily mean

% of daily mean

Daily max.

+1.0 +1.0 +0.7 +0.6

+3.9 +3.8 +2.6 +2.4

+2.3 +2.0 +1.1 +1.0

−8.6 −8.5 −6.2 −6.8

−8.9 −9.1 −6.8 −7.6

−18.4 ± 1.0 −14.6 ± 0.1 −9.7 ± 1.3 −9.7 ± 0.1

+351 +301 +275 +262

+234 +103 +70 +61

+917 +580 +577 +462

+0.5 +0.4 +1.0 +0.5

+2.1 +1.7 +3.9 +1.8

+0.5 +0.5 +1.1 +0.4

± ± ± ±

0.0 0.0 0.0 0.0

± ± ± ±

0.0 0.1 0.0 0.0

± ± ± ±

0.3 0.0 0.9 0.1

215

± ± ± ±

9 0 38 6

± ± ± ±

43 3 95 1

± ± ± ±

0.4 0.2 0.1 0.5

± ± ± ±

0.4 0.2 0.2 0.4

Agricultural and Forest Meteorology 252 (2018) 208–219

A. Meijide et al.

Fig. 6. Mean diurnal cycles of the microclimatic variables during August, September and October of 2013, 2014 and 2015 (ENSO year), expressed as mean (line) and the margin of error for a 95% confidence interval (shaded area) in the different land uses, i.e. forest (a, e, i), jungle rubber (b, f, j), oil palm (c, g, k) and rubber (d, h, l) plantations.

relevance of canopy cover as a driver of microclimatic conditions. As canopy cover changes seasonally in rubber plots due to leaf shedding, further studies should assess in detail possible effects of the leaf shedding of rubber trees on the seasonal variability of microclimatic conditions in this land-use system.

between canopy openness and air temperature and vapour pressure deficit but a strong negative relationship between canopy openness and relative air humidity, compatible with our hypothesis that canopy openness is positively correlated with air temperature and soil temperature, but negatively correlated with air humidity. Furthermore, during the day time, most of the sunlight incident on the plant canopy is absorbed by the leaves, twigs and branches while the remaining is reflected. The sunlight which is absorbed by the leaves will heat up the leaves and the canopy air. Solar radiation which is not absorbed or reflected by the plant canopy enters the soil and heats the soil surface. The heat from the soil surface is then transferred back into the air below the canopy. Therefore, the air temperature below the canopy strongly depends on the amount of heat absorbed by the plant canopy and soil surface. The land-use system with higher canopy openness will allow more heat to reach the soil surface and thus will have higher air temperature than a land-use system with lower canopy openness. Higher temperatures will then result in higher saturated water vapour pressure, which in turn increases vapour pressure deficit. Because of this, the areas with higher percentage of canopy openness and air temperature will have high vapour pressure deficit and low relative air humidity (i.e. O and R). Land-use change also has the potential of affecting the climate through modifications in the albedo (Betts, 2000; Perugini et al., 2017), with forested ecosystems leading to increased temperatures (Juang et al., 2007). In our study, we did not assess the albedo effects due to land-use change. However, a remote sensing study in the area suggest that albedo is not the dominant variable explaining land surface temperature (Sabajo et al., 2017), strengthening the hypothesis of the

4.3. Effect of ENSO Climate extremes such as heat waves affect carbon cycling (Ciais et al., 2005; Frank et al., 2015; Graf Pannatier et al., 2012) but information on how other ecosystem services are effected is very limited. ENSO events in Asia are associated with warmer temperatures (Kiladis and Diaz, 1989; Nicholls et al., 2005), for example, longer and warmer heat waves have been observed during El Niño years in India (Murari et al., 2016). In Indonesia, warmer temperatures and droughts (Allan, 2000; Harger, 1995; Susilo et al., 2013; Quinn et al., 1978), as well as increased forest fires (Page et al., 2011; van der Werf, 2004) are related to ENSO years, and the ENSO event of 2015 was among the strongest recorded (Varotsos et al., 2016). In our study, we observed an increase in temperature and Vpd and a decrease in RH in 2015, as recorded by our reference meteorological stations in open areas (Fig. 2). The drought caused by the ENSO might have led to a reduced evapotranspiration due to limited water availability (i.e. reduced plant transpiration and soil evaporation). This reduced evapotranspiration decreased atmospheric and surface cooling, making the air hotter and drier during the ENSO, amplifying the effect of the increased temperatures registered during this period. The temperature increase during the heat wave affected differently 216

Agricultural and Forest Meteorology 252 (2018) 208–219

A. Meijide et al.

rubber expansion. The relationships found in our study provide baseline information for model parameterization at regional and global scales, even though global modeling is still hampered by the use of different explanatory variables across available studies (e.g. canopy openness, leaf area index). The ENSO of 2015 led to warmer and drier conditions across land uses. Forest and jungle rubber, the less disturbed systems with low canopy openness, experienced larger increases of temperature and Vpd, and larger decreases of RH during ENSO than the monoculture plantations. In contrast, microclimatic conditions below the canopy of monocultures are more similar to those in open areas and the changes during ENSO thus less pronounced. Jungle rubber, however, reached comparable temperature, RH, and Vpd as in monocultures during ENSO. This suggests that the extreme event of ENSO in 2015 may have surpassed the buffering capacity of jungle rubber. We conclude that modified ecosystems such as jungle rubber, which are able to maintain stable microclimatic conditions under normal conditions, seem quite fragile under extreme climatic events, which may potentially have severe implications for other ecosystem functions.

the different land-use systems in the region (Fig. 6 and Table 3). During heat waves, forests provide a cool shelter, which is related to canopy closure (Renaud and Rebetez, 2009). Similarly to De Frenne et al., (2013), we expected an attenuation of the effect of ENSO in the systems with a denser canopy, and a lower attenuation in the agroecosystems (Anderson-Teixeira et al., 2012). The effect of the strong ENSO event of 2015 was pronounced in all four land-use systems although the forest, which have a denser canopy, maintained more stable microclimatic conditions (lower Ta and Vpd and higher RH) than those with less dense or more open canopies. These results are in agreement with Frey et al., (2016) who showed that single-species, even-aged plantations, would likely have reduced buffering capacity in comparison to forested and multiple-species sites. Surprisingly, there was a much larger relative increase in Ta and Vpd, and a greater relative decrease in RH in the more natural systems than in the monoculture plantations (up to 234% in Vpd in F while only 61% in R). The relative effect was especially important in JR, where in the pre-ENSO years, microclimatic conditions were intermediate between the F and the monocultures, but during the ENSO months exhibited similar conditions to the monocultures (Fig. 6). These results contradict our hypothesis, where we would have expected larger differences during ENSO in the more disturbed ecosystems (i.e. rubber and oil palm plantations, with higher canopy openness), but could be explained in terms of a collapse in the buffering capacity in JR. During the ENSO events, the variation of microclimatic conditions outside the canopy was much higher than in normal conditions (Fig. S1, see particularly the large increases in Vpd). In those land uses with a simpler structure and higher canopy openness (R and O), Ta inside the canopy was already similar to Ta outside the canopy (i.e. measured at the reference stations in open areas, see Fig. S1). When Ta outside canopy increased during the ENSO event, the change in below-canopy microclimatic conditions (increase in Ta and Vpd and decrease in RH) was not very strong for R and O, as it would have been expected for monoculture plantations (Frey et al., 2016). These results suggest that the monocultures might have very little thermal buffering capacity even under non-ENSO periods. On the other hand, in the forest, the increase of the temperature during ENSO was still buffered. Even if differences in microclimatic conditions with previous years were very large, forests are the land use which exhibits larger microclimatic differences with the reference stations in open areas (Fig. S1). For jungle rubber, our results suggest that its capacity of buffering the external microclimatic conditions was surpassed when temperature increased above a certain threshold, leading to similar microclimatic conditions as the monocultures.

Acknowledgements This study was financed by the Deutsche Forschungsgemeinschaft (DFG) in the framework of the collaborative German-Indonesian research project CRC990 (subprojects A03 and Z02). The authors would like to thank Heri Junedi from the University of Jambi (UNJA) for his collaboration with the project. Special thanks to Basri, Bayu, Anna Fitrina and Darwis, our field assistants in Indonesia, and to Edgar Tunsch, Dietmar Fellert, Frank Tiedemann and Malte Puhan for technical assistance. We thank the village leaders, PT REKI and Bukit Duabelas National Park for granting us access to sites and information. Finally, we would like to thank the Spanish project GEISpain (CGL2014-52838-C2) for financing A. Meijide during the preparation of this manuscript. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.agrformet.2018.01.013. References Abood, S.A., Lee, J.S.H., Burivalova, Z., Garcia-Ulloa, J., Koh, L.P., 2015. Relative contributions of the logging, fiber, oil palm, and mining industries to forest loss in Indonesia. Conserv. Lett. 8, 58–67. http://dx.doi.org/10.1111/conl.12103. Alkama, R., Cescatti, A., 2016. Biophysical climate impacts of recent changes in global forest cover. Science 351 (80), 600–604. Allan, R., 2000. ENSO and climatic variability in the past 150 years. In: Diaz, H., Markgraf, V. (Eds.), ENSO: Multiscale Variability and Global and Regional Impacts. University Press: Cambridge, Cambridge, pp. 3–55. Allen, K., Corre, M.D., Tjoa, A., Veldkamp, E., 2015. Soil nitrogen-cycling responses to conversion of lowland forests to oil palm and rubber plantations in Sumatra, Indonesia. PLoS One 10, e0133325. http://dx.doi.org/10.1371/journal.pone. 0133325. Anderson-Teixeira, K.J., Snyder, P.K., Twine, T.E., Cuadra, S.V., Costa, M.H., DeLucia, E.H., 2012. Climate-regulation services of natural and agricultural ecoregions of the Americas. Nat. Clim. Change 2, 177–181. Aussenac, G., 2000. Interactions between forest stands and microclimate: ecophysiological aspects and consequences for silviculture. Ann. For. Sci. 57, 287–301. http://dx. doi.org/10.1051/forest:2000119. Barnes, A.D., Jochum, M., Mumme, S., Haneda, N.F., Farajallah, A., Widarto, T.H., Brose, U., 2014. Consequences of tropical land use for multitrophic biodiversity and ecosystem functioning. Nat. Commun. 5, 5351. http://dx.doi.org/10.1038/ ncomms6351. Beltrán-Przekurat, A., Pielke, R.A., Peters, D.P.C., Snyder, K.A., Rango, A., 2008. Modeling the effects of historical vegetation change on near-surface atmosphere in the northern Chihuahuan desert. J. Arid Environ. 72, 1897–1910. http://dx.doi.org/ 10.1016/j.jaridenv.2008.05.012. Betts, R.A., 2001. Biogeophysical impacts of land use on present-day climate: near-surface temperature change and radiative forcing. Atmos. Sci. Lett. 2, 39–51. http://dx.doi. org/10.1006/asle.2001.0023. Betts, R.A., 2000. Offset of the potential carbon sink from boreal forestation by decreases in surface albedo. Nature 408, 187–190. http://dx.doi.org/10.1038/35041545.

5. Conclusions We provide first measurements of microclimatic conditions over several years below the canopy for tropical forests and land uses commonly replacing forests in Indonesia: jungle rubber, oil palm and rubber plantations. Our results point to large microclimatic variations among these land uses. The substitution of tropical forests with both rubber and oil palm monoculture plantations led to increases in the daily amplitude of air temperature, relative humidity, and Vpd, and resulted in increases of mean air temperature and Vpd of up to 2.3 °C and 665 Pa respectively, and in large decreases in mean RH (up to 12.8%). Despite all the attention that oil palm expansion has received, the effects of rubber expansion should thus not be neglected, e.g. regarding the effects of microclimatic changes on biodiversity. The continuing conversion of tropical forest to mosaic landscapes in South-East Asia, requires the assessment of all land uses. In this context, our results suggest that jungle rubber might stand out as a land use with has the capacity to generate some economic profit while avoiding significant changes in microclimate. Canopy cover is a valuable indicator of microclimate below the canopy. It can hence be used for modeling purposes and for helping to understand the climatic and microclimatic effects of oil palm and 217

Agricultural and Forest Meteorology 252 (2018) 208–219

A. Meijide et al.

38–44. http://dx.doi.org/10.1890/1540-9295(2003)001[0038:GSHTEC]2.0.CO;2. Frank, D., Reichstein, M., Bahn, M., Thonicke, K., Frank, D., Mahecha, M.D., Smith, P., van der Velde, M., Vicca, S., Babst, F., Beer, C., Buchmann, N., Canadell, J.G., Ciais, P., Cramer, W., Ibrom, A., Miglietta, F., Poulter, B., Rammig, A., Seneviratne, S.I., Walz, A., Wattenbach, M., Zavala, M.A., Zscheischler, J., 2015. Effects of climate extremes on the terrestrial carbon cycle: concepts, processes and potential future impacts. Global Change Biol. 21, 2861–2880. http://dx.doi.org/10.1111/gcb.12916. Frey, S.J.K., Hadley, A.S., Johnson, S.L., Schulze, M., Jones, J.A., Betts, M.G., 2016. Spatial models reveal the microclimatic buffering capacity of old-growth forests. Sci. Adv. 2, e1501392. http://dx.doi.org/10.1126/sciadv.1501392. Gibbs, H.K., Ruesch, A.S., Achard, F., Clayton, M.K., Holmgren, P., Ramankutty, N., Foley, J.A., 2010. Tropical forests were the primary sources of new agricultural land in the 1980s and 1990s. Proc. Natl. Acad. Sci. U. S. A. 107, 16732–16737. http://dx.doi. org/10.1073/pnas.0910275107. Gouyon, A., de Foresta, H., Levang, P., 1993. Does “jungle rubber” deserve its name? an analysis of rubber agroforestry systems in southeast Sumatra. Agrofor. Syst. 22, 181–206. http://dx.doi.org/10.1007/BF00705233. Graf Pannatier, E., Dobbertin, M., Heim, A., Schmitt, M., Thimonier, A., Waldner, P., Frey, B., 2012. Response of carbon fluxes to the 2003 heat wave and drought in three mature forests in Switzerland. Biogeochemistry 107, 295–317. http://dx.doi.org/10. 1007/s10533-010-9554-y. Guillaume, T., Muhammad, D., Kuzyakov, Y., 2015. Losses of soil carbon by converting tropical forest to plantations: erosion and decomposition estimated by δ 13 C. Global Change Biol. http://dx.doi.org/10.1111/gcb.12907. n/a-n/a. Hardwick, S.R., Toumi, R., Pfeifer, M., Turner, E.C., Nilus, R., Ewers, R.M., 2015. The relationship between leaf area index and microclimate in tropical forest and oil palm plantation: forest disturbance drives changes in microclimate. Agric. For. Meteorol. 201, 187–195. http://dx.doi.org/10.1016/j.agrformet.2014.11.010. Harger, J.R.E., 1995. Air-temperature variations and ENSO effects in Indonesia, the Philippines and El Salvador. ENSO Patterns Changes 1866-1993. Atmos. Environ. 29, 1919–1942. http://dx.doi.org/10.1016/1352-2310(95)00017-S. Jiang, J.-S.J.S., Wang, R.-S.R.S., 2003. Hydrological eco-service of rubber plantations in Hainan Island and its effect on local economic development. J. Environ. Sci. (China) 15 (September (5)), 701–709. https://www.ncbi.nlm.nih.gov/pubmed/14562935. Juang, J.Y., Katul, G., Siqueira, M., Stoy, P., Novick, K., 2007. Separating the effects of albedo from eco-physiological changes on surface temperature along a successional chronosequence in the southeastern United States. Geophys. Res. Lett. 34, 1–5. http://dx.doi.org/10.1029/2007GL031296. Kiladis, G.N., Diaz, H.F., 1989. Global climatic anomalies associated with extremes in the southern oscillation. J. Clim. http://dx.doi.org/10.1175/1520-0442(1989)002& 1069:GCAAWE&2.0.CO;2. Kotowska, M.M., Leuschner, C., Triadiati, T., Meriem, S., Hertel, D., 2015. Quantifying above- and belowground biomass carbon loss with forest conversion in tropical lowlands of Sumatra (Indonesia). Global Change Biol. 21 (10), 3620–3634. http://dx. doi.org/10.1111/gcb.12979. Kotowska, M.M., Leuschner, C., Triadiati, T., Hertel, D., 2016. Conversion of tropical lowland forest reduces nutrient return through litterfall, and alters nutrient use efficiency and seasonality of net primary production. Oecologica 180 (2), 601–618. Kovács, B., Tinya, F., Ódor, P., 2017. Stand structural drivers of microclimate in mature temperate mixed forests. Agric. For. Meteorol. 234–235, 11–21. http://dx.doi.org/ 10.1016/j.agrformet.2016.11.268. Laumonier, Y., Uryu, Y., Stuwe, M., Budiman, A., Setiabudi, B., Hadian, O., 2010. Ecofloristic sectors and deforestation threats in Sumatra: identifying new conservation area network priorities for ecosystem-based land use planning. Biodivers. Conserv. 19, 1153–1174. Laurance, W.F., Sayer, J., Cassman, K.G., 2014. Agricultural expansion and its impacts on tropical nature. Trends Ecol. Evol. 29, 107–116. http://dx.doi.org/10.1016/j.tree. 2013.12.001. Li, Y., Zhao, M., Motesharrei, S., Mu, Q., Kalnay, E., Li, S., 2015. Local cooling and warming effects of forests based on satellite observations. Nat. Commun. 6, 6603. http://dx.doi.org/10.1038/ncomms7603. Luskin, M.S., Potts, M.D., 2011. Microclimate and habitat heterogeneity through the oil palm lifecycle. Basic Appl. Ecol. 12, 540–551. http://dx.doi.org/10.1016/j.baae. 2011.06.004. Margono, B.A., Potapov, P.V., Turubanova, S., Stolle, F., Hansen, M.C., 2014. Primary forest cover loss in Indonesia over 2000–2012. Nat. Clim. Change 1–6. http://dx.doi. org/10.1038/nclimate2277. Margono, B.A., Turubanova, S., Zhuravleva, I., Potapov, P., Tyukavina, A., Baccini, A., Goetz, S., Hansen, M.C., 2012. Mapping and monitoring deforestation and forest degradation in Sumatra (Indonesia) using Landsat time series data sets from 1990 to 2010. Environ. Res. Lett. 7, 34010. http://dx.doi.org/10.1088/1748-9326/7/3/ 034010. Meijide, A., Röll, A., Fan, Y., Herbst, M., Niu, F., Tiedemann, F., June, T., Rauf, A., Hölscher, D., Knohl, A., 2017. Controls of water and energy fluxes in oil palm plantations: environmental variables and oil palm age. Agric. For. Meteorol. 239, 71–85. http://dx.doi.org/10.1016/j.agrformet.2017.02.034. Miettinen, J., Shi, C., Liew, S.C., 2011. Deforestation rates in insular Southeast Asia between 2000 and 2010. Global Change Biol. 17, 2261–2270. http://dx.doi.org/10. 1111/j.1365-2486.2011.02398.x. Murari, K.K., Sahana, A.S., Daly, E., Ghosh, S., 2016. The influence of the El Niño Southern Oscillation on heat waves in India. Meteorol. Appl. 23, 705–713. http://dx. doi.org/10.1111/fog.12167. Nicholls, N., Baek, H.J., Gosai, A., Chambers, L.E., Choi, Y., Collins, D., Della-Marta, P.M., Griffiths, G.M., Haylock, M.R., Iga, N., Lata, R., Maitrepierre, L., Manton, M.J., Nakamigawa, H., Ouprasitwong, N., Solofa, D., Tahani, L., Thuy, D.T., Tibig, L., Trewin, B., Vediapan, K., Zhai, P., 2005. The El Niño-Southern Oscillation and daily

Bhatti, V., Gupta, A., Gupta, N., 2016. Vegetation and microclimate (case study of Shri Mata Vaishno Devi University, Katra J & K). Int. J. Sci. Res. 5, 1–3. Böhnert, T., Wenzel, A., Altenhövel, C., Beeretz, L., Tjitrosoedirdjo, S.S., Meijide, A., Rembold, K., Kreft, H., 2016. Effects of land-use change on vascular epiphyte diversity in Sumatra (Indonesia). Biol. Conserv. 202, 20–29. http://dx.doi.org/10. 1016/j.biocon.2016.08.008. Bolton, D., 1980. The computation of equivalent potential temperature. Mon. Weather Rev. 108, 1046–1053. http://dx.doi.org/10.1175/1520-0493(1980)108& 1046:TCOEPT&2.0.CO;2. Bright, R.M., Davin, E., Halloran, T.O., Pongratz, J., Zhao, K., Cescatti, A., 2017. Local temperature response to land cover and management change driven by non-radiative processes. Nat. Clim. Change 7, 296–302. http://dx.doi.org/10.1038/nclimate3250. Carlson, D.W., Groot, A., 1997. Microclimate of clear-cut, forest interior, and small openings in trembling aspen forest. Agric. For. Meteorol. 87, 313–329. http://dx.doi. org/10.1016/S0168-1923(95)02305-4. Carlson, K.M., Curran, L.M., Asner, G.P., Pittman, A.M., Trigg, S.N., Marion Adeney, J., 2012a. Carbon emissions from forest conversion by Kalimantan oil palm plantations. Nat. Clim. Change 3, 283–287. http://dx.doi.org/10.1038/nclimate1702. Carlson, K.M., Curran, L.M., Ratnasari, D., Pittman, A.M., Soares-Filho, B.S., Asner, G.P., Trigg, S.N., Gaveau, D.A., Lawrence, D., Rodrigues, H.O., 2012b. Committed carbon emissions, deforestation, and community land conversion from oil palm plantation expansion in West Kalimantan. Indonesia Proc. Natl. Acad. Sci. U. S. A. 109, 7559–7564. http://dx.doi.org/10.1073/pnas.1200452109. Chen, J., Franklin, J.F., Spies, T.A., 1993. Contrasting microclimates among clearcut, edge, and interior of old-growth Douglas-fir forest. Agric. For. Meteorol. 63, 219–237. http://dx.doi.org/10.1016/0168-1923(93)90061-L. Chen, J., Saunders, S., Crow, T., Naiman, R., Brosofske, K., Mroz, G., Brookshire, B., Frankiln, J., 1999. Microclimate in forest ecosystem and landscape ecology variations in local climate can be used to monitor and compare the effects of different management regimes. Bioscience 49, 288–297. http://dx.doi.org/10.1086/250095. Ciais, P., Reichstein, M., Viovy, N., Granier, A., Ogée, J., Allard, V., Aubinet, M., Buchmann, N., Bernhofer, C., Carrara, A., Chevallier, F., De Noblet, N., Friend, A.D., Friedlingstein, P., Grünwald, T., Heinesch, B., Keronen, P., Knohl, A., Krinner, G., Loustau, D., Manca, G., Matteucci, G., Miglietta, F., Ourcival, J.M., Papale, D., Pilegaard, K., Rambal, S., Seufert, G., Soussana, J.F., Sanz, M.J., Schulze, E.D., Vesala, T., Valentini, R., 2005. Europe-wide reduction in primary productivity caused by the heat and drought in 2003. Nature 437, 529–533. http://dx.doi.org/10.1038/ nature03972. Clough, Y., Barkmann, J., Juhrbandt, J., Kessler, M., Wanger, T.C., Anshary, A., Buchori, D., Cicuzza, D., Darras, K., Putra, D.D., Erasmi, S., Pitopang, R., Schmidt, C., Schulze, C.H., Seidel, D., Steffan-Dewenter, I., Stenchly, K., Vidal, S., Weist, M., Wielgoss, A.C., Tscharntke, T., 2011. Combining high biodiversity with high yields in tropical agroforests. Proc. Natl. Acad. Sci. U. S. A. 108, 8311–8316. http://dx.doi.org/10. 1073/pnas.1016799108. Clough, Y., Krishna, V.V., Corre, M.D., Darras, K., Denmead, L.H., Meijide, A., Moser, S., Musshoff, O., Steinebach, S., Veldkamp, E., Allen, K., Barnes, A.D., Breidenbach, N., Brose, U., Buchori, D., Daniel, R., Finkeldey, R., Harahap, I., Hertel, D., Holtkamp, A.M., Hörandl, E., Irawan, B., Jaya, I.N.S., Jochum, M., Klarner, B., Knohl, A., Kotowska, M.M., Krashevska, V., Kreft, H., Kurniawan, S., Leuschner, C., Maraun, M., Melati, D.N., Opfermann, N., Pérez-Cruzado, C., Prabowo, W.E., Rembold, K., Rizali, A., Rubiana, R., Schneider, D., Tjitrosoedirdjo, S.S., Tjoa, A., Tscharntke, T., Scheu, S., 2016. Land-use choices follow profitability at the expense of ecological functions in Indonesian smallholder landscapes. Nat. Commun. 7, 13137. http://dx.doi.org/10. 1038/ncomms13137. De Frenne, P., Rodríguez-Sánchez, F., Coomes, D.A., Baeten, L., Verstraeten, G., Vellend, M., Bernhardt-Römermann, M., Brown, C.D., Brunet, J., Cornelis, J., Decocq, G.M., Dierschke, H., Eriksson, O., Gilliam, F.S., Hédl, R., Heinken, T., Hermy, M., Hommel, P., Jenkins, M.A., Kelly, D.L., Kirby, K.J., Mitchell, F.J.G., Naaf, T., Newman, M., Peterken, G., Petrík, P., Schultz, J., Sonnier, G., Van Calster, H., Waller, D.M., Walther, G.-R., White, P.S., Woods, K.D., Wulf, M., Graae, B.J., Verheyen, K., 2013. Microclimate moderates plant responses to macroclimate warming. PNAS 110, 18561–18565. http://dx.doi.org/10.1073/pnas.1311190110. Drescher, J., Rembold, K., Allen, K., Beckscha, P., Buchori, D., Clough, Y., Faust, H., Fauzi, A.M., Gunawan, D., Hertel, D., Irawan, B., Jaya, I.N.S., Klarner, B., Kleinn, C., Knohl, A., Kotowska, M.M., Krashevska, V., Krishna, V., Leuschner, C., Lorenz, W., Meijide, A., Melati, D., Steinebach, S., Tjoa, A., Tscharntke, T., Wick, B., Wiegand, K., Kreft, H., Scheu, S., 2016. Ecological and socio-economic functions across tropical land use systems after rainforest conversion. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 231, 1–7. http://dx.doi.org/10.1098/rstb.2015.0275. Ellison, D., Morris, C.E., Locatelli, B., Sheil, D., Cohen, J., Murdiyarso, D., Gutierrez, V., van Noordwijk, M., Creed, I.F., Pokorny, J., Gaveau, D., Spracklen, D.V., Tobella, A.B., Ilstedt, U., Teuling, A.J., Gebrehiwot, S.G., Sands, D.C., Muys, B., Verbist, B., Springgay, E., Sugandi, Y., Sullivan, C.A., 2017. Trees, forests and water: cool insights for a hot world. Global Environ. Change 43, 51–61. http://dx.doi.org/10.1016/j. gloenvcha.2017.01.002. Ewers, R.M., Didham, R.K., Fahrig, L., Ferraz, G., Hector, A., Holt, R.D., Kapos, V., Reynolds, G., Sinun, W., Snaddon, J.L., Turner, E.C., 2011. A large-scale forest fragmentation experiment: the stability of altered forest ecosystems project. Philos. Trans. R. Soc. B Biol. Sci. 366, 3292–3302. FAO (Food and Agricultural Organization), 2017. FAOSTAT Database. available at: http://faostat3.fao.org/browse/Q/QC/E (last Accessed: 25 July 2017). Fitzherbert, E.B., Struebig, M.J., Morel, A., Danielsen, F., Bruhl, C.A., Donald, P.F., Phalan, B., 2008. How will oil palm expansion affect biodiversity? Trends Ecol. Evol. 23, 538–545. http://dx.doi.org/10.1016/j.tree.2008.06.012. Foley, J.A., Costa, M.H., Delire, C., Ramankutty, N., Costaz, M.H., Snyder, P., 2003. How terrestrial ecosystems could affect earth â€TM s climate. Front. Ecol. Environ. 1,

218

Agricultural and Forest Meteorology 252 (2018) 208–219

A. Meijide et al.

of oil palm and other cash crops causes an increase of land surface temperature in Indonesia. Biogeosciences 14, 4619–4635. http://dx.doi.org/10.5194/bg-14-46192017. Suggitt, A.J., Gillingham, P.K., Hill, J.K., Huntley, B., Kunin, W.E., Roy, D.B., Thomas, C.D., 2011. Habitat microclimates drive fine-scale variation in extreme temperatures. Oikos 120, 1–8. http://dx.doi.org/10.1111/j.1600-0706.2010.18270.x. Susilo, G.E.B., Yamamoto, K., Imai, T., Ishii, Y., Fukami, H., Sekine, M., 2013. The effect of ENSO on rainfall characteristics in the tropical peatland areas of Central Kalimantan. Indonesia Hydrol. Sci. J. 58, 539–548. http://dx.doi.org/10.1080/ 02626667.2013.772298. Tölle, M.H., Engler, S., Panitz, H.-J., 2017. Impact of abrupt land cover changes by tropical deforestation on Southeast Asian climate and agriculture. J. Clim. 2587–2600. http://dx.doi.org/10.1175/JCLI-D-16-0131.1. van der Molen, M.K., Dolman, A.J., Waterloo, M.J., Bruijnzeel, L.A., 2006. Climate is affected more by maritime than by continental land use change: A multiple scale analysis. Global Planet Change 54, 128–149. http://dx.doi.org/10.1016/j.gloplacha. 2006.05.005. van der Werf, G.R., 2004. Continental-scale partitioning of fire emissions during the 1997 to 2001 El Nino/La Nina period. Science 80 (303), 73–76. http://dx.doi.org/10. 1126/science.1090753. Varotsos, C.A., Tzanis, C.G., Sarlis, N.V., 2016. On the progress of the 2015–2016 El Niño event. Atmos. Chem. Phys. 16, 2007–2011. http://dx.doi.org/10.5194/acp-16-20072016. Wilcove, D.S., Koh, L.P., 2010. Addressing the threats to biodiversity from oil-palm agriculture. Biodivers. Conserv. 19, 999–1007. http://dx.doi.org/10.1007/s10531009-9760-x. Ziegler, A.D., Fox, J.M., Xu, J., 2009. The rubber juggernaut. Source Sci. New Ser. 324, 1024–1025. http://dx.doi.org/10.1126/science.1173833.

temperature extremes in east Asia and the west Pacific. Geophys. Res. Lett. 32, 1–4. http://dx.doi.org/10.1029/2005GL022621. OECD-FAO, 2012. Agricultural outlook. Chp. 5. Oilseeds and Oilseed Products. Page, S.E., Morrison, R., Malins, C., Hooijer, A., Rieley, J.O., Jauhiainen, J., 2011. Review of peat surface greenhouse gas emissions from oil palm plantations in Southeast Asia. ICCT White Pap. 15, 1–78. Perugini, L., Caporaso, L., Marconi, S., Cescatti, A., Quesada, B., Noblet-Ducoudré, N., de House, J.I., Arneth, A., 2017. Biophysical effects on temperature and precipitation due to land cover change. Environ. Res. Lett. 12, 1–13. http://dx.doi.org/10.1088/ 1748-9326/aa6b3f. Porté, A., Huard, F., Dreyfus, P., 2004. Microclimate beneath pine plantation, semi-mature pine plantation and mixed broadleaved-pine forest. Agric. For. Meteorol. 126, 175–182. http://dx.doi.org/10.1016/j.agrformet.2004.06.001. Quinn, W.H., Zorf, D.O., Short, K.S., Yang, R.T.W.K., 1978. Historical trends and statistics of the southern oscillation, El Nino and Indonesian Droughts. Fish. Bull. 76, 663–678. https://www.st.nmfs.noaa.gov/spo/FishBull/76-3/quinn.pdf. Ramdani, F., Hino, M., 2013. Land use changes and GHG emissions from tropical forest conversion by oil palm plantations in Riau Province, Indonesia. PLoS One 8, 1–6. http://dx.doi.org/10.1371/journal.pone.0070323. Rasband, W.S., 1997-2015. ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA, https://imagej.nih.gov/ij/. Rembold, K., Mangopo, H., Tjitrosoedirdjo, S.S., Kreft, H., Rembold, K., 2017. Plant diversity, forest dependency, and alien plant invasions in tropical agricultural landscapes in Sumatra. Biol. Conserv. 213, 1–25. http://dx.doi.org/10.1016/j.biocon. 2017.07.020. Renaud, V., Rebetez, M., 2009. Comparison between open-site and below-canopy climatic conditions in Switzerland during the exceptionally hot summer of 2003. Agric. For. Meteorol. 149, 873–880. http://dx.doi.org/10.1016/j.agrformet.2008.11.006. Sabajo, C.R., le Maire, G., June, T., Meijide, A., Roupsard, O., Knohl, A., 2017. Expansion

219