Abstract

The Chilean north Patagonian lakes formerly called “Araucanian lakes” (38oS-41° S), were described originally as oligotrophic, nevertheless since 1978, a transition from oligotrophy to mesotrophy has been reported in some bays due to human intervention, such as towns and industrial activities. The aim of the present study was to analyze the temporal variations between five years in a north Patagonian lake comparing chlorophyll concentration with spectral properties based on LANDSAT ETM+ sensor. The results revealed significant direct correlation between chlorophyll concentrations with B2, B3. B4, B5 and B7 reflection bands. The multiple regression analysis revealed a direct association between chlorophyll concentrations with B4, and inverse with B5.  The obtained results agree with observations for other Chilean north Patagonian lakes about chlorophyll increase in determined bays with human intervention, and the use of remote sensing techniques would be an optimal tool for study changes.

1. Introduction

1The Chilean North Patagonian lakes formerly called “Araucanian lakes” (38-41° S) characterized originally by their marked oligotrophy (Campos 1984, Soto and Zúñiga, 1991, Soto, 2002). Nevertheless in the last two decades a transition from oligotrophy to mesotrophy has been reported (Woelfl et al. 2003), indeed to changes in their surrounding basins in some bays due to human intervention, specifically towns, agricultural and industrial activities with the consequent replacement of original native forest (Soto 2002). An example can be Riñihue lake that has reported a transition from oligotrophy to mesotrophy, from 1978 to 1998 (Woelfl et al., 2003), or Llanquihue lake with bays with a wide gradient of trophic status due human activities associated such as towns, agriculture or aquaculture activities (Soto 2002, De los Ríos-Escalante et al, 2017a). Also, according to the literature, the main regulator of phytoplankton activity in Chilean Patagonian lakes is the mixing depth; a marked inverse correlation exists between both parameters, consequently, the northern Patagonian lakes, due their low mixing depth are more exposed to increase their chlorophyll concentration due to nutrients inputs (Soto, 2002).

2The literature reports the use of remote sensing techniques for limnological monitoring (Kondratyev and Filaratov, 1999), some of them are related with phytoplankton pigments monitoring specifically for avoid toxic algal blooms (Li et al. 2015, Lunetta et al. 2015, Oyama et al. 2015). There are very scarce reports about use of remote sensing techniques in Chilean lakes, the existing few studies are based on spectral properties in spatial gradient in glacial influenced lakes in Chilean Patagonia (De los Ríos-Escalante and Acevedo 2016a,b), where the gradient of glacial influence generate changes in water color with consequent changes in zooplankton composition. Also, the use of remote sensing techniques was reported for Chilean Patagonian mountain lakes with difficult access, as first descriptions of these water bodies (De los Ríos-Escalante et al., 2017b,c,d). The aim of the present study is to find correlations between chlorophyll concentrations and spectral properties in Riñihue Lake, because this lake was one of the most studied in Chile, considering its temporal variations in trophic status during three decades (Woelfl et al. 2003).

2. Material and methods

3Data of Chlorophyll concentrations in Riñihue lake (Fig. 1), from May 2004 to October 2009, during the period corresponding to late winter to late summer, that corresponds to maximum chlorophyll concentrations (Wölfl, 1996; Table 1), were obtained from “Dirección General de Aguas” database (www.dga.cl). These sites corresponds to sampling point located at 39°48’53’’S and 72°24’44’ W (Figure 1). Satellite data was obtained from LANDSAT/ETM+ image dated between May 2004 to October 2009 (Table 1) provided by the Land Processess Distributed Active Archive Center (LP DAAC), U.S. Geological Survey (http://LPDAAC.usgs.gov). The bands of visible, near, and mid-infrared (Table 2) were calibrated radiometrically to spectral irradiance and then to reflectance with   correction being applied (Table 1). All data analysis was applied using software “R” (R Development Core Team, 2009). At afirst step, data were analyzed by matrix correlation analysis using Hmisc R package, (Harrell, 2016) for determine the associations between studied variables. At a second step multiple regression analysis was applied considering chlorophyll concentration as dependent variable, whereas the spectral properties B1, B2, B3, B4, B5 and B7 reflectance bands were considered as independent variables. This statistical analysis was applied with the HSAUR R package (Everitt and Hothorn, 2016).

Figure 1: Map with studied site.

Table 1 : Chlorophyll “a” concentration (“Chl(a)”, in mg/L), and reflectance in Landsat ETM+ bands, for sampled site in Riñihue Lake.

Sampling date	Chl(a)	Satellite date	B1	B2	B3
26-05-2004	1.10	13-05-2004	0.1040	0.0580	0.0310
11-02-2005	0.50	09-02-2005	0.0810	0.0475	0.0235
18-11-2005	0.40	08-11-2005	0.0775	0.0440	0.0220
09-08-2006	1.80	30-07-2006	0.1011	0.0587	0.0349
25-10-2006	0.60	26-10-2006	0.0777	0.0445	0.0226
26-05-2007	1.10	06-05-2007	0.1106	0.0611	0.0360
21-08-2007	3.40	18-08-2007	0.0918	0.0566	0.0339
20-11-2007	0.70	14-11-2007	0.0772	0.0424	0.0219
26-02-2008	0.50	26-02-2008	0.0829	0.0486	0.0287
23-08-2008	4.40	13-09-2008	0.1062	0.0703	0.0479
25-02-2009	0.36	28-02-2009	0.0793	0.0447	0.0252
03-06-2009	1.40	12-06-2009	0.1246	0.0721	0.0409
28-10-2009	0.70	11-11-2009	0.0789	0.0485	0.0275
26-05-2004	1.10	13-05-2004	0.0180	0.0008	0.0006
11-02-2005	0.50	09-02-2005	0.0177	0.0041	0.0024
18-11-2005	0.40	08-11-2005	0.0171	0.0020	0.0010
09-08-2006	1.80	30-07-2006	0.0235	0.0026	0.0026
25-10-2006	0.60	26-10-2006	0.0165	0.0022	0.0011
26-05-2007	1.10	06-05-2007	0.0212	0.0027	0.0020
21-08-2007	3.40	18-08-2007	0.0246	0.0029	0.0027
20-11-2007	0.70	14-11-2007	0.0164	0.0016	0.0011
26-02-2008	0.50	26-02-2008	0.0234	0.0064	0.0045
23-08-2008	4.40	13-09-2008	0.0531	0.0235	0.0159
25-02-2009	0.36	28-02-2009	0.0199	0.0031	0.0018
03-06-2009	1.40	12-06-2009	0.0235	0.0024	0.0015
28-10-2009	0.70	11-11-2009	0.0209	0.0034	0.0028

Table 2: Technical characteristics of the Landsat7/ETM+ sensor.

Band	Spectral Range [µm]	Wavelength Center [µm]	GSD [m]
1	0.450 – 0.515	0.479	30
2	0.525 – 0.605	0.561	30
3	0.630 – 0.690	0.661	30
4	0.775 – 0.900	0.835	30
5	1.550 – 1.750	1.650	30
7	2.090 – 2.350	2.208	30

3. Results

4The results revealed low chlorophyll “a” concentrations that varied between 0.36 to 4.40 g/L that would correspond to transition from oligotrophic to mesotrophic status, similar to the descriptions for Patagonian lakes (Table 1). There is not a regular trend of chlorophyll concentration variations during the sampled period (Table 1).

5The correlation analysis revealed significant direct associations between chlorophyll “a” concentration with B2, B3, B4, B5, B7 reflectance, B1 with B3, B3 with B2, B4 with B2, B4 with B3, B5 with B3, B5 with B4, B7 with B3, B7 with B4, B7 with B5 (Tables 2 and 3).

Table 3: Correlation matrix for parameters considered in the present study (“p” values between brackets, “p” values lower than 0.05 denotes significant correlations).

	B1	B2	B3	B4	B5	B7
Chl(a)	0.48 (0.09) n.s	0.69 (< 0.01)*	0.80 (< 0.01)*	0.84 (< 0.01)*	0.70 (< 0.01)*	0.75 (< 0.01)*
	B1	0.95 (< 0.01)*	0.85 (< 0.01)*	0.41 (0.16) n.s	0.21 (0.49) n.s	0.24 (0.43) n.s
		B2	0.96 (< 0.01)*	0.65 (0.01)*	0.45 (0.11)n.s	0.49 (0.09) n.s
			B3	0.81 (< 0.01)*	0.63 (0.02)*	0.67 (0.01)*
				B4	0.96 (< 0.01)*	0.97 (< 0.01)*
					B7	0.71 (< 0.01)*

6Multiple regression analysis yielded an inverse association between chlorophyll concentration and B5 reflectance, and direct association with B4 reflectance. The regression equation describing this association was:

7Y = -3.4873 + 260.6268X₁ – 256.9269X₂

8Where:

9Y = chlorophyll concentration,

10X₁ = B4 reflectance,

11X₂ = B5 reflectance,

12Multiple R-squared: 0.8333

13Adjusted R-squared: 0.7999

14F statistic: 24.99

15P = 0.00012.

4. Discussion

16The results denoted the oligomesotrophic status observed and predicted for Chilean northern Patagonian lakes during the last decade due changes in their surrounding basin (Soto, 2002, Woelfl et al. 2003, Woelfl 2007, De los Ríos-Escalante et al., 2011, in 2017d). It is a marked different situation in comparison to northern Patagonian lakes that are oligotrophic (Trochine et al. 2015, Modenutti et al. 2013). Rinihue lake was intensively studied between 1978 to 1998, period where it was reported marked nutrient inputs due changes in soil use in their surrounding basin, mainly in sampled point (Woelfl et al., 2003).

17The use of remote sensing techniques in limnology of Chilean lakes have been a good predictive tool for determine the zooplankton community based in water column transparence (De los Ríos-Escalante and Acevedo, 2016a,b; De los Ríos-Escalante et al., 2013). These variations in water column transparence are a consequence of glacier melt inputs, that generate light limitation that would affect the phytoplankton activity necessary for sustain zooplankton communities such as been observed for Argentinean glacial Patagonian lakes (Laspoudamares et al, 2013, Hylander et al. 2011).

18The use of remote sensing techniques has important applications for limnology, mainly in lakes with difficult access like  in northern Patagonia with many mountain lakes, located in protected areas, where some few of them are accessible by long mountain paths (De los Ríos-Escalante et al. in press 2017b,c,d). Some of these lakes were exposed during the last decade to catastrophic events such as volcanic activity with consequences in their water composition, such as been observed in mountain lakes of Conguillio National Park. These changes can be detected using remote sensing techniques specifically LANDAT ETM+ images (De los Ríos-Escalante et al. 2017c). In this scenario, and if we considerate the literature descriptions about use of remote sensing applications to detect with phytoplankton pigments (Li et al. 2015, Lunetta et al. 2015, Oyama et al. 2015), the exposed results would agree with the literature descriptions.

19The literature reports that studied site originally has mixotrophic ciliates in its original oligotrophic status, and these were gradually replaced by phytoplankton, specifically diatoms and chlorophytes due the nutrients inputs (Wölfl, 1995, Woelfl, 2007). In this scenario, if we considere the study of mixotrophic ciliates concentration and phytoplankton pigments concentrations independently, it would have probably more robust predictive models, that would explain also, all pelagic structure in Patagonian lakes. As conclusion the remote sensing applications for limnological studies of Chilean Patagonian lakes is auseful and robust technique that would have interesting applications for water monitoring procedures at temporal and spatial scales.

5. Acknowledgements

20The present study was founded by projects Tides Grant Foundation TRF13-03011 MECESUP UCT 0804 and the Research Direction of the Catholic University of Temuco and M.I for her valuable comments for improve the manuscript.

6. References

21Campos H., 1984. Limnological study of Araucanian lakes (Chile). Verhandlungen International Vereinung Angewaldte Limnologie, 22: 1319-1327.

22De los Ríos Escalante P. & Acevedo P., 2016a First observations on zooplankton and optical properties in a Glacial North Patagonian lake (Tagua Tagua lake, 41° S, Chile). Polish Journal of Environmental Studies, 25: 453-457.

23De los Ríos Escalante P. & Acevedo P., 2017b. First observations of Boeckella michaelseni Mrázek, 1901, and optical properties of a Central Patagonian lake. Polish Journal of Environmental Studies, 25:  1781-1785.

24De los Ríos Escalante P., Quinan E., & Acevedo P., 2013. Crustacean zooplankton communities in lake General Carrera (46° S) and their posible association with optical properties. Crustaceana, 86: 507-513.

25De los Ríos Escalante P., Hauenstein E., & Romero-Mieres M., 2011. Microcrustacean assemblages composition and environmental variables in lakes and ponds of the Andean region - South of Chile (37-39° S). Brazilian Journal of Biology, 71: 353-358.

26De los Ríos Escalante P., Soto D., Santander-Massa R., & Acevedo P.,  2017a. Plankton crustaceans in bays with different trophic status in Llanquihue lake (41° S Chile). Brazilian Journal of Biology, 77: 469-475.

27De los Ríos Escalante, P., Acevedo P., Esse C., & Castro M., 2017b.  Satellite spectral property observations in Chileaen lakes. Polish Journal of Environmental Studies 26: 1413-1418.

28De los Ríos Escalante, P., Acevedo P., Esse C., & Castro M.,  2017c. Spectral properties in lakes surrounding a Chilean volcano. Polish Journal of Environmental Studies 26: 935-939

29De los Ríos Escalante, P., Acevedo P., Esse C., & Castro M., 2017c Spectral properties in lakes of a Chilean National Park. Polish Journal of Environmental Studies 26: 425-429.

30Everitt B.S. & Hothorn T., 2016 HSAUR: A handbook of statistical analyses using R (first edition). Available at: https://cran.r-project.org/web/packages/HSAUR/index.html, visited 17th October, 2016

31Harrell, F.E. Jr. 2016 Hmisc: Miscellaneous. Available at: https://cran.r-project.org/web/packages/Hmisc/index.html, visited, 17th October, 2016.

32Hylander S, Jephson T., Lebret K., Von Einem J., Fagerberg T., Balseiro E., Modenutti B., Souza M.S., Laspoumaderes C., Jöhnson M., Ljungberg P., Nicolle, A., Nilsson P.A., Ranaker L., & Hansson L-A., 2011. Climate-induced input of turbid glacial meltwater affects vertical distribution and community composition of phyto-and zooplankton. Journal of Plankton Research, 33: 1239-1248.

33Kondratyev K.Y.., & Filaratov N.N., 1999. (Eds). Limnology and remote sensing. A contemporany approach. Springer-Praxis Series in Remote Sensing, 406 p.

34Laspoumaderes C., Modenutti B., Souza M.S., Bastidas Navarro M., Cuassolo F., & Balseiro E., 2013. Glacier melting and stoichiometry implications for lake community structure: zooplankton species distributions across a natural light gradient. Global Change Biology, 19: 316-326.

35Li L., L. Li, & Song K.,  2015. Remote sensing of freshwater cyanobacteria: an extended IOP Inversion Model of Inland Waters (IIMIW) for partitioning absorption coefficient and estimating phycocyamin. Remote Sensing of  Environment, 157: 9-23

36Lunetta R.S., Schaeffer B.A., Stunpf R.P., Keith D., Jacobs S.A., & Murphy M.S.,  2015 Evaluation of cyanobacteria cell count detection derived from MERIS imagery across the eastern USA. Remote Sensing of  Environment 157: 23-34

37Modenutti B., Balseiro E., Bastidas Navarro M., Laspoumaderes C., Souza M.S., & Cuassolo F., 2013. Environmental changes affecting light climate in oligotrophic mountain lakes: the Deep chlorophyll máxima and sensitive variable.  Aquatic Sciences, 75: 361-371.

38Trochine C., Diaz C., Bastidas M., Balseiro E., & Modenutti B., 2015. The abundance of mixotrophic algae drives the carbon isotope composition of the copepod Boeckella gracilipes in shallow Patagonian lakes. Journal of Plankton Research, 37: 441-451.

39Oyama Y., Matsushita B., & Fukushima T., 2015. Distinguishing surface cyanobacterial blooms and aquatic macrophytes using LANDSAT/TM and ETM+ software infrared bands. Remote Sensing of  Environment 157: 35-47

40R Development Core Team. 2009. R: A language and environment for statistical computing. R foundation for statistical computing, Vienna, Austria.

41Soto D. & Zuñiga L.R., 1991. Zooplankton assemblages of Chilean temperate lakes: a comparison with North American counterparts. Revista Chilena de Historia Natural, 64: 569-581.

42Soto D. 2002. Oligotrophic patterns in southern Chile lakes: the relevance of nutrients and mixing depth. Revista Chilena de Historia Natural, 75: 377-393.

43Wölfl, S. 1995. Untersuchungen zur Zooplanktonstruktur einschliesslich der mikrobiellen Gruppen unter besonderer Berücksichtigung der mixotrophen Ciliaten in zwei Südchilenischen Andenfußseen. Doctoral Thesis, Universität Konstanz.

44Woelfl S, Villalobos L., & Parra O., 2003. Trophic parameters and method validation in Lake Riñihue (North Patagonia: Chile) from 1978 through 1997. Revista Chilena de Historia Natural, 76: 459-474.

45Woelfl S., 2007. The distribution of large mixotrophic ciliates (Stentor) in deep North Patagonian lakes (Chile): first results. Limnologica, Ecology and Management of Inland Waters, 37: 28-36.

Pour citer cet article